0

ASME Conference Presenter Attendance Policy and Archival Proceedings

2011;():i. doi:10.1115/DSCC2011-NS2.
FREE TO VIEW

This online compilation of papers from the ASME 2011 Dynamic Systems and Control Conference and Bath/ASME Symposium on Fluid Power and Motion Control (DSCC2011) represents the archival version of the Conference Proceedings. According to ASME’s conference presenter attendance policy, if a paper is not presented at the Conference, the paper will not be published in the official archival Proceedings, which are registered with the Library of Congress and are submitted for abstracting and indexing. The paper also will not be published in the ASME Digital Library and may not be cited as a published paper.

Commentary by Dr. Valentin Fuster

Novel Actuation

2011;():1-7. doi:10.1115/DSCC2011-5920.

Pneumatic artificial muscle (PAM) is a class of flexible muscle-like actuator with low structure weight and high power density. In this paper, an integrated PAM design is presented, which incorporates peripheral elements, especially the control valves, in the interior space of the actuator to reduce the size and complexity of the entire system. More importantly, the incorporated components are expected to largely eliminate the dead volume in the PAM actuator, and improve its energy efficiency. An energy consumption analysis is presented, which estimates the maximum percentage of achievable energy savings. A practical design, incorporating standard poppet valves is also presented. To demonstrate the energy saving effect of the proposed design, experiments were conducted with a commercial PAM actuator modified to incorporate a solid bar. Experimental results indicate an energy savings of up to 13%, which is expected to increase significantly with the custom-made integrated PAM actuators in the future.

Commentary by Dr. Valentin Fuster
2011;():9-16. doi:10.1115/DSCC2011-5962.

In this paper, a single-moving-part planar positioner with 6 coils is designed and implemented. A concentrated-field permanent-magnet matrix is employed as the stationary part. The moving platen has a compact size (185.4 mm × 157.9 mm), light mass (0.64 kg) and low center of mass. The moving platen carries three planar-motor armatures with two phases per motor. Force calculation is based on the Lorentz force law and conducted by volume integration. In order to deal with the nonlinearity due to trigonometric terms in the force-current relation, modified PID (proportional-integral-derivative) and lead-and-PI controllers are designed with computed currents to close the control loop and obtain the desired performances. Experimental results verify the commutation law and the force calculation. The new design with only 6 coils allows for simplification of the control algorithm and reduced power consumption of the positioner. The maximum travel ranges in x, y, and the rotation about the vertical axis are 15.24 cm, 20.32 cm, and 12.03°, respectively. The positioning resolution in x and y is 8 μm with the root-mean-square (rms) position noise of 6 μm. The positioning resolution in rotations about the vertical axis is 100 μrad.

Commentary by Dr. Valentin Fuster
2011;():17-24. doi:10.1115/DSCC2011-6022.

This paper presents the modelling of an actuator based on Magnetic Shape Memory Alloys (MSMA). The actuation principle relies on the ability of the material to change its shape under the application of a magnetic field. Previous models proposed by authors were based on canonical (symplectic) Hamiltonian modeling and thermodynamics of irreversible processes. These models, though physically cogent, are non-minimal differential algebraic dynamical models and hence less adapted for control purposes. This paper therefore proposes a modified and system-oriented modeling procedure which lends itself naturally to a port-Hamiltonian model. The latter is found to be a minimal realization of the above whereby interconnection between subsystems is clearly visible. Using Lagrange multipliers, constraints which arise due to causality and interconnection are expressed. In the last section, Differential Algebraic Equations (DAE) resulting from previous models are reduced to Ordinary Differential Equations (ODE) and by using coordinate transformations, constraints are decoupled from the system input/output. The resulting model is well-suited for control.

Commentary by Dr. Valentin Fuster
2011;():25-32. doi:10.1115/DSCC2011-6070.

Displacement amplification mechanisms have been a topic of research for piezoelectric actuators for decades to overcome their significantly small strain, but still utilize their high power density, force, and efficiency. This paper further analyzes a nonlinear buckling mechanism to improve its efficiency, defined as the ratio of mechanical work output of the buckling actuator to the mechanical work output of the PZT actuator, as well as, employing two methods, preload and loading conditions, that improve its work output per cycle. This is accomplished by running a numerical analysis of the geometry of the flexure joints in the buckling mechanism which found a maximum mechanical efficiency of 48%. The preload is applied using shape memory alloy wire to exploit the low stiffness of the super elastic regime; which in turn allows a larger work output due to a loading condition supplied by a novel gear design. Finally, a prototype was fabricated to provide a baseline of comparison against these concepts.

Commentary by Dr. Valentin Fuster
2011;():33-40. doi:10.1115/DSCC2011-6150.

Due to very low power throughput and behavior as primarily capacitive loads, voltage conversion for micro-scale piezoelectric and electrostatic actuators can be very inefficient, while analog drive circuit supply currents may also exceed actuator power consumption. When limited instead to switching commands at discrete time instants to control micro-actuator motion, it may be advantageous to coordinate operating periods of any voltage conversion circuitry with actuators. A sample scenario of a traditional boost converter driving a thin-film piezoelectric micro-robotic appendage along a trajectory is considered through simulation studies. Some ranges of target motion accuracy are achieved using less energy, up to a 25% reduction at 30 V, for a 4 V battery, when coordinating voltage converter operation, even when conversion efficiency is very low.

Commentary by Dr. Valentin Fuster
2011;():41-48. doi:10.1115/DSCC2011-6153.

In this paper, we present the design, fabrication, and testing of a moving magnet actuator (MMA) for large range (∼10mm) nanopositioning. MMAs are direct-drive, single-phase electromagnetic linear actuators that provide frictionless and backlash-free motion. These qualities, along with an adequate motion range, make MMAs promising candidates for large range nanopositioning. In this work, we identify actuator- and system-level performance criteria and associated design tradeoffs, and use this knowledge to systematically and concurrently design an MMA and a double parallelogram flexure bearing. The resulting actuator provides a force output per unit square root power of 4.56N/W, better than 9% force uniformity with respect to stroke, and a low moving mass of 106g. An integrated thermal management system is also incorporated as part of the actuator in order to mitigate the heat dissipated from the MMA coils. The overall single-axis motion system was fabricated and tested to demonstrate a 36Hz open-loop bandwidth and less than 4nm (RMS) steady-state positioning noise over a 10mm motion range. Preliminary closed-loop design and testing highlight the potential of the proposed actuator in nanopositioning.

Topics: Magnets , Actuators
Commentary by Dr. Valentin Fuster

Robotics

2011;():49-56. doi:10.1115/DSCC2011-5908.

In this work a mathematical model of the motion of a cylinder moving on a plane is deduced using screw theory. The linear Coulomb friction equations are applicable for the maximum static and kinetic friction forces. In the case of the rolling motion of a cylinder, the friction forces are not necessarily maxima. This paper describes the dynamic states of motion of a cylindrical part moving in three separate scenarios by the Euler dual equation. The first scenario is when the cylinder is moving on a horizontal static plane due to an external harmonic force proportional to the mass of the part. For this case, the sliding conditions are expressed as a function of the vibration parameters and generalized based on a harmonic dimensionless variable. The second and third scenarios are when the cylinder is moving by translational displacements on a horizontal and inclined plane.

Topics: Screws , Cylinders
Commentary by Dr. Valentin Fuster
2011;():57-64. doi:10.1115/DSCC2011-5918.

A new approach to localize an indoor wheeled mobile robot (WMR) using a low-cost laser sensor is proposed. An onboard laser sensor moving inside a guide way is used to recognize the unique pattern of a target, by virtue of which the vehicle is localized globally. A kinematic model is developed, which determines the vehicle orientation and location with respect to the center of the target object. Here, the design parameters of the target and guide way lengths are analyzed in terms of their influence on the localization accuracy in the presence of noise in measurements. The kinematic model and the design parameter analysis are validated by the simulation results.

Commentary by Dr. Valentin Fuster
2011;():65-71. doi:10.1115/DSCC2011-6000.

PC-based video games are commonly controlled through a combination of pointing device and keyboard input. Although this conventional type of interface has a long history of successful implementation, a small group of people remains excluded due to accessibility issues. This is unfortunate, because virtual environment immersion can provide great benefits to people suffering from temporary or permanent physical disabilities. This work describes the development of a novel human-computer interface system that incorporates biosensors in order to replace conventional input devices. It will engage this group of overlooked users by addressing and eliminating the previously mentioned accessibility issues. In this system, the input from an eye-tracking device will be used to replace on-screen pointing, and a data-glove will be used to replace keyboard input for in-game navigation and interaction. It is expected that previously limiting physical conditions will be eliminated by the system’s effective, effortless, and pleasant interaction control implications.

Topics: Biosensors
Commentary by Dr. Valentin Fuster
2011;():73-80. doi:10.1115/DSCC2011-6069.

We have developed a novel type of submersible called a Synergistically Propelled Ichthyoid [SPI] which is propelled by the combined jet and tail action of a fluttering fluid-conveying pipe. Two dynamic models for an SPI are presented and solved; an analytically tractable model in an inertial reference frame and a more complex model in a general planar reference frame. The latter model has been solved numerically—these simulations show a benefit to using a fluttering tail relative to a dimensionally identical rigid tail. Construction details of an experimental realization are provided. Preliminary measurements taken using that platform are also provided which qualitatively confirm the simulation’s conclusion that a fluttering flexible tail is capable of higher speed than a dimensionally identical rigid tail.

Commentary by Dr. Valentin Fuster
2011;():81-86. doi:10.1115/DSCC2011-6096.

Mobile robot navigation in unknown, unstructured environments such as hilly terrains is often complex due to uncertainties associated with classification of terrain features as traversable or non-traversable and identification of paths between start and goal that pose minimum danger to the robot during navigation. Humans possess an uncanny ability to identify paths under the presence of uncertainties and determine whether a terrain feature is safe to traverse or not. Therefore, it is beneficial if such human knowledge is taken advantage of when available. This paper discusses the use of reinforcement learning techniques for navigation in hilly type terrains with and without the human assistance. When a prior knowledge such as low resolution aerial view about the terrain is available, a human expert can suggest paths that are relatively safe to traverse. A value matrix is developed using those expert paths that could be used by the robot to steer from any start position in the terrain to the goal position. Matlab based simulations have been carried out and the results have been presented for robot navigation in a terrain in the presence of multiple expert suggested paths.

Commentary by Dr. Valentin Fuster
2011;():87-94. doi:10.1115/DSCC2011-6147.

Passive dynamic walking is a gait developed, partially or in whole, by the energy provided by gravity. An improved kneed bipedal walking mechanism was designed and built to study passive gait patterns. The first aim of this study is to determine if testing a passive dynamic biped walker on a ramp is equivalent to testing on a treadmill. Based on the small difference between the gait patterns measured on the two test platforms, testing on a treadmill is equivalent to testing on a ramp. Measurement of the gait parameters were then conducted on the treadmill to evaluate the effects of the treadmill angle of inclination, mass distribution of the biped, treadmill belt speed and length of flat feet. Our experimental results are presented and compared with previous experimental and simulation results. Research on passive dynamic bipedal walking helps to develop an understanding of walking mechanics. Moreover, experimental passive dynamic walking results provide information to validate mathematical models of passive dynamic walking.

Commentary by Dr. Valentin Fuster
2011;():95-102. doi:10.1115/DSCC2011-6161.

An integrated approach to underwater wireless communication and orientation estimation using visible light is presented and demonstrated using a simplified 1-DOF system. This system utilizes a long range, high bandwidth optical communication system as a reference with which to rectify error in the prediction of the orientation state. This is done through the use of a Discrete Kalman Filter. In this way, a dual-use system is developed. Using the same hardware, this system can simultaneously communicate using a modulated light source, and estimate, with accuracy, absolute orientation in space. This orientation knowledge is vital to maintaining a line-of-sight link for the communication system. In addition, orientation estimation is important for the various tasks that the underwater robot must perform.

Commentary by Dr. Valentin Fuster
2011;():103-110. doi:10.1115/DSCC2011-6162.

Research on dynamics of micro devices indicates that contact forces are complex in a micro-scale environment. Previous work tried to identify how small-scale forces numerically affect a walking micro system, using a similar structure to a micro-robot leg and a simple analytical model for system dynamics. This paper extends that work by focusing on further experimental testing and simulation studies on a prototype micro-robot. Evaluation of experimental data indicates that characteristics of the ground, such as material properties and static charge on the surface should be also considered in understanding foot-terrain interaction. This leads to modifications to the former foot-terrain interaction algorithm. The refined analytical model is validated through experimental comparison.

Commentary by Dr. Valentin Fuster
2011;():111-114. doi:10.1115/DSCC2011-6164.

In this paper it is proposed a central pattern generator (CPG) based on workspace intentions, where the parameters of modulation have physical meaning and the walking can be adapted to overcome irregular terrains by changing few parameters. The walking features are independently modulated since there is no coupling relationship among WCPG parameters. Simulation results are presented to demonstrate the WCPG performance for a simplified quadruped robot model in different terrains.

Topics: Robots , Generators
Commentary by Dr. Valentin Fuster
2011;():115-120. doi:10.1115/DSCC2011-6185.

A new control algorithm was developed for tremor estimation and suppression in a second order linear model of the human elbow joint. An adaptive method developed to estimate a simple harmonic disturbance was generalized for tremorous motion with spectral composition similar to the clinical reports for action tremor. Numerical simulations showed the ability of proposed controller to reduce tremor amplitude without generating significant resistance against voluntary motion of the arm. The designed algorithm can be used in an upper-limb orthosis to suppress debilitating tremorous motion of the arm.

Topics: Human elbow
Commentary by Dr. Valentin Fuster
2011;():121-127. doi:10.1115/DSCC2011-6186.

In this paper, we propose a linear parameter varying (LPV) control design approach for trajectory tracking in a robotic system, intended to be involved in an image-guided teleoperated cardiac surgery. The robot is eventually aimed to guide a 3 degree-of-freedom medical tool (a catheter) inside the left ventricle (LV) and achieve the implantation of a prosthetic aortic valve. The successful delivery of the valve from the apical entrance to the aortic annulus strongly depends on the precise navigation of the catheter such that its probable collision with the LV’s changing environment is avoided. The LPV control strategy is utilized here due to its ability to capture the nonlinearities of the designed robot manipulator and adapt in real-time based on the varying end effector’s angle. The simulation studies demonstrate promising results achieved for a guaranteed safe navigation through LV.

Topics: Valves , Manipulators
Commentary by Dr. Valentin Fuster
2011;():129-136. doi:10.1115/DSCC2011-6188.

In this paper we study the classical set-point control problem for rigid robots when there are time-varying delays in the input-output channels. It has been demonstrated earlier that scattering variables together with additional gains can be utilized to stabilize the closed loop system in the presence of time-varying delays. However, this architecture is not able to guarantee asymptotic regulation to the desired configuration, and the stability depends on the maximum rate of change of the time-varying delays in the communication. Hence, in this paper, we present a new architecture where scattering variables and position feedback are utilized to guarantee stability and asymptotic convergence of the regulation error to the origin while simultaneously relaxing a significant assumption on the rate of change of delays. The proposed algorithm is numerically verified on a two-degree-of-freedom manipulator.

Topics: Delays , Manipulators
Commentary by Dr. Valentin Fuster

Sensing

2011;():137-144. doi:10.1115/DSCC2011-5963.

In this paper, we investigate the feasibility of off-the-shelf buoyant fluorescent microspheres as particle tracers in turbid water flows. Microspheres’ fluorescence intensity is experimentally measured and detected in static aqueous suspensions of increasing concentrations of clay to simulate typical conditions of natural drainage networks. We conduct experiments by using photoconductive cells and image-based sensing methods. Results obtained with both approaches exhibit comparable trends and show that the considered particles are detectable in critically turbid water flows. Further information on the performance and integration of such microspheres in low-cost measurement instrumentations for field observations is obtained through experiments on a custom built water channel. Findings from this study show that the proposed technology may serve as a minimally invasive sensing system for hazardous events, such as pollutant diffusion in natural streams and flash flooding due to extreme rainfall.

Commentary by Dr. Valentin Fuster
2011;():145-151. doi:10.1115/DSCC2011-5999.

A magnetic field-based sensing system utilizing statistically optimized concurrent multi-sensor outputs for non-contact precise field-position association is presented. The rationale and principle of capitalizing on simultaneous spatial field measurements to induce unique correspondence between field and position to achieve accurate translational motion over large travel distances for feedback control is illustrated using a single-source-multi-sensor configuration. Principal component analysis (PCA) is used as a pseudo filter to optimally reduce the dimension of the multi-sensor output space for field-position mapping with artificial neural networks (ANNs). The effects of PCA on the sensing accuracy and closed-loop tracking performance are experimentally investigated using a voice-coil motor and a 9 sensor network with an optical encoder as a comparison.

Topics: Engines
Commentary by Dr. Valentin Fuster
2011;():153-158. doi:10.1115/DSCC2011-6031.

Electrical Capacitance Tomography (ECT) is a method to determine the material distribution within the interior of a closed object by measuring the capacitance values across externally mounted electrodes. Traditionally, an AC excitation pulse is applied to a pair of electrodes that form a capacitor during each measurement step, in order to determine the capacitance from the output current measured. This paper investigates how the speed of inter-electrode capacitance measurement can be improved by comparatively studying three methods that affect the way electrodes are excited and signals are received: 1) multiple-excitation-single-receiving, 2) single-excitation-multiple-receiving, and 3) multiple-excitation-multiple-receiving. A PSPICE circuit model was built to simulate the bandwidth and interference between the simultaneously sampled measurement channels. Simulations using ECT sensors with 8 and 12-electrodes have shown that measurement speed can be increased by up to 3 ∼ 30 times as compared to the traditional mode of capacitance measurement. Such new capability opens up new possibilities for ECT as an effective tool for online, real-time monitoring of a wide range of dynamical processes in the industry.

Topics: Capacitance
Commentary by Dr. Valentin Fuster
2011;():159-165. doi:10.1115/DSCC2011-6039.

This paper presents design and analysis of a flexonic mobile node (FMN) for structural health monitoring. Unlike rigid configurations with a fixed distance between the front and rear axles, this FMN features various deformable configurations in three-dimensional (3D) space by incorporating a compliant beam (connecting the front and rear axles with a controlled pin joint). Hence, this FMN has more flexibility to negotiate corners and attach sensors on ferromagnetic structures. These functions are facilitated by the guidance of an embedded vision controlled system. Controlling a continuous compliant beam is a challenging task in terms of nonlinear deformation with infinite degrees of kinematic redundancy. Detailed discussion focuses on a theoretical beam model for simulating its 3D deformed shape. Results include the relations between input/output displacements and rotation angle for control implementation in sensor attachment and corner negotiation. Experiment is provided for model validation by comparing with the analysis for sensor attachment.

Topics: Deformation
Commentary by Dr. Valentin Fuster
2011;():167-172. doi:10.1115/DSCC2011-6059.

Online measurement of pressure, temperature, velocity, and viscosity of polymer melt within the injection mold is key to improving process control for quality assurance in injection molding. A multivariate sensing method has been introduced to enable simultaneous determination of the four parameters in injection molding using one single sensor package. Comparing with the method of measuring different physical parameters using separate sensors, the new method is advantageous in terms of system miniaturization and energy efficiency. This paper addresses the aspect of polymer melt velocity sensing based on the melt temperature measured by an Infrared (IR) sensing element integrated within the sensor package. From the Stephan-Boltzmann model, an analytical relationship between the melt front velocity and the ramping rate of the IR detector voltage output has been established. Using a Finite Element (FE) model, the process of polymer melt flowing over the IR sensor lens has been simulated. The result shows that the melt front velocity can be determined within an error of ± 0.25%, under a broad range of melt temperature and IR detector diameters. The result is experimentally confirmed by temperature measurement in a realistic injection molding machine instrumented with a reference IR sensor.

Commentary by Dr. Valentin Fuster
2011;():173-180. doi:10.1115/DSCC2011-6092.

In this paper, a new class of Gaussian processes is proposed for resource-constrained mobile sensor networks. Such a Gaussian process builds on a GMRF with respect to a proximity graph over a surveillance region. The main advantages of using this class of Gaussian processes over standard Gaussian processes defined by mean and covariance functions are its numerical efficiency and scalability due to its built-in GMRF and its capability of representing a wide range of non-stationary physical processes. The formulas for Bayesian posterior predictive statistics such as prediction mean and variance are derived and a sequential field prediction algorithm is provided for sequentially sampled observations. For a special case using compactly supported kernels, we propose a distributed algorithm to implement field prediction by correctly fusing all observations in Bayesian statistics. Simulation results illustrate the effectiveness of our approach.

Topics: Sensor networks
Commentary by Dr. Valentin Fuster

Vibration Dynamics and Control

2011;():181-188. doi:10.1115/DSCC2011-5949.

This paper presents a procedure for designing an adaptive H∞ controller to reject sinusoidal disturbances with unknown constant frequencies. The proposed adaptive controller identifies the tonal frequencies in the disturbance and updates the controller to attenuate these frequencies. The controller design builds upon a closed-form solution for a simplified single-frequency attenuation problem. An application to optical beam jitter control illustrates the design procedure and its properties.

Commentary by Dr. Valentin Fuster
2011;():189-196. doi:10.1115/DSCC2011-6072.

Magneto-rheological (MR) dampers that belong to the family of semi-active devices are widely used for vibration attenuation in space and civil engineering structures. In this paper, we study the use of MR dampers for seismic protection of a model two-story structure. A modified Bingham model of the MR damper is considered for linear parameter varying (LPV) modeling and control of the system. The main contribution of the paper is the design and experimental validation of an LPV anti-windup compensator to tackle the effect of actuator saturation on control design performance. The designed LPV anti-windup control scheme is advantageous from the implementation standpoint because it can be considered as an addition to the existing control system. Experimental results demonstrate the effective vibration suppression of the structure in the presence of the seismic excitation inputs by utilizing an LPV control strategy. An inner/outer loop control strategy is further developed and implemented considering the actuator saturation effect to reduce the control effort and saving the MR damper power consumption.

Commentary by Dr. Valentin Fuster
2011;():197-200. doi:10.1115/DSCC2011-6155.

Various quasi-zero stiffness (QZS) passive vibration isolators were developed for protecting a system against vibration. This paper presents active control of a QZS vibration isolator. First, a unified model of QZS passive vibration isolators with symmetric nonlinearity is presented, which is a Duffing equation with odd power polynomials. Based the unified model, vertical and horizontal actuations are proposed for active control of QZS isolators. Finally, active control scheme for a QZS vibration isolator are presented and the isolation performances are compared with simulations.

Topics: Stiffness
Commentary by Dr. Valentin Fuster
2011;():201-208. doi:10.1115/DSCC2011-6160.

In this paper the Active Vibration Control (AVC) of a three-story gantry type structure is designed. AVC is a viable solution to decrease drift (relative displacement between the slabs) in civil structures under seismic excitation. The structure under consideration is composed by six individual gantries. Dynamic models of these individual components are obtained using classical structural dynamic methods and include a reduction of its degree of freedom (DOF). The complete model of the three-story structure is obtained assuming a rigid diaphragm behavior of the slabs. This model is a multi-input, multi-output linear ordinary differential equation (ODE) including nine DOF. The obtained ODE equation is transformed to its equivalent multi-input MIMO state space realization. The state space format is the require representation to easily apply the proposed LQR, LQG and H∞ control techniques. In addition, this work considers the possible location of the PZT actuators within the structure. Real seismic data were used to test the system performance. The controlled system response shows a substantial improvement when compared to the non-controlled structure behavior.

Commentary by Dr. Valentin Fuster
2011;():209-216. doi:10.1115/DSCC2011-6175.

Remote bridge condition monitoring can be accomplished using sensors equipped with transmitters. This allows for timely detection of structural degradation and thereby increases safety. These sensors require a more permanent power source than batteries to enable operation without costly periodic maintenance. Since bridges exhibit structural vibration, electromagnetic vibration energy harvesting is a viable power source candidate. Unfortunately conventional electromagnetic vibration energy harvesters (VEHs) operate efficiently only at their natural frequency. Since a single bridge will vibrate differently, depending on its traffic conditions, the following work proposes a method for augmenting the excitation frequency bandwidth for electromagnetic VEHs, thereby enabling them to be feasible power supplies for bridge condition monitoring. Specifically, the paper provides experimental results showing the effectiveness of the proposed method in inducing resonance type behavior in an electromagnetic VEH at excitation frequencies that otherwise produce little to no motion in the harvester.

Topics: Vibration
Commentary by Dr. Valentin Fuster

Alternative Propulsion Systems Modeling and Control

2011;():217-223. doi:10.1115/DSCC2011-5950.

In this paper, we adopted a simple yet useful capacity fade model of a Li-ion battery for a hybrid electric vehicle application. Its simple structure permitted us to embed it in the formulation of the optimal energy management control problem, which we solved using dynamic programming. In this way, we were able to study the trade-off between globally optimal fuel economy and battery life. This trade-off was quantified in terms of total cost of fuel and battery projected to the entire vehicle life. Furthermore, a sensitivity analysis was performed to investigate the influence of time-varying fuel and battery prices on total cost of ownership.

Commentary by Dr. Valentin Fuster
2011;():225-232. doi:10.1115/DSCC2011-5969.

This paper presents a model predictive control approach to solving the energy management problem within a series hydraulic hybrid powertrain. The hydraulic hybrid utilizes a high pressure accumulator for energy storage which has superior power density than conventional battery technology. This makes fluid power attractive for urban driving applications in which there are frequent starts and stops and large startup power demands. Model predictive control was chosen for control design because this technique requires no information about the future drive cycle, which can be highly uncertain in urban settings. The proposed control strategy was validated experimentally using an electro-hydraulic powertrain testbed which includes energy storage. The experimental study demonstrates the controller’s ability to track a reference trajectory while achieving efficient engine operation.

Commentary by Dr. Valentin Fuster
2011;():233-238. doi:10.1115/DSCC2011-5982.

In predictive adaptive cruise control systems, a major challenge is estimating the future driving pattern of the lead car. This paper proposes an adaptive cruise control system that acts more smoothly and fuel efficiently by utilizing probabilistic information of velocity transition of the front car. The car following problem is formulated in a chance constrained model predictive control framework in which the inter-vehicle gap constraints are enforced probabilistically. The probability distribution of the position of the front car is estimated through a Markov Chain Monte Carlo (MCMC) simulation. The position probability distribution is then utilized to convert the chance constrained MPC problem to a deterministic linear MPC problem. Two case studies with two real driving cycle profiles are presented to show the potential improvement in fuel economy.

Commentary by Dr. Valentin Fuster
2011;():239-246. doi:10.1115/DSCC2011-6043.

This paper discusses the development of a hybrid neural net and physics based battery model. A control oriented one dimensional physics based model of a battery is developed. A neural net based modeling approach is used to compensate for the lack of knowledge of material parameters for the battery cell. Given the knowledge of the physics of the battery, sparse recurrent neural nets are used. Multiple types of standalone neural nets as well as hybrid neural net and physics based battery models are developed and tested to determine the appropriate configuration for an optimal performance. All neural nets are trained as open-loop (feed-forward) systems and are tested as recurrent systems, with the battery state of charge being fed back. The neural nets are trained, tested and validated using test data from a 4.4Ah Boston Power lithium ion battery cell. The modeling approach presented in this paper is able to accurately simulate battery performance for multiple current profiles.

Commentary by Dr. Valentin Fuster
2011;():247-254. doi:10.1115/DSCC2011-6047.

Plug-in electric vehicles (PEVs) are equipped with sizable batteries to replace fossil fuel with electric energy. The electric grid may be stressed when a large number of PEVs charge their batteries, especially during peak hours. This paper presents a PEV charging control algorithm, which achieves balance between the system-level objective (valley filling) and individual-level objective (full battery charging). In addition, grid frequency regulation is achieved through a two-level control algorithm, which ensures robustness and reduces reliance on conventional ancillary services. This two-level control algorithm is scalable because it is designed for decentralized implementation. In other words, the algorithm works with indefinite numbers of PEVs as long as the load is within the grid capacity.

Commentary by Dr. Valentin Fuster
2011;():255-262. doi:10.1115/DSCC2011-6062.

In this paper a new strategy for controlling the power flow in hybrid electric vehicles is described. The strategy focuses in the planetary gear system where kinematic and dynamic constraints must be satisfied. The aim is to satisfy driver demands and to reduce fuel consumption. The resultant power flow control is continuous and uses the internal combustion engine with the maximum possible efficiency. The strategy is not optimal, although it is inspired by the solution to most optimization problems. The main advantages are that the computational cost is low, when compared to optimization based approaches, and that it is easy to tune. The strategy is tested with simulations using a mathematical model of a power train of a hybrid diesel-electric bus subjected to the power demands of representative urban area driving cycles. Simulation results indicate that the strategy achieves small speed tracking errors and attains good fuel consumption reduction levels.

Commentary by Dr. Valentin Fuster
2011;():263-270. doi:10.1115/DSCC2011-6114.

This paper builds on our previous published works in which we had employed nonlinear model predictive control for the (sub)optimal power management of a power-split hybrid electric vehicle (HEV). In addition to the battery’s state of charge, in this work we include the effect of inertial powertrain dynamics in the control-oriented model that are usually ignored because of their fast dynamics. We show how inclusion of the new state removes the need for a separate rule-based strategy for engine start/stop and can result in considerable improvement in the fuel economy as shown by closed-loop simulations over a high-fidelity power-split HEV model.

Commentary by Dr. Valentin Fuster
2011;():271-278. doi:10.1115/DSCC2011-6115.

This paper presents an operational energy optimization method for pure electric ground vehicles (EGVs) based on terrain profile preview. Dynamic programming is applied to obtain the global optimal energy point for torque distributions. The vehicles considered in this paper are powered by batteries and in-wheel motors. Different from the fuel optimization methods of conventional hybrid electric vehicles which adjust battery power supplies, a pure EGV actuated by in-wheel motors has only one energy source, whose power consumption is mainly determined by the torque distributions among in-wheel motors in terms of their operating efficiencies. Moreover, the optimization results show that different terrains will influence the optimal torque distributions and energy savings.

Commentary by Dr. Valentin Fuster
2011;():279-286. doi:10.1115/DSCC2011-6138.

This paper proposes a self-learning approach to develop optimal power management with multiple objectives, e.g. to minimize fuel consumption and transient engine-out NOx and particulate matter emission for a series hydraulic hybrid vehicle. Addressing multiple objectives is particularly relevant in the case of a diesel powered hydraulic hybrid since it has been shown that managing engine transients can significantly reduce real-world emissions. The problem is formulated as an infinite time horizon stochastic sequential decision making/markovian problem. The problem is computationally intractable by conventional Dynamic programming due to large number of states and complex modeling issues. Therefore, the paper proposes an online self-learning neural controller based on the fundamental principles of Neuro-Dynamic Programming (NDP) and reinforcement learning. The controller learns from its interactions with the environment and improves its performance over time. The controller tries to minimize multiple objectives and continues to evolve until a global solution is achieved. The control law is a stationary full state feedback based on 5 states and can be directly implemented. The controller performance is then evaluated in the Engine-in-the-Loop (EIL) facility.

Commentary by Dr. Valentin Fuster
2011;():287-294. doi:10.1115/DSCC2011-6183.

The parallel hybrid electric vehicle (HEV) with a single pre-transmission electric motor (the P2 configuration) has a challenging drivability issue during engine-starts because the motor must simultaneously provide the propulsion torque demand and start the engine. The goal of this study is to present a design process to achieve optimal balance between these two conflicting goals. We first develop a control-oriented HEV powertrain model to accurately predict the engine, clutch, and vehicle dynamics. Assuming that the clutch torque can be accurately estimated and perfectly cancelled, an optimal engine-start control problem is formulated to minimize engine-start time while accurately supplying the driver torque demand simultaneously. This nonlinear constrained optimal control problem is solved numerically using Dynamic Programming (DP). DP results show that a proper constant clutch pressure level, which leads to constant torque disturbance that is easy to cancel out using the motor, can achieve near-optimal performance. Furthermore, the DP control policy is found to be time-invariant, and thus can be directly implemented in the form of a full state feedback controller.

Commentary by Dr. Valentin Fuster
2011;():295-302. doi:10.1115/DSCC2011-6184.

This paper presents a comprehensive assessment of the power that is available for harvesting in the vehicle suspension system and the tradeoff among energy harvesting, ride comfort, and road handing with analysis, simulations and experiments. The excitation from road irregularity is modeled as a stationary random process with road roughness suggested in the ISO standard. The concept of system H2 norm is used to obtain mean value of power generation and the root mean square values of vehicle body acceleration (ride quality) and dynamic tire-ground contact force (road handling). For a quarter car model, analytical solution of the mean power is obtained. The influence of road roughness, vehicle speed, suspension stiffness, shock absorber damping, tire stiffness, wheel and chasses masses to the vehicle performances and harvestable power are studied. Experiments are carried out to verify the theoretical analysis. The results suggest that road roughness, tire stiffness, and vehicle driving speed have great influence to the harvesting power potential, where the suspension stiffness, absorber damping, vehicle masses are insensitive. At 60mph on good and average roads 100–400 watts average power is available in the suspensions of a middle-size vehicle.

Commentary by Dr. Valentin Fuster

Bath/ASME Fluid Power Symposium

2011;():303-310. doi:10.1115/DSCC2011-5954.

This paper discusses the theoretical and experimental results of a position control system using a pneumatic actuator driven by fast switching on/off valves. To begin with, there is a brief introduction about servo-pneumatic systems driven by fast switching valves. Subsequently, the mathematical model of the servo-pneumatic system is discussed involving directional fast switching on/off valves and the flow control valves, modeled according to the mass flow equation based on ISO 6358 and the actuator modeled by Newton’s second law and the continuity equation. The system control is performed through the pulse width modulation (PWM) technique used with a proportional-integrative-derivative (PID) controller. The system performance is analyzed in relation to its application for power control of wind turbines. The simulations of the servo-pneumatic system were carried out by Matlab/Simulink with the experimental results obtained through a test bench.

Commentary by Dr. Valentin Fuster
2011;():311-318. doi:10.1115/DSCC2011-5990.

The frequency bandwidth of conventional electro-hydraulic vibrator is always limited to a rather narrow range. This restriction is attributed not only to the limited response capability of the servo valve but also to the electro-hydraulic servo system itself the bandwidth of which is unable to be extended. Thus, the parallel design of a two-dimensional valve (here within defined as a 2D valve) and a servo valve is adopted to control an electro-hydraulic vibrator in half-close-loop mode so that the frequency bandwidth is greatly extended and the precision control of the vibration characteristic variables such as waveform shape is preserved. The analysis and experimental results demonstrate that the waveform distortion is less than 10%, up to 2000 Hz. It was concluded that the feasibility of the half closed loop design is established for the realization of the high frequency vibration.

Commentary by Dr. Valentin Fuster
2011;():319-326. doi:10.1115/DSCC2011-6121.

This article discusses the application of a novel model-based fault detection method. The method is based on the interacting multiple model (IMM) strategy, which makes use of a finite number of known operating modes. A filter is used in conjunction with the IMM in order to estimate the states and parameters of the system. The smooth variable structure filter (SVSF) is a relatively new estimation strategy, and is based on sliding mode concepts which introduces an inherent amount of robustness and stability. The combined SVSF-IMM strategy is applied on an electrohydrostatic actuator (EHA), which is a device used in the aerospace industry. Two different operating modes were created, based on varying degrees of friction acting on the EHA cylinder. The results of the friction fault detection were compared with the popular Kalman filter (KF) based IMM strategy.

Commentary by Dr. Valentin Fuster
2011;():327-334. doi:10.1115/DSCC2011-6127.

In this paper, we consider the problem of designing an algorithm for estimating the stroke of a pushrod of the air brake system. The stroke of pushrod directly relates to the braking force available at the wheels and also affects the response time. The longer the stroke, the volume available for expansion is larger and correspondingly, the response is slower. The stroke depends on the clearance between the brake pad and the drum, which can vary due to variety of factors such as thermal expansion of drum and mechanical wear. Typical safety inspections of air brakes include the measurement of the stroke of the pushrod of each brake chamber. Regulations on trucks such Federal Motor Vehicle Safety Standard (FMVSS) 121 require the inspection to be carried out at 90 psi supply pressure and at full brake application. The evolution of the brake pressure depends on the stroke of the pushrod and the area of the treadle valve, which is controlled by the driver. The treadle valve meters compressed air from the supply reservoir to the brake chamber. The proposed scheme requires the measurement of pressure and a model for predicting the evolution of brake chamber pressure in response to full application of the brake (brake pedal is fully depressed). We experimentally corroborate the effectiveness of the proposed algorithm.

Topics: Algorithms , Brakes
Commentary by Dr. Valentin Fuster
2011;():335-342. doi:10.1115/DSCC2011-5900.

In this paper, the dynamic characteristics of the three-way hydromechanical valve are studied using numerical and analytical techniques. The full nonlinear model for this valve is derived by taking into consideration the capacitance of the downstream load chamber and the fluid momentum effects acting on the valve. Using non-dimensional analysis it is shown that the higher order dynamics of the valve itself are negligible compared to other forces acting on the valve, and that the transient flow-forces are also of negligible size. Using this information, a reduced order model is developed for studying the bandwidth frequency in closed form and for employing perturbation techniques to identify the relative impact of various design changes for the valve. In conclusion it is shown that the most significant design change that can be made for increasing the bandwidth frequency of the valve is to reduce the load capacitance by either shrinking the load-chamber volume or increasing the fluid bulk modulus. The impact of other design perturbations is also discussed including the effect of increasing the overall valve size, the supply pressure to the valve, and the mechanical spring rate.

Commentary by Dr. Valentin Fuster
2011;():343-350. doi:10.1115/DSCC2011-5940.

Servovalves are compact, accurate, fast flow modulating valves. However, cost reduction pressures exist, not least due to the electomagnetically actuated pilot stage. This paper describes a servovalve with a jet deflector pilot stage actuated by a multilayer piezoelectric bimorph. The electrical power and voltage requirements are relatively low (+/−30V), and mechanical spool feedback is used as opposed to the more complex electrical feedback alternative. A mathematical model of the valve is presented, which is used to simulate its performance. Finite element analysis is used to model the bimorph actuator and the feedback wire assembly to verify an Euler-Bernoulli beam analysis. A Moog 26 Series servovalve is used as a basis for the prototype. Experimental test results are in good agreement with the simulation results. The high order nonlinear model is also approximated by a first order transfer function to identify the parameters that dictate the main design tradeoffs.

Commentary by Dr. Valentin Fuster
2011;():351-354. doi:10.1115/DSCC2011-5947.

When using an independent metering valve configuration with a multivariable control concept the position and the rod side pressure level and therefore the pressurization level of the cylinder drive can be controlled. By controlling the rod side pressure and the position separately the operating area of the cylinder drive changes dependent on the actual nominal rod-side pressure. In order to reach every operation point with optimal stiffness a nominal pressure generation dependent on the actual operating point is needed. In this article a novel idea how to realize such a logic control for the optimal nominal pressurization level generation is presented. The main focus in this paper is the principal structure of such a logic control. The functionality is shown by simulation and measurement results of a differential cylinder drive.

Topics: Valves
Commentary by Dr. Valentin Fuster
2011;():355-362. doi:10.1115/DSCC2011-5972.

This paper presents the design of a flow control valve in which the area-schedule i.e., the relationship between the spool position and the metering orifice area, can be chosen arbitrarily. The motivation for such a flow controller comes from a novel hydraulic actuator intended for camless valve actuation in internal combustion engines. The proposed actuator has a unique internal feedback system in which the motion of the flow regulator’s spool is hydro-mechanically coupled to the motion of the actuator. Lack of direct control over the spool position necessitates the appropriate modification of the area-schedule in order to control the variation of the orifice area. The design modifications required to realize the area-schedules are first discussed. A systematic procedure which combines CFD analysis and geometry based analysis is then developed to characterize the variation of the effective orifice area for various spool designs. The proposed analysis procedure is validated with experimental data from a prototype spool valve. The fast turn around time of the proposed analysis technique is used to develop an automated procedure to design the 3D features on the spool required to realize a specified area-schedule.

Commentary by Dr. Valentin Fuster
2011;():363-366. doi:10.1115/DSCC2011-6007.

This article describes how on/off valves can be used in the closed loop control. After studying different industrial hydraulic applications typical requirements will be compiled. From these requirements a basic on/off valve architecture will be derived. In this article the number of parallel connected on/off valves per control edge is limited to one valve. So the typical digital hydraulics with a high number of parallel connected valves is not considered. The on/off valves could not be controlled like normal directional valves. To control on/off valves digital control methods are common to use. The digital control methods are not optimized for the hydraulic on/off valves, because the origins of the digital control methods will be found in the electrical engineering. The novel idea for the digital control methods is the optimized pulse control which eliminates the disadvantages of the common digital control methods. How the optimized pulse control works is described in this article.

Topics: Valves
Commentary by Dr. Valentin Fuster

Automotive Systems Modeling and Control

2011;():367-374. doi:10.1115/DSCC2011-5938.

Prior experiments have confirmed that specific terrain-based localization algorithms, designed to work in GPS-free or degraded-GPS environments, achieve vehicle tracking with tactical-grade inertial sensors. However, the vehicle tracking performance of these algorithms using low-cost inertial sensors with inferior specifications has not been verified. The included work identifies, through simulations, the effect of inertial sensor characteristics on vehicle tracking accuracy when using a specific terrain-based tracking algorithm based on Unscented Kalman Filters. Results indicate that vehicle tracking is achievable even when low-cost inertial sensors with inferior specifications are used. However, the precision of vehicle tracking decreases approximately linearly as bias instability and angle random walk coefficients increase. The results also indicate that as sensor cost increases, the variance in vehicle tracking error asymptotically tends to zero. Put simply, as desired precision increases, increasingly larger and quantifiable investment is required to attain an improvement in vehicle tracking precision.

Topics: Sensors , Vehicles
Commentary by Dr. Valentin Fuster
2011;():375-381. doi:10.1115/DSCC2011-6085.

The development of new stochastic terrain gridding methods are necessitated by new tire and vehicle modeling applications. Currently, grid node locations in the horizontal plane are assumed to be known and only the uncertainty in the vertical height estimates is modeled. This work modifies the current practice of weighting the importance of a particular measured data point (the terrain height at some horizontal location) by the inverse distance between the grid node and that point. A new weighting function is developed to account for the error in the horizontal position of the grid nodes. The geometry of the problem is described and the probability distribution is developed in steps. Although the solution cannot be determined in closed form, an estimate of the median distance is developed within 1% error. This more complete stochastic definition of the terrain can then be used for advanced tire modeling and vehicle simulation.

Commentary by Dr. Valentin Fuster
2011;():383-390. doi:10.1115/DSCC2011-6165.

A dynamic model for the crankshaft/connecting-rod/piston-assembly for a single cylinder engine is developed. The model considers the rigid body motion of the crank-slider mechanism including the piston secondary motions such as the piston-slap and piston-tilting. The formulation considers the ring to have three rigid body degrees of freedom in addition to its longitudinal and in-plane transverse deformations. The structural flexibility terms are approximated by using curved beam finite element method. The dynamic model has a variable structure whereby the number of degrees of freedom depends on the piston-liner and piston-ring interactions. Its formulation does not include frictional losses. The simulation results illustrate the piston secondary motions along with the ring tilting angles relative to the piston orientation for the total duration of the engine cycle. In addition, they exhibit the translational motion of the ring within the piston groove.

Commentary by Dr. Valentin Fuster
2011;():391-398. doi:10.1115/DSCC2011-6166.

The comparison between PD de-centralized and centralized control for an EMS maglev system that uses combined magnets with an inverted U-rail for levitation and guidance is presented. A simple 2-DOF maglev system model (rigid and flexible body cases) that comprises heave and lateral modes is used. The comparison is based on two aspects. First, by sketching the multi-input multi-output (MIMO) root loci with every controller individually for system rigid and flexible body cases. Second, a gradient-like search algorithm based on an optimal criterion for PD de-centralized and centralized controllers’ gains tuning is used. Simulation results of the maglev system with these tuned controllers show that the centralized control scheme is capable of lateral displacements suppression that may result from disturbing lateral forces than the de-centralized one.

Commentary by Dr. Valentin Fuster
2011;():399-406. doi:10.1115/DSCC2011-6173.

The stability control problem of single-unit commercial vehicles, from large pickup trucks to motorcoaches, is analyzed. Detailed dynamic models of four baseline vehicles are used to estimate the feasibility and the performance limits of rollover stability and directional stability control systems for such vehicles on dry and wet roads.

Commentary by Dr. Valentin Fuster

Bio-Inspired Systems

2011;():407-410. doi:10.1115/DSCC2011-5953.

Members of the animal kingdom produce motion by muscle contraction. Biological muscle can be viewed as a unidirectional actuator. To achieve bidirectional motion, each muscle has a corresponding antagonist muscle whose contraction produces motion in the opposite direction. This gives biological systems the unique ability to modulate the stiffness of a joint, which is important when interacting with the environment. Certain bio-inspired robotic systems incorporate antagonistic pairs in an attempt to produce similar desirable properties. The cellular actuator employs nested compliant mechanisms to produce human-scale motion from piezoelectric stack actuators, which on their own have a small displacement. The expression for the stiffness of the actuator composed of these mechanisms takes the form of a continued fraction, which results from the nested structure. In this way, the stiffness can be easily approximated to a desired degree of accuracy by considering only the outermost mechanisms.

Commentary by Dr. Valentin Fuster
2011;():411-418. doi:10.1115/DSCC2011-6023.

Smart materials have been used in various applications. In this paper, a walking robot with six two-degree-of-freedom (2-DOF) legs made of ionic polymer metal composite (IPMC) is designed and implemented. Each leg can work as both a supporter and a driver, closely mimicking a real insect. To support and drive the robot, thicker (around 1 mm in thickness) IPMC strips were fabricated and used, and a 0.2-rad/s square wave is given as an input signal. The IPMC strips exhibit better performance in response to the square wave (8 mm) than sawtooth (4 mm) and sinusoidal (6 mm) waves in deflection. By applying this input signal in sequence, all the IPMC strips bend and walk in the form of six legs. In addition, thin magnet wires were used to supply power to each strip to prevent from confining the motion of our robot. Six lower legs are divided into two groups that work in the opposite directions to move the robot forward by turns. Upper legs are also divided into two groups to lift up their lower legs from making the robot to move back to the same place. The sizes of the IPMC strips and our robot (102 mm × 80 mm × 43 mm) were decided to exhibit better performance (0.5 mm/s) according to our tests.

Commentary by Dr. Valentin Fuster
2011;():419-425. doi:10.1115/DSCC2011-6042.

For those who have lost the ability to walk due to paralysis or other injuries, eLEGS, a mobile robotic exoskeleton, offers the chance to walk again. The device is a mobile exoskeleton with actuated sagittal plane hip and knee joints which supports the user and moves their legs through a natural gait. The device uses a multi-leveled controller that consists of a state machine to determine the user’s intended motion, a trajectory generator to establish desired joint behavior, and a low level controller to calculate individual joint controller output. The system can be controlled by a physical therapist or can be controlled by the user. Subject testing results are presented from a seven subject pilot study including patients with complete and incomplete injuries. The testing resulted in six of the seven subjects walking unassisted using forearm crutches after a single two hour testing session.

Topics: Testing , Wounds , Spinal cord
Commentary by Dr. Valentin Fuster
2011;():427-434. doi:10.1115/DSCC2011-6108.

Nature provides various alternative locomotion strategies which could be applied to robotic systems. One such strategy is that of jumping, which enables centimeter to millimeter-scaled insects to traverse highly unstructured environments quickly and efficiently. These insects generate the required high magnitude power through specialized structures which store and rapidly release large amounts of energy. This paper presents an investigation into the morphology of natural jumpers and derives a generalized mathematical model based on them. The model describes mathematically the relationships present in a jumping system which uses a pause-and-leap jumping strategy. The use of springs as energy storage elements for such a jumping system is assessed. The discussion is then further extended to another bioinspired approach that can be applied to a jumping robot: that of gliding using foldable wings. The developed jumping and gliding mobility paradigm is analyzed and its feasibility for mobile robot applications is discussed.

Commentary by Dr. Valentin Fuster
2011;():435-442. doi:10.1115/DSCC2011-6125.

The objective of this paper is to develop a simple active control scheme for a kneeless bipedal walking robot. Our control objective is to generate asymptotically stable walking gait pattern by impulsive control actions at the feet of the robot. Consequently, a nonlinear impulsive controller is designed for the hybrid dynamics of the system that can actively generate a wide range of desired ballistic walking gaits on a flat surface. Furthermore, the role of continuous control in the closed loop dynamics of the system is investigated. We show a combination of continuous and impulsive control can maintain a non-scuffing bipedal walking on arbitrary ground slope angles. Our results show that the controlled systems successfully generate active gaits without significant degradation in the energetic efficiencies of the equivalent passive systems.

Commentary by Dr. Valentin Fuster
2011;():443-450. doi:10.1115/DSCC2011-6174.

Snakes are one of the world’s most versatile locomotors, at ease slithering through rubble or ratcheting up vertical tree trunks. Their adaptations for movement across complex dry terrain thus serve naturally as inspirations for search-and-rescue robotics. In this combined experimental and theoretical study, we perform experiments on inclined surfaces to show a snake’s scales are critical anatomical features that enable climbing. We find corn snakes actively change their scale angle of attack by contracting their ventral muscles and lifting their bodies. We use this novel paradigm to design Scalybot, a two-link limbless robot with individually controlled sets of belly scales. The robot ascends styrofoam plates inclined up to 45°, demonstrating a climbing ability comparable to that of a corn snake in the same conditions. The robot uses individual servos to provide a spatial and temporal dependence of its belly friction, effectively anchoring the stationary part of its body while reducing frictional drag of its sliding section. The ability to actively modulate friction increases both the robot’s efficiency over horizontal surfaces and the limiting angles of inclination it can ascend.

Topics: Friction , Robots
Commentary by Dr. Valentin Fuster

Biomechatronics and Medical Robotics

2011;():451-458. doi:10.1115/DSCC2011-5998.

This paper presents experimental results of a myoelectric impedance controller designed for reciprocal stair ascent with an active-knee powered transfemoral prosthesis. The controller is modeled from non-amputee (normal) motion capture data, estimating knee torque with a linear two-state (stance/swing) impedance control form that includes proportional myoelectric torque control. The normal gait model is characterized by small stiffness and damping in both stance and swing, a low angle set-point in stance, a high angle set-point in swing, and proportional myoelectric control in stance but not swing. Clinical tests with a single unilateral transfemoral amputee indicate good performance of the controller; however, subject feedback suggests a reduction in the extensive myoelectric torque parameter and the need for constant, balanced myoelectric torque parameters in both stance and swing. Average prosthesis knee joint kinetics from a stairwell test using the amputee-tuned controller compare favorably with non-amputee gait data.

Commentary by Dr. Valentin Fuster
2011;():459-462. doi:10.1115/DSCC2011-6027.

Force sensing is an important component for a number of surgical procedures as it can help to prevent undesirable damage to the tissue and at the same time provides the surgeons with a better “feel” of the tool-tissue interaction. However, most of the current commercially available multi-DOF force sensors are relatively large in size and it is a challenge to incorporate them into the surgical tool. Hence, a multi-DOF miniature force sensor is desired and this paper presents the design and development of a miniature 2-DOF force sensor. In order to achieve a miniature force sensor, microfabrication technique is used and the proposed force sensor is a capacitive-based sensor. The proposed force sensor can be used in a number of percutaneous procedures as well as catheter-based procedures. This paper presents the design and microfabrication process of the proposed miniature force sensor.

Commentary by Dr. Valentin Fuster
2011;():463-470. doi:10.1115/DSCC2011-6065.

A novel hybrid (i.e., discrete/continuous) control system is studied on a cellular microinjector technology called the Ros-Drill© (Rotationally Oscillating Drill). Ros-Drill© is developed primarily for ICSI (Intra-Cytoplasmic Sperm Injection). It is an inexpensive set-up, which creates high-frequency rotational oscillations at the tip of an injection pipette tracking a harmonic motion profile. These rotational oscillations enable the pipette to drill into cell membranes with minimum biological damage. Such a motion control procedure presents no particular difficulty when it uses sufficiently precise motion sensors. However, size, costs and accessibility of technology on hardware components may severely constrain the sensory capabilities. Then the trajectory tracking is adversely affected. In this paper we handle such a practical case, and present a novel adaptive-hybrid control logic to overcome the hurdles. The control is implemented using a commonly available microcontroller and extremely low-resolution position measurements. First, the continuous control system is analyzed and designed. Then, an adaptive, robust and optimal PID (proportional-integral-derivative) control strategy is performed. We demonstrate via simulations and experiments that the tracking of the harmonic rotational motion is achieved with desirable fidelity.

Commentary by Dr. Valentin Fuster
2011;():471-478. doi:10.1115/DSCC2011-6067.

This paper describes the design and control of a compact and flexible pneumatic artificial muscle (PAM) actuation system for bio-robotic systems. The entire paper is divided into two parts, with the first part covering the mechanism design and the second part covering the corresponding controller design. This novel system presented in this part incorporates two new features, including a variable-radius pulley based PAM actuation mechanism, and a spring-return mechanism to replace the PAM in the “weak” direction. With the pulley radius as a function of the joint angle, this new feature enables the designer to freely modulate the shape of the torque curve, and thus achieves a significantly higher flexibility than the traditional configuration. The other new feature, the spring-return mechanism, is inspired by the fact that a large number of bio-robotic systems require a significantly larger torque in one direction than the other.

Topics: Design , Muscle
Commentary by Dr. Valentin Fuster
2011;():479-486. doi:10.1115/DSCC2011-6143.

In this paper, an adaptive block backstepping controller is developed for stimulation of agonist-antagonist muscles by using Functional Electrical Stimulation (FES). FES is an alternative method that stimulates a paralyzed muscle in lieu of inactive motor neurons, which was first implemented in the 1960s ([1]). However, many challenges need to be addressed before clinical trials with Spinal Cord Injury (SCI) patients can be performed, which include designing a controller, modeling muscles, developing interface equipment, just to mention a few. To contribute to the FES control field, a nonlinear controller and its evaluation through a computer simulation are presented.

Topics: Design , Muscle
Commentary by Dr. Valentin Fuster
2011;():487-494. doi:10.1115/DSCC2011-6167.

In this paper, we develop a numerical mixed flexible-rigid body model to take into account the effects of the external disturbances acting on a flexible manipulator secondary to the oscillatory transmitral blood flow in the left ventricle. The manipulator is made of a flexible rubber-like material to further extend the surgical robotic-based catheters’ degrees of freedom and steer-ability in beating-heart prosthetic aortic valve implantation procedure. Along with the developed numerical model, a detailed description of the catheter’s mechanical architecture and the actuation system is also provided. Necessity of employing such a model for the designed system is clearly justified using simulation studies.

Commentary by Dr. Valentin Fuster

Biosystems and Biomedical Applications

2011;():495-501. doi:10.1115/DSCC2011-5924.

This paper describes motion intention classifiers which utilize reaction forces signals from heel and toe; and hip velocity information to predict subject’s intention. Those classifiers using Bayes method to predict (i) walk-to-stop, (ii) walking-speed-changing, and (iii) stop-to-motions. They are very accurate (most of them have accuracy rate higher than 90%) and a significant step in order to develop a multifunctional knee-ankle-foot orthosis.

Topics: Orthotics , Knee
Commentary by Dr. Valentin Fuster
2011;():503-510. doi:10.1115/DSCC2011-6048.

This paper introduces a new seven-mass biomechanical model for the vibration of vocal folds. The model is based on the body-cover layer concept of the vocal fold biomechanics, and segments the cover layer into three masses. This segmentation facilitates the model comparison with the motion of the vocal glottis contour derived from modern high-speed digital imaging systems. The model simulation is compared to experimental data from a pair of healthy vocal folds showing good agreement in the frequency and time domains. The proposed model is also shown to outperform a previously-developed model that does not explicitly account for the body layer dynamics.

Commentary by Dr. Valentin Fuster
2011;():511-518. doi:10.1115/DSCC2011-6082.

Epilepsy is one of the most common disorders of the brain. Currently, studies of epileptic seizures often involve tedious and time-consuming visual inspection of multi-channel long EEG data by medical experts. To better monitor seizures and make medications more effective, we propose a recurrence time based approach to characterize brain electrical activity. Unlike many other nonlinear methods, the proposed approach does not require that the EEG data be chaotic and/or stationary. It only contains a few parameters that are largely signal-independent, and hence, is very easy to use. The method detects epileptic seizures with accuracy close to 100% (when subclinical seizures are not counted) and false alarm rate per hour close to 0. Most critically, the method is very fast: with an ordinary PC (CPU speed less than 2 GHz), computation of the recurrence time from one channel EEG data of duration one hour with sampling frequency of 200 Hz takes about 1 minute CPU time. Therefore, with an ordinary PC, the method is able to process all 28 channels of 1-hour EEG data in about half an hour, and thus faster than the data being continuously collected. The method can also effectively monitor propagation of seizures in the brain. Therefore, it has the potential to be an excellent candidate for real-time monitoring of epileptic seizures in a clinical setting.

Commentary by Dr. Valentin Fuster
2011;():519-525. doi:10.1115/DSCC2011-6132.

This article presents an inversion-based iterative feedforward-feedback (II-FF/FB) approach to achieve high-speed nanomechanical property measurement of soft materials in liquid. Measurement of nanomechanical properties in liquid is needed for studying a wide variety of soft materials, particularly live biological samples such as live cell. Moreover, high-speed nanomechanical property measurement is needed when dynamic evolution of the sample occurs and/or the rate-dependent viscoelasticity of materials is measured. Nanomechanical measurement in liquid, however, is challenged by the adverse effects including the thermal drift effect, the reduction of the signal to noise ratio, and the hysteresis and the vibrational dynamics effects of the piezoelectric actuators (used to position the probe relative to the sample). Thus, the main contribution of the article is the development of the II-FF/FB approach to tackle these challenges. The proposed method is illustrated through force-curve measurement on a poly (dimethylsiloxane) (PDMS) sample in liquid at high-speed. The experimental results demonstrated the efficacy of the proposed approach in achieving high-speed force-curve measurements of soft samples in liquid.

Commentary by Dr. Valentin Fuster
2011;():527-529. doi:10.1115/DSCC2011-6133.

Toxoplasma gondii is a protozoan capable of replicating sexually in cats and asexually in other warm-blooded animals. By using a three dimensional mesh of both the brain and spleen, it is possible to simulate using a computational model to demonstrate the entire life-cycle within an intermediate host of the parasite as it completes the life-cycle using host cells of these organs. A cellular automata model is developed to demonstrate the dynamics of the parasite, where each cell follows the same set of rules for each discrete time-step. This cellular automata model allows for data simulations to be run of the parasite within a mouse and display graphical images and animations.

Commentary by Dr. Valentin Fuster
2011;():531-536. doi:10.1115/DSCC2011-6171.

The enzyme acetyl-CoA carboxylase catalyzes the first committed step in fatty acid synthesis in all organisms. The E. coli form of the carboxyltransferase subunit was recently found to regulate its own activity and expression by binding its own mRNA. By binding acetyl-CoA or the mRNA encoding its own subunits, Carboxyltransferase is able to sense the metabolic state of the cell and attenuate its own translation and enzymatic activity using a negative feedback mechanism. In this paper, this network of interactions is modeled mathematically using mass action kinetics. Numerical simulations of the model show agreement with experimental results.

Commentary by Dr. Valentin Fuster

Dynamics and Control in Medicine and Biology

2011;():537-544. doi:10.1115/DSCC2011-5946.

Reactive oxygen species (ROS) and peroxidase-oxidase (PO) reactions are Janus-faced contributors to cellular metabolism. At low concentrations, reactive oxygen species serve as signaling molecules; at high concentrations, as destroyers of proteins, lipids and DNA. Correspondingly, PO reactions are both sources and consumers of ROS. In the present paper, we study a well-tested model of the PO reaction based on horseradish peroxidase chemistry. Our principal predictions are these: 1. Under hypoxia, the PO reaction can emit pulses of hydrogen peroxide at apparently arbitrarily long intervals. 2. For a wide range of input rates, continuing infusions of ROS are transduced into bounded dynamics. 3. The response to ROS input is hysteretic. 4. With sufficient input, regulatory capacity is exceeded and hydrogen peroxide, but not superoxide, accumulates. These results are discussed with regard to the episodic nature of neurodevelopmental and neurodegenerative diseases that have been linked to oxidative stress and to downstream interactions that may result in positive feedback and pathology of increasing severity.

Commentary by Dr. Valentin Fuster
2011;():545-552. doi:10.1115/DSCC2011-5984.

Electroporation is an effective means to deliver molecules into the cellular cytoplasm, and has been widely applied in both biological research and clinical applications. The process however suffers from low viability and efficiency. In this work, we present a multi-pulse feedback control scheme for enhancing viability and efficiency of electroporation process. We first present a spherical model for cell transmembrane potential (TMP) dynamics to describe the spatial distribution of electric potential on cell membrane. Then we analyze the models and discuss the bifurcation property of the system. Based on the dynamic models, we design a nonlinear model predictive controller (NMPC) to track pore size profiles in electroporation. We demonstrate that by using NMPC, the pore size can be precisely regulated to targeted values for drug delivery applications. The control design is illustrated through simulation examples.

Commentary by Dr. Valentin Fuster
2011;():553-555. doi:10.1115/DSCC2011-6003.

We study electromechanical dynamics in paced cardiac tissue using numerical simulations of a mathematical model that accounts for excitation-contraction coupling as well as mechanoelectrical feedback. A previously developed finite element based parallel platform is adopted. Extensive numerical simulations are carried out on a 2d tissue and a 3d tissue to investigate the influences of various parameters on the stability of propagating cardiac waves, including conduction velocity, pathological scars, contraction, and stretch activated channels.

Commentary by Dr. Valentin Fuster
2011;():557-563. doi:10.1115/DSCC2011-6046.

Designing a system that adequately processes the input and that rejects the effects of disturbance is a central theme in feedback control theory. In this paper, we use the concept of “disturbance rejection” to analyze the oscillatory behavior of p53, a well-known tumor suppressor protein. Our analysis reveals that the p53 oscillation is not completely dictated by the p53-MDM2 negative feedback loop—it is also modulated by periodic DNA repair-related fluctuations. According to our disturbance rejection model, the feedback loop normally filters the effects of noise and fluctuations on p53, but upon DNA damage, it stops performing the filtering function so that DNA repair-related fluctuations can modulate the p53 oscillation. Our analysis suggests that the overexpression of MDM2, observed in many types of cancer, can make the feedback mechanism less responsive to the modulating signals after DNA damage occurs.

Topics: Oscillations
Commentary by Dr. Valentin Fuster
2011;():565-572. doi:10.1115/DSCC2011-6081.

Understanding the causal relation between neural inputs and movements is very important for the success of brain machine interfaces (BMIs). In this study, we perform systematic statistical and information theoretical analysis of neuronal firings of 104 neurons, and employ three different types of fractal and multifractal techniques (including Fano factor analysis, multifractal detrended fluctuation analysis (MF-DFA), and wavelet multifractal analysis) to examine whether neuronal firings related to movements may have long-range temporal correlations. We find that MF-DFA and wavelet multifractal analysis (but not Fano factor analysis) clearly indicate that when neuronal firings are not well correlated with movement trajectory, they do not have or only have weak temporal correlations. When neuronal firings are well correlated with movements, they are characterized by very strong temporal correlations, up to a time scale comparable to the average time between two successive reaching tasks. This suggests that neurons well correlated with hand trajectory experienced a “re-setting” effect at the start of each reaching task. We further discuss the significance of the coalition of those important neurons in executing cortical control of prostheses.

Topics: Motion
Commentary by Dr. Valentin Fuster
2011;():573-577. doi:10.1115/DSCC2011-6083.

Chaos and random fractal theories are among the most important for fully characterizing nonlinear dynamics of complicated multiscale biosignals. Chaos analysis requires that signals be relatively noise-free and stationary, while fractal analysis demands signals to be non-rhythmic and scale-free. To facilitate joint chaos and fractal analysis of biosignals, we report an adaptive multiscale decomposition algorithm, which: (1) can readily remove nonstationarities from the signal, (2) can more effectively reduce noise in the signals than linear filters, wavelet denoising, and chaos-based noise reduction schemes; (3) can readily decompose a multiscale biosignal into a series of intrinsically bandlimited functions; (4) offers a new formulation of fractal and multifractal analysis that is better than the popular detrended fluctuation analysis when a biosignal contains a strong oscillatory component. The effectiveness of the approach is demonstrated by applying it to classify EEGs for the purpose of detecting epileptic seizures.

Topics: Filtration , Chaos , Fractals
Commentary by Dr. Valentin Fuster
2011;():579-586. doi:10.1115/DSCC2011-6084.

Complex systems often generate highly nonstationary and multiscale signals, due to nonlinear and stochastic interactions among their component systems and hierarchical regulations imposed by the operating environments. The further advances in the fields of life sciences, systems biology, nano-sciences, information systems, and physical sciences, have made it increasingly important to develop complexity measures that incorporate the concept of scale explicitly, so that different behaviors of the signals on varying scales can be simultaneously characterized by the same scale-dependent measure. Here, we propose such a measure, the scale-dependent Lyapunov exponent (SDLE), and develop a unified theory of multiscale analysis of complex data. We show that the SDLE can readily characterize low-dimensional chaos and random 1/f α processes, as well as accurately detect epileptic seizures from EEG data and distinguish healthy subjects from patients with congestive heart failure from heart rate variability (HRV) data. More importantly, our analyses of EEG and HRV data illustrate that commonly used complexity measures from information theory, chaos theory, and random fractal theory can be related to the values of the SDLE at specific scales, and useful information on the structured components of the data is also embodied by the SDLE.

Topics: Signals
Commentary by Dr. Valentin Fuster
2011;():587-594. doi:10.1115/DSCC2011-6117.

Neurons in the nervous system communicate by spiking, which activates synaptic connections via the release of neurotransmitter molecules. Modification of the strength of these synaptic connections, known as plasticity, is a mechanism by which networks of neurons can exhibit learning. Previously, a biophysical model of a rodent lateral amygdala was developed that could learn and store auditory fear and extinction memories following classical Pavlovian fear conditioning [1]. We propose a novel reduced order model that preserves the learning capabilities of the detailed model with considerably fewer computations while providing additional insights into the synaptic learning process. To capture the dynamics of individual cells, we propose enhancements to the Wilson-Cowan firing rate model that permit “full” spike frequency adaptation and a non-zero threshold. To incorporate synaptic learning dynamics, we propose a regression technique to capture the nonlinear relationship between firing rate and synaptic [Ca2+ ]. The resulting method provides a general technique to develop neuronal networks that employ [Ca2+ ]-dependent synaptic learning.

Commentary by Dr. Valentin Fuster
2011;():595-601. doi:10.1115/DSCC2011-6134.

Endothelial cells (ECs) create a vascular network with a tubular structure in response to growth factors diffused into the gel and interactions with the surrounding environment. Individual cells migrate in response to all of these cues, leading to network pattern formation. This paper presents a dynamic model predicting EC sprout growth that is tuned to time-lapse experimental cell migration data obtained from microfluidic 3D culture. Simple cell migration equations with just a few parameters are formulated and a Maximum Likelihood estimator is used for estimating model parameters from experimental data. The tuned model is used to evaluate the influence of different sprout elongation rates on cell density in the sprout stalk. This quantitative modeling approach will lead to input shaping and feedback control to optimize sprouting metrics such as stalk cell density.

Commentary by Dr. Valentin Fuster
2011;():603-605. doi:10.1115/DSCC2011-6146.

The mechanical properties of biomaterials have long been a subject of interest for researchers due to their potential in predicting biologically relevant questions, like proliferation of cancer in tissue. A popular technique of estimating material properties of biomaterials is the AFM, which consists of a probe that indents the material of interest. However, region localization for AFM indentation is challenging, especially when probing large sections of the tissue. Furthermore, identifying the point of contact between AFM tip and the specimen on the force-displacement curve involves uncertainties that are difficult to predict. In this work, we try to address these two issues. We use a vision-guided positioning system to achieve region localization, and we use a resistance based-electrical circuit to identify the point of contact between AFM tip and the specimen.

Commentary by Dr. Valentin Fuster
2011;():607-610. doi:10.1115/DSCC2011-6148.

A beat-to-beat variation in cardiac action potential durations (APD) is a phenomenon known as electrical alternans. Alternans desynchronizes depolarization, increases dispersion of refractoriness and creates a substrate for ventricular fibrillation. In the heart, APD alternans can be accompanied by alternans in intracellular calcium ([Ca2+ ]i ) transients. Recently, we demonstrated experimentally that the onset of APD alternans in the heart is a local phenomenon that undergoes complex spatiotemporal dynamics as pacing rate increases. Moreover, the local onset of APD alternans can be predicted by measuring the restitution properties of periodically paced cardiac tissue. The purpose of this research is to investigate the interplay between local onsets of APD and [Ca2+ ]i alternans using 2D simulation of action potential model of cardiac myocytes.

Commentary by Dr. Valentin Fuster

Engine Modeling and System Identification

2011;():611-617. doi:10.1115/DSCC2011-5944.

The prediction of dynamic phenomena in compressible fluids, such as the air path systems of Internal Combustion Engines (ICEs) has seen an enormous growth in the past years. Striving to improve engine performance, fuel economy and emissions has led to the understanding that significant gains can only be achieved if improvements in engine design can be matched by the ability to closely control engine breathing and combustion performance. The current state of the art in the modeling of ICEs air path systems presents two main approaches, namely the high-fidelity, computationally intensive numerical methods and the low-fidelity, calibration intensive lumped-parameter models. This paper introduces a novel approach for modeling unsteady phenomena in compressible fluids that combines the advantages of numerical methods (high accuracy and low calibration effort) with the limited computation time of lumped-parameter models based on ordinary differential equations (ODEs). The approach is here presented for the one-dimensional nonlinear Euler equations for compressible fluid flow systems, which are particularly relevant for modeling the air path systems of internal combustion engines.

Commentary by Dr. Valentin Fuster
2011;():619-626. doi:10.1115/DSCC2011-5992.

The gas exchange process in a modern diesel engine is generally modeled using manufacturer-provided performance maps that describe mass flows through, and efficiencies of, the turbine and compressor. These maps are typically implemented as look-up tables requiring multiple interpolations based on pressure ratios across the turbine and compressor, as well as the turbocharger shaft speed. In the case of variable-geometry turbochargers, the nozzle position is also an input to these maps. This method of interpolating or extrapolating data is undesirable when modeling for estimation and control, and though there have been several previous efforts to reduce dependence on turbomachinery maps, many of these approaches are complex and not easily implemented in engine control systems. As such, the aim of this paper is to reduce turbocharger maps to analytical functions for models amenable to estimation and control.

Topics: Modeling
Commentary by Dr. Valentin Fuster
2011;():627-634. doi:10.1115/DSCC2011-6001.

Accurate calculation of the conditions (i.e., temperature, pressure, and enthalpy) of internal combustion engine cylinder exhaust is critical to the modeling of, and control design development for, gas exchange in modern and future diesel engine systems. In this paper, a physically-based model for cylinder exhaust temperature, pressure, and enthalpy for engines equipped with variable valve actuation is outlined and extensively validated against experimental data from 193 operating points. The model takes the known conditions when the intake valves close and steps through a polytropic compression process, constant pressure combustion process beginning at top-dead center, and a polytropic expansion process to achieve the desired results when the exhaust valves open. To incorporate the flexibility of modulating the intake valve opening and closing, the effective compression ratio is used to establish the conditions when the intake valves close. Experimental model validation, via a unique multi-cylinder diesel engine utilizing fully flexible intake valve actuation, shows that the model captures the influences of all of the model inputs: engine speed, charge flow, total fueling quantity, intake manifold pressure, and effective compression ratio.

Commentary by Dr. Valentin Fuster
2011;():635-642. doi:10.1115/DSCC2011-6015.

Presented in this paper is a Recursive Least Squares (RLS) based algorithm for system identification of Linear Parameter-Varying (LPV) systems. An identification method based on the input-output representation of LPV systems is employed, where the coefficients of the model depend on external parameters assumed to be measurable in real-time. The identification problem is reduced to a problem of linear regression. Application of the proposed method to a quasi-LPV system developed from an intake manifold model of a spark ignition (SI) engine is demonstrated. Simulations performed using the GT-Power simulation tool and experiments performed on a 5.4-L V8 spark-ignition engine are used validate the accuracy of the proposed method.

Commentary by Dr. Valentin Fuster
2011;():643-650. doi:10.1115/DSCC2011-6180.

This paper presents the synthesis of driver inputs for evaluating diesel soot emissions using iterative learning control. Transient soot emissions from diesel engine vehicles are extremely sensitive to driver aggressiveness. Using closed-loop tracking controllers to follow a vehicle over a prescribed drive cycle usually do not account for the fact that drivers potentially adapt their driving styles to a given powertrain design. This work develops an algorithm producing driver input traces that significantly reduces the soot emissions for a given drive cycle, thus providing a consistent basis for evaluating the influence of powertrain design changes on soot emissions. Possible improvements are first explored using conventional optimal techniques and results are obtained using linear programming. It is then shown that a first-order PD-type iterative learning control based algorithm can deliver good performance, substantially reducing the total soot emissions at a fraction of the computational cost.

Topics: Diesel , Soot , Emissions
Commentary by Dr. Valentin Fuster
2011;():651-654. doi:10.1115/DSCC2011-6202.

Identification of the injection timing drift in modern diesel engine is investigated. The cylinder balancing action which is most common in modern diesel engine is utilized in locating the injection timing (injection timing with the minimum cylinder balancing correction). To find the minimum, several approaches are examined, including a parameterization based approach, sliding mode based self optimizing control and the simplex search. The proposed approaches are evaluated with experimental data from a vehicle.

Topics: Fuels , Cylinders
Commentary by Dr. Valentin Fuster

HCCI Engines Modeling and Control

2011;():655-662. doi:10.1115/DSCC2011-5983.

This paper analyzes the stability of the dynamic coupling among the homogeneous charge compression ignition (HCCI) combustion, gas exchange dynamics and the piston dynamics in a hydraulic free piston engine. The intrinsic feedback nature of the system makes the stability of the engine operation difficult to predict. Specifically, output from the gas exchange process of the previous cycle affects the combustion timing of the current cycle. The combustion force and hydraulic loading together determine how far the piston can travel (dead center positions) and provide input for the current cycle gas exchange process. A discrete cycle-to-cycle based model that captures this dynamic coupling is described in the paper. Linearization of the model gives us the insight of the engine stability under any given operation conditions. The stability analysis is validated with Matlab numerical simulation results of a high-order nonlinear model under a specified set of operation conditions. The proposed methodology can be transformed and applied to analyze stability of systems with similar dynamics couplings.

Commentary by Dr. Valentin Fuster
2011;():663-670. doi:10.1115/DSCC2011-6020.

The combustion mode transition between spark ignition (SI) and homogeneously charged compression ignition (HCCI) combustions of an internal combustion (IC) engine is challenging due to the distinct engine operational parameters over these two combustion modes and the cycle-to-cycle residue gas dynamics of the HCCI combustion. The control problem becomes even more complicated when multi-cylinder operation is involved. This paper studies the combustion mode transition problem of a multi-cylinder IC engine with dual-stage valve lifts and electrical variable valve timing systems. A control oriented engine model was used to develop a multistep mode transition control strategy via iterative learning for combustion mode transition between SI to HCCI with minimal engine torque fluctuations. The hardware-in-the-loop (HIL) simulations demonstrated the effectiveness of the developed control strategy for the combustion mode transition under both constant load and transient engine operational conditions.

Commentary by Dr. Valentin Fuster
2011;():671-676. doi:10.1115/DSCC2011-6036.

Increasing need for clean and efficient energy conversion devices in the automotive sector has pushed technologies such as homogeneous charge compression ignition (HCCI) to the forefront of research. While it offers a number of major advantages in terms of efficiency and emissions, HCCI has significant challenges to overcome before it can be used in a production setting. Among those challenges is the need to control the start of combustion, and one method for doing that is to use variable valve timing to utilize hot, residual exhaust gas. Critical to these residual-affected methods is the role of mixing on the thermodynamic state. This work presents a method for validating a control-oriented two-zone mixing model for HCCI using CFD software. A method of discretizing, extracting, and analyzing the results of the CFD simulation is described, and it is compared with the two-zone model. There is strong correlation, but there is also evidence that the two-zone model needs to be further developed to fully capture the dynamics of mixing.

Commentary by Dr. Valentin Fuster
2011;():677-684. doi:10.1115/DSCC2011-6050.

A zero dimensional, mean-value, control-oriented model for recompression homogeneous charge compression ignition (HCCI) combustion with two discrete states representing temperature and composition dynamics is presented. This model captures steady state magnitudes and trends in combustion phasing, residual gas fraction, and mass flows caused by sweeps in valve timings, fueling rate, and fuel injection timing. It is shown that the coupling of the composition state with the mainly thermally-driven combustion dynamics causes competing slow and fast dynamics that shape the transient response of the phasing. A decoupled version of the model, where composition does not affect combustion phasing, is also developed in an effort to further simplify the model. This version matches the steady state fidelity of the coupled model, but has a qualitatively different dynamical behavior. Both models exhibit complex behaviors such as limit cycles at extremely late phasing. Both realizations are valid contenders as low order steady state representations of HCCI behavior. High-fidelity transient data will be necessary to further clarify the necessity of including composition effects on combustion phasing.

Commentary by Dr. Valentin Fuster
2011;():685-692. doi:10.1115/DSCC2011-6091.

Late-phasing homogeneous charge compression ignition (HCCI) operating conditions have the potential to expand the useful operating range of HCCI. However, these conditions exhibit significant variation in combustion timing and work output from one cycle to the next. Cyclic variations in the combustion timing of HCCI combustion at late-phasing operating conditions can be removed through the use of cycle-to-cycle control of pilot fuel injection timing. A nonlinear, discrete-time model of the recompression HCCI process captures the oscillations in late-phasing HCCI; when this model is linearized, it represents these dynamics as a pole on the negative real-axis. A simple lag compensator eliminates the oscillations in combustion phasing and drastically improves the operability of late-phasing HCCI. in both simulation and experiment.

Commentary by Dr. Valentin Fuster
2011;():693-700. doi:10.1115/DSCC2011-6118.

The probability distribution shape of the ignition timing ensemble allows detection of unstable operation near misfire in Homogeneous Charge Compression Ignition (HCCI) engines. The acceptable range of cyclic variation in HCCI combustion timing is determined by linking the experimental measurements with the shape factors of Generalized Extreme Value (GEV) probability distribution. A combined physical-statistical model is incorporated to analyze the range of cyclic variability in CA50 (crank angle of 50% mass fraction fuel burnt) for two single-cylinder engines. The model is validated with the experimental data at 227 operating points with five different Primary Reference Fuels (PRF). Good agreement between simulation and the experiment with an average error of 0.36 crank angle degree for predicting standard deviation of CA50 is obtained. Low, medium and high cyclic variability zones are identified as a function of intake manifold pressure, equivalence ratio, and intake manifold temperature. This information can be integrated into the design of an engine controller strategy to maintain acceptable levels of cyclic variation during a commanded engine load change.

Commentary by Dr. Valentin Fuster

Modeling and Estimation for Automotive and Energy Systems

2011;():701-708. doi:10.1115/DSCC2011-5951.

Alternative energy storage systems (AESS) are receiving considerable interest today for low-cost mild-hybrid vehicles where the electrical system is substituted with mechanical or hydraulic energy storage. As these technologies are being explored, simulation tools become helpful to predict the behavior of the energy storage system during vehicle use, as well as to conduct comparative studies evaluating the energy and power density, fuel economy improvement, system weight and costs. This paper presents an energy-based modeling approach to characterize the low-frequency dynamic behavior of alternative energy storage systems for hybrid vehicle applications, with the ability to predict the energy flows and sources of energy loss during driving operations. The model aims at evaluating the potential, in terms of efficiency and fuel economy improvement, offered by non-electrified energy storage systems, such as mechanical (flywheels) or hydraulic (accumulators). The modeling tool developed is able to provide a characterization of the performance of each of the two systems starting from a characterization of the components energy conversion behavior. The paper includes a simulation study where the performance of a mechanical and hydraulic energy storage system are compared on a forward-oriented hybrid vehicle simulator, with the objective of characterizing and comparing the energy recuperation process and the energy efficiency of the two systems.

Commentary by Dr. Valentin Fuster
2011;():709-716. doi:10.1115/DSCC2011-6013.

Lithium-ion batteries continue to garner interest as an energy storage system in stationary and vehicular applications. Considerable research effort is currently devoted to investigating the physical and chemical phenomena leading to aging, namely internal resistance growth and capacity fade. This paper presents a reduced-order model that characterizes the dynamic behavior of a Lithium-ion battery cell. The model is derived from the governing electrochemical principles and is applied to a Li-ion cell based upon a natural graphite negative electrode and iron phosphate positive electrode. The paper describes the modeling approach and equations, followed by a validation with experimental data. A sensitivity analysis is then conducted to investigate the influence of the model parameters on the cell internal resistance and capacity. The results of this study allows one to identify a subset of model parameters that may evolve throughout the battery’s life, providing guidance towards establishing which parameter trajectories must be quantified as batteries age.

Commentary by Dr. Valentin Fuster
2011;():717-724. doi:10.1115/DSCC2011-6024.

The accelerated durability test can be achieved by using the GlyghWorks software through the process of Mission Profile and Test Synthesis. Mission profile comprises a number of load events that simulate the real world driving conditions, which contain dynamic information about the vehicle. Test Synthesis is accomplished with extracting the information on the frequency of the events expected in the service life. The challenge of accurately separating the time series signal that correlate the load events becomes essential in order to efficiently conduct the accelerated durability test analysis. In this paper, a method is developed for automatic event identification from a time series based on wavelet analysis, clustering, and Fourier analysis, which are applied respectively to denoise, cluster and refine the segmentation of respective events embedded in the time series signal. The identified events and their corresponding data are used to generate the accelerated durability testing profiles. A systematic approach of acceleration time series generation for the partial test based on the durability testing PSD profile will also be shown, and then the created loading profile can be directly applied as the driven profile in the lab test. Subsequently, method of validation process for the partial test will be introduced.

Commentary by Dr. Valentin Fuster
2011;():725-731. doi:10.1115/DSCC2011-6025.

The localization of vehicles on roadways without the use of a GPS has been of great interest in recent years and a number of solutions have been proposed for the same. The localization of vehicles has traditionally been divided by their solution approaches into two different categories: global localization which uses feature-vector matching, and local tracking which has been dealt by using techniques like Particle Filtering or Kalman Filtering. This paper proposes a unifying approach that combines the feature-based robustness of global search with the local tracking capabilities of a particle filter. Using feature vectors produced from pitch measurements from Interstate I-80 and US Route 220 in Pennsylvania, this work demonstrates wide area localization of a vehicle with the computational efficiency of local tracking.

Commentary by Dr. Valentin Fuster
2011;():733-740. doi:10.1115/DSCC2011-6045.

This paper proposes a payload parameter estimation method for lightweight vehicles (LWVs), whose dynamics and control are substantially affected by their payload variations due to the LWVs’ significantly reduced sizes and weights. Accurate and real-time estimation of payload parameters, including payload mass and its onboard planar location, will be helpful for controller designs and load condition monitoring. The proposed payload parameter estimator (PPE) is divided into two parts: tire nominal normal force estimator (NNFE) based on a recursive least squares (RLS) algorithm using signals measured from LWV constant speed maneuvers, and parameter calculator based on estimated nominal normal forces. The prototype LWV is an electric ground vehicle with separable torque control of the four wheels by in-wheel motors, which allow redundant input injections in the designed maneuvers. Simulation results, based on a CarSim® model, show that the proposed PPE is capable of accurately and quickly estimating payload parameters, and is independent of the road condition as long as the tire forces are kept within their linear ranges.

Topics: Vehicles
Commentary by Dr. Valentin Fuster

Powertrain Systems Modeling and Control

2011;():741-748. doi:10.1115/DSCC2011-5937.

In this paper, a series of closed-loop system identification tests was performed for a variable valve timing cam phaser system on a test bench to obtain a family of linear models for an array of engine speeds and oil pressures. Using engine speed and oil pressure as the system parameters, the family of linear models was translated into a linear parameter varying (LPV) system. The engine speed and oil pressure can be measured in real-time by these sensors equipped on the engine, thus allowing their use as scheduling parameters. An observer-based gain-scheduling controller for the obtained LPV system is then designed based on the numerically efficient convex optimization or linear matrix inequality (LMI) technique. Test bench results show the effectiveness of the proposed scheme.

Topics: Engines , Actuators
Commentary by Dr. Valentin Fuster
2011;():749-754. doi:10.1115/DSCC2011-5941.

A gearshift controller for automatic transmission is developed in this paper, which regulates turbine speed to track the desired trajectory and suppresses the output shaft oscillation at the same time by single clutch pressure control. In order to achieve this task, the proposed controller is designed, by integrating backstepping technique and sliding mode output feedback control theory. Simulation of 1–2 upshift is illustrated to verify the validity of the proposed gearshift controller.

Commentary by Dr. Valentin Fuster
2011;():755-762. doi:10.1115/DSCC2011-5965.

High automotive hydrocarbon emission during cold start is a well recognized challenge with increasing importance in moving towards green vehicles. In this work the application of a linear model reduction technique on the design of a controller for a nonlinear system is discussed. A reduced order cold start model of an SI engine and aftertreatment system is realized using a balanced truncation technique. Sliding mode controllers, derived from a nonlinear physical model and the reduced order model, are designed to reduce tailpipe hydrocarbon emissions. The comparison results indicate the controller derived from the balanced truncated model performs better since it adjusts the control inputs such that it favors the certain desired trajectories which are more influential on the final control target.

Commentary by Dr. Valentin Fuster
2011;():763-768. doi:10.1115/DSCC2011-6004.

Most modern spark ignited (SI) internal combustion engines maintain their air-to-fuel ratio (AFR) at a desired level to maximize the three-way catalyst conversion efficiency and to extend its life. However, maintaining the engine AFR during its transient operation is quite challenging due to rapid changes of driver demands. Conventional transient AFR control is based upon the inverse dynamics of the engine port-fuel-injection well-wetting dynamics and the measured mass air flow rate. This paper develops a dynamic linear quadratic (LQ) tracking controller to regulate the AFR using a control oriented model of the wall wetting dynamics of a port fuel injector (PFI) and estimated transport delays of the airflow travel and throttle dynamics. The LQ tracking controller is designed to optimally track the measured airflow through the throttle during engine transients over a given time interval. The performance of the optimal LQ tracking controller was compared with the conventional inverse fueling dynamics through simulations and showed improvement over the baseline controller.

Commentary by Dr. Valentin Fuster
2011;():769-776. doi:10.1115/DSCC2011-6019.

The paper presents a dynamic model of a dual clutch lever-based electromechanical actuator. Bond graph modeling technique is used to describe the clutch actuator dynamics. The model is parameterized and thoroughly validated based on the experimental data collected by using a test rig. The model validation results are used for the purpose of analysis of the actuator behavior under typical operating modes.

Commentary by Dr. Valentin Fuster

Surgical and Rehabilitation Robotics

2011;():777-782. doi:10.1115/DSCC2011-5976.

This paper presents identification and modeling of dynamic ankle mechanical impedance in relaxed muscles. A multi-variable estimation method using a wearable therapeutic robot enabled clear interpretation of dynamic ankle impedance both in the sagittal and frontal planes. Estimation results showed that dynamic ankle behavior apparently cannot be reconciled with a simple 2nd order model. Measurements in a seated and standing position verified that ankle impedance changes substantially depending on lower-limb posture. Identification results were fitted with a modified Hill model with a mass between the muscle and tendon elements. When coupled with foot inertia, either singly or antagonistically, this model successfully captured the dynamic behavior of the ankle impedance both in the seated and standing positions up to 20 Hz. At least a 4th order model having 2 complex zero and 1 complex pole pairs was required to describe relaxed ankle impedance either in the sagittal or frontal plane up to 20Hz. In the seated position, a 6th order model was slightly better than the 4th order model but with the expense of complexity, and a 8th order model might be used to describe dynamic ankle behavior up to 30∼40Hz.

Commentary by Dr. Valentin Fuster
2011;():783-790. doi:10.1115/DSCC2011-5977.

An energy-efficient circuit for driving PZT actuators using a charge recovery technique is explored in this paper. Mobile and wearable devices for physical assists requiring extended battery life and/or minimal battery weight will benefit from this technology. PZT is a capacitive transducer and can produce constant force for extended periods with little power consumption. Furthermore, energy can be saved by moving charge from one PZT stack to another instead of draining to ground. This paper describes an efficient charge-recovery circuit that can capture 40∼65% of the energy in one PZT unit by transferring it to another PZT unit. The battery then must only supply the remaining charge thereby increasing battery life. First, the basic principle of the charge-recovery technique is described. The idealized circuit architecture and possible implementation are presented. Then, the electrical circuit behavior is analyzed. PZT hysteresis is discussed, and results are presented.

Commentary by Dr. Valentin Fuster
2011;():791-794. doi:10.1115/DSCC2011-6029.

Control of surgical continuum robot manipulators and steerable needles requires accurate real-time sensing of tip position and/or shaft shape. Medical image feedback provides the most straightforward and widely available method of measuring device and clinical target positions and shapes during insertion or tissue manipulation. In this paper we present a method for automatic robot/needle curve segmentation from fluoroscopic images, as well as a method for 3D reconstruction of the curve using biplane fluoroscopy images.

Commentary by Dr. Valentin Fuster
2011;():795-797. doi:10.1115/DSCC2011-6075.

Brain tumors are among the most feared complications of cancer and their treatment is challenging due to the lack of good continuous imaging modality during the procedure and the inability to remove the complete tumor due to obstructions. A highly dexterous, small cross-section robot is being developed to overcome these limitations. The robot is being designed to remove the tumor which is outside the direct “line-of-sight” of the physician. In this paper, we report the design of a Minimally Invasive Neurosurgical Intracranial Robot (MINIR) using a tendon-driven mechanism. In the current prototype presented in this paper, the actuators for actuating the robot are not MRI compatible. The primary goal of this paper is to evaluate the performance of the robot motion and not the MRI compatibility of the entire system. The robot contains four links and four revolute joints. Pulleys and cables are put inside the robot to make the robot compact. The four revolute joints are placed orthogonally to have out-of-plane motion capability and can be controlled independently.

Commentary by Dr. Valentin Fuster
2011;():799-806. doi:10.1115/DSCC2011-6154.

Nonlinearities inherent in soft-tissue interactions create roadblocks to realization of high-fidelity real-time haptics-based medical simulations. While finite element (FE) formulations offer greater accuracy over conventional spring-mass-network models, computational-complexity limits achievable simulation-update rates. Direct interaction with sensorized physical surrogates, in offline or online modes, allows a temporary sidestepping of computational issues but hinders parametric analysis and true exploitation of a simulation-based testing paradigm. Hence, in this paper, we develop Radial-Basis Neural-Network approximations, to FE-model data within a Modified Resource Allocating Network (MRAN) framework. Real-time simulation of the reduced order neural-network approximations at high temporal resolution provided the haptic-feedback. Validation studies are being conducted to evaluate the kinesthetic realism of these models with medical experts.

Commentary by Dr. Valentin Fuster
2011;():807-814. doi:10.1115/DSCC2011-6170.

Ankle-foot orthoses (AFOs) are used to assist persons with lower-limb neuromuscular impairments. We have developed the portable powered AFO (PPAFO). This device uses a bidirectional pneumatic actuator powered by a CO2 bottle to provide dorsiflexor and plantarflexor torque assistance. The PPAFO operates tether-free, allowing for use outside of the laboratory. This system has been tested on one impaired and multiple healthy subjects. Timing of the assistance provided by the PPAFO has been determined by: 1) direct event detection using sensor feedback with threshold triggers, and 2) state estimation in which gait events are estimated using a cross-correlation based algorithm. Direct event detection, while simple to implement, can be unreliable for subjects with certain gait impairments. State estimation, while more complicated to implement, provides access to state information that cannot be directly measured by the AFO, which allows for greater flexibility in assistance timing. Current hardware limitations and future work are also discussed.

Topics: Orthotics
Commentary by Dr. Valentin Fuster

Vehicle Dynamics and Control

2011;():815-822. doi:10.1115/DSCC2011-5913.

This work presents the validation of control strategies for a semi-active steering damper aimed at improving the stability of two-wheeled vehicles by controlling the weave and wobble modes. A mixed frequency/time-domain method is introduced to evaluate the performance of the control strategies. The proposed cost functions allow one to evaluate the influence of the algorithms on the damping of the weave and wobble modes and also the overall performance in terms of stability of the steering assembly and of the chassis. The performance of the control algorithms is assessed on a multibody motorcycle simulator considering three challenging maneuvers that excite both the weave and wobble modes, such as kick-back and strong braking while cornering at high speed.

Topics: Dampers
Commentary by Dr. Valentin Fuster
2011;():823-830. doi:10.1115/DSCC2011-5934.

The design of an active stability control system for two-wheeled vehicles is a fully open problem and it constitutes a challenging task due to the complexity of two-wheeled vehicles dynamics and the strong interaction between the vehicle and the driver. This paper describes and compares two different methods, a model-based and a data-driven approach, to tune a Multi-Input-Multi-Output controller which allows to enhance the safety while guaranteeing a good driving feeling. The two strategies are tested on a multibody motorcycle simulator on challenging maneuvers such as kick-back and strong braking while cornering at high speed.

Commentary by Dr. Valentin Fuster
2011;():831-837. doi:10.1115/DSCC2011-5939.

In this paper a hybrid force-based Anti Locking Braking System (ABS) is presented. The proposed control system directly considers the lateral tyre behaviour during heavy braking. In this way it is possible to guarantee drivability and stability also when braking on surfaces without a clear tyre characteristic peak. The goal of the paper is to show the potentialities of load-based vehicle dynamics control. It is shown that, thanks to the availability of the lateral and longitudinal tyre force measurements, a stable braking can be achieved with a minimal increase of complexity of the algorithm in conditions that are critical for traditional ABS. The general control concept is that of reducing the longitudinal slip if the measured lateral force is smaller than a desired minimum. Two different ways of computing the minimum lateral forces are presented: one for the front axle with the objective of guaranteeing steerability and one for the rear axle to guarantee stability. Simulations on a nonlinear vehicle simulator confirm that the controller can maintain the desired steering behaviour and vehicle yaw stability in case of heavy braking.

Commentary by Dr. Valentin Fuster
2011;():839-846. doi:10.1115/DSCC2011-5955.

This paper investigates the appropriate level of model complexity when designing optimal vehicle active suspension controllers using the Linear Quadratic Regulator (LQR) method. The LQR method requires the formulation of a performance index with weighting factors to penalize the three competing objectives in suspension design: suspension travel (rattle space), sprung mass acceleration (ride quality) and tire deflection (road-holding). The optimal control gains are determined from the solution of a matrix Riccati equation with dimension equal to the number of state variables in the model. A quarter car model with four states thus poses a far less onerous formulation problem than a half or full car model with eight or more states. However, half and full car models are often assumed to be more accurate than quarter car models, and necessary for capturing and controlling degrees of freedom such as pitch and roll motion which are not directly available from a quarter car. The vertical acceleration, pitch acceleration and roadholding of a pitch plane vehicle are controlled in this paper using both quarter and half car-based controllers. First, optimal gains are calculated for each of the front and rear actuators assuming that the front and rear of the vehicle can be separately modeled as quarter cars with four states each. Then, half car-based optimal gains, based on feedback of eight states for the entire vehicle, are computed. Using quarter car-based controllers at the front and rear of a half car gives superior performance in reducing sprung mass inertial acceleration, and can effectively control pitch motion even when interactions between front and rear suspensions are not decoupled. Minimizing vertical motion of the front and rear ends indirectly regulates pitch motion. Improvements resulting from the additional complexity of the half car-based controller are seen only when the weighting factor for pitch suppression is very high in the performance index.

Commentary by Dr. Valentin Fuster
2011;():847-852. doi:10.1115/DSCC2011-6052.

This paper presents design of a neighboring optimal controller for vehicle evasive maneuvers. The developed controller is tested in presence of varying and constant side force disturbances. A model of wind side force is presented and the controller is tested. On simulated vehicle maneuver cases, the controller shows good disturbance rejection capabilities. The controller is able to perform the maneuver even in presence of 30 m/s (67 mph) wind gust with less than 1cm of error. Also, testing the controller with different bank angles shows that even with 10% bank angle the vehicle performs the maneuver with only 5 cm of error.

Commentary by Dr. Valentin Fuster
2011;():853-860. doi:10.1115/DSCC2011-6097.

During a race, professional drivers follow a racing line using specific maneuvers that allow them to utilize as much of the car’s tire force as possible. These lines could be used to create trajectories for obstacle avoidance in autonomous vehicles if they could be analytically defined. In fact, many of the techniques described by professional drivers can be expressed by a family of simple curves including straights, clothoids, and constant radius arcs. By comparing different members of this family of curves, different racing techniques can be examined. In particular, the differences between two phase and three phase corners described by professional drivers can be easily captured and analyzed in a single parameter. Experimental results on an autonomous race-car highlight the advantages of two phase cornering over three phase cornering and demonstrate the types of comparisons that can be made with this approach.

Commentary by Dr. Valentin Fuster
2011;():861-868. doi:10.1115/DSCC2011-6124.

Loss of control accidents result in thousands of fatalities in the United States each year. Production stability control systems are highly effective in preventing these accidents, despite their reliance on a hand-tuned response to data from a small set of sensors. However, improvements in sensing offer opportunities to determine stabilizing actions in a more systematic manner. This paper presents an approach that utilizes the yaw-sideslip phase plane to choose boundaries that eliminate unstable and undesirable driving regimes. These boundaries may be varied to obtain desirable performance and driver acceptance and form the basis for a driver assistance system that augments the driver input to maintain the vehicle within the bounds of a safe handling envelope. Experimental results from a model predictive controller used to enforce the envelope boundaries on a steer-by-wire vehicle are presented to demonstrate the viability of this framework for implementing stability boundaries.

Topics: Vehicles
Commentary by Dr. Valentin Fuster
2011;():869-876. doi:10.1115/DSCC2011-6152.

Agile maneuvers performed by professional racing car drivers have shown superior travel time and agility. These agile unstable yet safe maneuvers can be potentially used for active safety feature for high-speed obstacle avoidance or other emergency driving. Understanding the driving strategies during these agile maneuvers is however still not clear. In this paper, we first present coupled longitudinal/lateral vehicle dynamics and stability. Based on the stability analysis, we discuss agile maneuver driving strategies under which the vehicle is operated outside the stability region while still maintaining the safety. A new concept of safety boundary is used to quantitatively capture the safe region in the state space of vehicle dynamics. We use pendulum-turn agile maneuver experiments and CarSim simulations as an example to illustrate the analysis. We demonstrate that driving strategies are not unique to achieve high agile maneuvers such as pendulum-turn maneuvers.

Commentary by Dr. Valentin Fuster
2011;():877-884. doi:10.1115/DSCC2011-6159.

An integrated vehicle chassis control system was developed to improve vehicle handling (yaw) responses while maintain vehicle roll stability using an 8 DOF vehicle model, a simplified tire model, and a model predictive control method. The proposed control system incorporates active wheel torque distribution, active front steering, and active anti-rollbar to enhance vehicle handling and its ability to track the desired trajectory when the risk of vehicle rollover is low. As vehicle rollover risks increase, the proposed control system shifts its control focus from only handling enhancement to vehicle roll stabilization by adjusting the gains in the controller. The simulation results show that the proposed control system can improve vehicle handling responses while ensuring vehicle roll stability at high speed vehicle maneuvers.

Topics: Stability , Vehicles
Commentary by Dr. Valentin Fuster
2011;():885-892. doi:10.1115/DSCC2011-6168.

A Maglev-girder interaction system experiences a variation of its dynamics as the vehicle’s position varies over the girder’s span at different speeds. This paper investigates this variation using Multi-input Multi-output root locus criteria. Thus, the change of the system’s poles and zeros is visualized with the change of the vehicle’s position, while the speed of the vehicle represents the rate of these changes. These Multi-input Multi-output root locus criteria are used to study and design both centralized and de-centralized controllers for Maglev system.

Commentary by Dr. Valentin Fuster
2011;():893-900. doi:10.1115/DSCC2011-6195.

We present and compare vehicle maneuver stability under two vehicle dynamics models, one with the rear tire slip angle dynamics and the other with the vehicle side slip angle dynamics. Instead of using vehicle mass center side slip angle, we consider to use rear axial slip angle as one of the state variables for studying vehicle lateral dynamics. Using rear wheel slip angle as a state variable for studying vehicle dynamics has been reported in practices in industry but not rigorously studied. We analyze the new vehicle dynamics and compare the stability results with existing reported results. Both analytical and numerical results have shown that the stability region of the vehicle dynamics by using the rear slip angle is less conservative comparing with using the vehicle side slip angle.

Commentary by Dr. Valentin Fuster
2011;():901-908. doi:10.1115/DSCC2011-6200.

This paper presents a controller framework for autonomous drifting of a rear wheel drive vehicle. The controller uses a successive loop structure, where yaw rate is treated as a synthetic input to control the vehicle’s sideslip dynamics, and yaw rate is in turn controlled through coordination of steering and rear drive force. Relative to prior designs, the drift controller presented in this work enables a straightforward, physically insightful stability analysis where local closed-loop stability of the desired high sideslip “drift equilibrium” is demonstrated. When implemented on a by-wire testbed, the new controller achieves experimental performance that matches or exceeds prior designs, generating sustained and robust autonomous drifts.

Commentary by Dr. Valentin Fuster

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In