0

ASME Conference Presenter Attendance Policy and Archival Proceedings

2012;():i. doi:10.1115/DSCC2012-MOVIC2012-NS1.
FREE TO VIEW

This online compilation of papers from the ASME 2012 5th Annual Dynamic Systems and Control Conference joint with the JSME 2012 11th Motion and Vibration Conference (DSCC2012-MOVIC2012) 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to ASME’s conference presenter attendance policy, if a paper is not presented at the Conference, the paper will not be published in the official archival Proceedings, which are registered with the Library of Congress and are submitted for abstracting and indexing. The paper also will not be published in The ASME Digital Collection and may not be cited as a published paper.

Commentary by Dr. Valentin Fuster

Adaptive Control

2012;():1-9. doi:10.1115/DSCC2012-MOVIC2012-8529.

Solutions already exist for the problem of canceling sinusoidal disturbances by measurement of the state or by measurement of an output for linear and nonlinear systems. In this paper, we design an adaptive backstepping controller to cancel unmatched sinusoidal disturbances forcing a linear time-invariant system which is augmented by a linear input subsystem by using only measurement of state-derivatives of the original subsystem and state of the input subsystem. Our design is based on four steps, 1) parametrization of the sinusoidal disturbance as the output of a known feedback system with an unknown output vector that depends on unknown disturbance parameters, 2) design of an adaptive disturbance observer for both disturbance and its derivative, 3) design of an adaptive controller for virtual control input, and 4) design final controller by defining error system and using backstepping procedure. We prove that the equilibrium of the closed-loop adaptive system is stable and state of the considered error system goes to zero as t → ∞ with perfect disturbance estimation. The effectiveness of the controller is illustrated with a simulation example of a third order system.

Topics: Feedback
Commentary by Dr. Valentin Fuster
2012;():11-18. doi:10.1115/DSCC2012-MOVIC2012-8547.

In system identification and adaptive control, the problem of designing strictly positive real (SPR) transfer functions in the presence of uncertain adaptation parameters is essential for stability and convergence in a group of parameter adaptation algorithms. This paper proposes a convex optimization approach to address the robust SPR problem. Besides achieving the robust SPR condition, the presented solution is optimal in the sense of minimizing the distance from the transfer function to unity. Such consideration is important for parameter convergence in practical applications. New topics such as minimum-order compensator and minimum high-frequency magnitude design are also introduced.

Topics: Optimization
Commentary by Dr. Valentin Fuster
2012;():19-27. doi:10.1115/DSCC2012-MOVIC2012-8551.

We apply an extension of retrospective cost adaptive control (RCAC) to a command-following problem for the uncertain electromagnetically controlled oscillator (ECO). We assume that an estimate of the first Markov parameter of the discretized and linearized plant is known, but RCAC does not require knowledge of the inertia, damping, or stiffness of the plant. RCAC uses a setpoint feedback path and an auxiliary nonlinearity to stabilize the unstable ECO at the commanded equilibria.

Commentary by Dr. Valentin Fuster
2012;():29-38. doi:10.1115/DSCC2012-MOVIC2012-8573.

We apply retrospective cost adaptive control (RCAC) with auxiliary nonlinearities to a command-following problem for uncertain Hammerstein systems with rate-dependent hysteretic input nonlinearities. The only required modeling information of the linear plant is a single Markov parameter. To account for the hysteretic input nonlinearity, RCAC uses auxiliary nonlinearities that reflect the monotonicity properties of the input nonlinearity. The hysteresis nonlinearity is modeled using the rate-dependent Prandtl-Ishlinskii model.

Commentary by Dr. Valentin Fuster
2012;():39-48. doi:10.1115/DSCC2012-MOVIC2012-8580.

We apply retrospective cost adaptive control (RCAC) to a broadband disturbance rejection problem under limited modeling information and assuming that the performance variable is measured. The goal is to compare the asymptotic performance (that is, after convergence of the controller) of the adaptive controller with the performance of discrete-time LQG controller, which uses complete modeling information but does not require a measurement of the performance variable. For RCAC we assume that the first nonzero Markov parameter of the plant is known. We show that if the plant zeros are also known, the retrospective cost can be modified to recover the high-control-authority LQG performance.

Commentary by Dr. Valentin Fuster
2012;():49-57. doi:10.1115/DSCC2012-MOVIC2012-8597.

We apply retrospective cost adaptive control (RCAC) to command-following and disturbance-rejection problems for a diesel engine model. The engine is a multi-input, multi-output system with strong static and dynamic interactions, nonlinearities, uncertainties and nonminimum phase characteristics. We demonstrate that RCAC is effective for both the linearized and nonlinear engine models provided that two Markov parameters of the linearized engine plant model are known, either analytically or through system identification. For the command-following and disturbance-rejection problems, we consider the case when the disturbance is harmonic but otherwise unknown, and while the command signal is harmonic and known but no advance knowledge of its spectrum is assumed to be available.

Commentary by Dr. Valentin Fuster
2012;():59-64. doi:10.1115/DSCC2012-MOVIC2012-8635.

This paper deals with a dynamic analysis of a racing kart considering an elastic deformation of the kart frame. As the first step of this research, an FEM kart frame model was validated by means of static and dynamic tests. In the static test, the strain on the kart frame caused by a steering operation was measured by strain gauges, and compared to the simulation result. The dynamic response of the frame was evaluated by hammering test, and the test result was compared to that of the modal analysis. Next, a flexible multibody vehicle model was developed with this FEM frame model. In this process, model reduction based on the modal analysis was applied to reduce the degree of freedom of the flexible body. Some running tests with actual racing kart were carried out to evaluate the handling characteristic, and the comparison between simulation and experiment are discussed.

Commentary by Dr. Valentin Fuster
2012;():65-73. doi:10.1115/DSCC2012-MOVIC2012-8679.

The main contribution of this paper is to realize a stability-based model-free control approach based on a conservative online-judgment of measured states with several assumptions. This new type of adaptive control with embedded stability-based active control adaption is based on a cognition-based framework, which consists of three parts: (1) a dynamic recurrent neural network (DRNN) used for local identification and multi-step-ahead prediction of the system; (2) a geometrical criterion based on a suitable definition of quadratic stability applied for judging the stability of motion of the system numerically; (3) a suitable strategy according to a cost function used for choosing most suitable control input value for the next predefined time interval. The proposed controller is able to gain useful local knowledge and define autonomously suitable local control input according to the stability criterion. Numerical examples are shown to demonstrate the successful application and performance of the method.

Commentary by Dr. Valentin Fuster
2012;():75-84. doi:10.1115/DSCC2012-MOVIC2012-8752.

Decentralized control is a longstanding challenge in systems theory. A decentralized controller may consist of multiple local controllers, connected to disjoint or overlapping sets of sensors and actuators, and where each local controller has limited ability to communicate directly with the remaining local controllers and, in addition, may lack global knowledge of the plant and operation of the remaining local controllers. In the present paper we apply adaptive control to investigate the ability of the local controllers to cooperate globally despite uncertainty, communication constraints, and possibly conflicting performance objectives. The approach we apply in this paper is based on retrospective cost adaptive control (RCAC). The development of RCAC assumes a centralized controller structure; the goal of the present paper is to investigate the stability and performance of RCAC in a decentralized setting.

Commentary by Dr. Valentin Fuster
2012;():85-92. doi:10.1115/DSCC2012-MOVIC2012-8767.

A performance oriented multi-loop approach to the tracking control of linear motor drive systems with input saturation, state constraints, parametric uncertainties and input disturbances is proposed. In the inner loop, a constrained adaptive robust control (ARC) law is synthesized to achieve the required robust tracking performances with respect to on-line replanned trajectory in the presence of input saturation and various types of uncertainties. In the middle loop, a set-membership identification (SMI) algorithm is implemented to obtain a monotonically decreasing estimate of the upper bound of the inertia so that more aggressive trajectory replanning can be done. In the outer loop, a replanned trajectory is generated to minimize the converging time of the overall system response to the desired target while not violating various constraints. It is theoretically shown that the resulting closed-loop system can track any feasible desired trajectory with a guaranteed converging time and steady-state tracking accuracy without violating the state constraints. Experimental results obtained on a HIWIN linear motor show that the proposed algorithm indeed achieves closed-loop stability and small steady-state tracking errors with a transient performance much better than that of the unconstrained ARC algorithm.

Commentary by Dr. Valentin Fuster
2012;():93-100. doi:10.1115/DSCC2012-MOVIC2012-8783.

Due to the potential of achieving high speed and high precision, linear motor driven gantry systems are widely used in industrial applications such as machine tools, semiconductor manufacturing equipments and microelectronics manufacturing equipments. To have large enough driving force, a H-type structure with dual parallel driving motors is usually adopted. Though dual-motor driven structure can deliver higher driving power, the “pull and drag” problem between two motors exists when they are controlled separately. In this paper, a synchronous control scheme is proposed which has an effective thrust allocation to deal with the chattering of the “pull and drag” effect, and uses adaptive robust control (ARC) technique to obtain a guaranteed performance with the presence of both parametric uncertainties and uncertain nonlinearities. Comparative experiments have been done on a dual-linear-motor-driven industrial gantry. The results show that the control input chattering is significantly reduced without sacrificing the good tracking performance of ARC controllers.

Commentary by Dr. Valentin Fuster
2012;():101-110. doi:10.1115/DSCC2012-MOVIC2012-8800.

This paper develops an adaptive PDE observer for battery state-of-charge (SOC) and state-of-health (SOH) estimation. Real-time state and parameter information enables operation near physical limits without compromising durability, thereby unlocking the full potential of battery energy storage. SOC/SOH estimation is technically challenging because battery dynamics are governed by electrochemical principles, mathematically modeled by partial differential equations (PDEs). Simultaneous state and parameter estimation is extremely challenging in PDE models. Consequently, several new theoretical ideas are developed, integrated together, and tested. These include a backstepping PDE state estimator, a Padé-based parameter identifier, nonlinear parameter sensitivity analysis, and adaptive inversion of nonlinear output functions. The end result is the first combined SOC/SOH battery estimation algorithm that identifies physical system variables via an electrochemical model, from measurements of voltage and current only.

Topics: Batteries
Commentary by Dr. Valentin Fuster

Advanced Vehicle Propulsion Systems

2012;():111-118. doi:10.1115/DSCC2012-MOVIC2012-8531.

This paper presents a fault diagnosis approach for a class of over-actuated systems where the actuators share the same unknown parameters. With the persistence of excitation condition being satisfied, the estimated unknown parameters from all the healthy and identical actuators should have a great similarity. A significant parameter estimation difference from other estimations may be attributed to the fault on the specific actuator. Such a feature was utilized to design the residuals and evaluate the actuator fault for over-actuated systems with identical actuators and applied to a four wheel independently actuated (FWIA) electric ground vehicle where the tire-road friction coefficient (TRFC) can be treated as the common unknown parameter for the four wheels. Simulation and experimental results from a FWIA electric ground vehicle are given to show the effectiveness of the proposed fault diagnosis method.

Commentary by Dr. Valentin Fuster
2012;():119-128. doi:10.1115/DSCC2012-MOVIC2012-8555.

The EcoCAR 2 team at the Ohio State University is designing a Parallel-Series Plug-in Hybrid Electric Vehicle (PHEV) that will be capable of 50 miles all-electric range using a 18.9 kWh battery pack, with range extension in parallel operation using a 1.8L dedicated E85 engine. The vehicle is designed to significantly reduce fuel consumption while still meeting the EPA’s Tier II Bin 5 emissions standards. This paper details the rapid vehicle architecture selection process the team completed to select the parallel-series PHEV architecture in less than two months. The team used Argonne National Lab’s Autonomie software to simulate four hybrid vehicle architectures considered by the team as well as the conventional vehicle. The simulation results were used to compare the performance and fuel economy of each of the hybrid vehicles, in order select the vehicle architecture that best met the goals of the Ohio State team.

Topics: Vehicles
Commentary by Dr. Valentin Fuster
2012;():129-136. doi:10.1115/DSCC2012-MOVIC2012-8645.

This paper presents a study of how different methods for designing energy management strategies (EMS) are affected by variations in drive cycle and system parameters. Specifically a rule-based strategy extracted from analysis of dynamic programming results over the Federal Urban Drive Cycle and a model predicative controller are considered. These strategies were validated on a hydraulic hybrid powertrain testbed. The hydraulic hybrid powertrain 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. Through a simulation study we will demonstrate the need to carefully consider the application and confidence one has in knowledge of the drive cycle and system model when choosing an EMS design method.

Commentary by Dr. Valentin Fuster
2012;():137-144. doi:10.1115/DSCC2012-MOVIC2012-8692.

There is currently a high level of uncertainty surrounding the evolution of personal transportation. A variety of new types of vehicle powertrains have been proposed or implemented, including alternative fuels, hybrid electric vehicles, and fully electric vehicles. It is also possible, as shown by Mechtenberg [1], to combine multiple fuels and batteries to design this 36 mode hybrid vehicle. The hybrid vehicle presented here features multiple modes of operation with a wide range of possible combinations of fuel and battery usage. While the many degrees of freedom offered by this hybrid vehicle design present an opportunity to operate under a variety of different conditions, it also presents a control challenge, as the vehicle’s control system must decide how best to use the various modes available, given the driver’s optional selection and the current status of the vehicle. In this paper, we discuss the various modes of operation, degree of driver involvement in their selection, and automatic switching between various options. The optimal control is found for various different driving cycles, based on the objective of maximizing the efficiency of the powertrain, and it is shown that this type of hybrid vehicle can operate efficiently under a variety of different scenarios. This model is built upon Wagner and Papalambros’ engine optimization [2] and Ahn’s continuously variable transmission model [3].

Commentary by Dr. Valentin Fuster
2012;():145-150. doi:10.1115/DSCC2012-MOVIC2012-8699.

This paper presents a detailed analysis of the optimal energy management problem for a plug-in hybrid electric vehicle (PHEV) solved using the Pontryagin’s minimum principle (PMP). The aim of this analysis is to study the relationships between the control parameters and the vehicle and driving characteristics. In this study, a relationship between the state and co-state trajectories with the battery characteristics has been developed which has not been explored in a similar fashion in prior literature. Results from sensitivity analysis show a strong asymmetry in higher and lower estimation of the initial value of the co-state. A spatial domain analysis is also carried out which shows quasi linearity of the optimal state of charge (SOC) with respect to trip length for a combination of driving cycles. Knowledge gained from this exercise enables us to develop an adaptive energy management strategy.

Commentary by Dr. Valentin Fuster
2012;():151-156. doi:10.1115/DSCC2012-MOVIC2012-8763.

A wedge clutch with a wedge ramp transfers the tangential force into an axial force. It has unique features of self-reinforcement, and can be packaged into tight spaces. This wedge clutch is developed to apply to an automatic transmission as an implementation example.

The slipping decay time is found to be critical for the shifting quality. This paper focuses on the experimental study and control of slipping decay time of the wedge clutch through the influencing factors. The mechanical system of the wedge clutch applied in an automatic transmission is described and the sensors for measuring signals are installed. A transmission dynamometer is set up for experiments.

The torque magnitude and direction of the motor motion are considered as actuation factors; the driveline input speed, load torque, and oil temperatures are considered as the driveline factors. The results show how influencing factors affect the slipping decay time during gear shifting.

Topics: Wedges
Commentary by Dr. Valentin Fuster

Aerospace Systems

2012;():157-165. doi:10.1115/DSCC2012-MOVIC2012-8512.

An innovative implementation of attitude estimation in 3 degrees of freedom (3-DOF) combining the TRIAD algorithm [1] and a time-varying nonlinear complementary filter (TVCF) is derived. This work is inspired by the good performance of the TVCF in 1-DOF [2] developed for applications limited to small mobile platforms with low computational power. To demonstrate robust 3-DOF estimation, information from vector and rate-gyroscope measurements are fused. Simulation and experimental results demonstrate comparable performance to the extended Kalman filter (EKF) and improved performance over alternative methods such as sole gyroscope rate-integration and the TRIAD algorithm without the TVCF as a pre-filter.

Topics: Algorithms , Filters
Commentary by Dr. Valentin Fuster
2012;():167-175. doi:10.1115/DSCC2012-MOVIC2012-8736.

In this paper, we present the use of optimal estimation and control algorithms to guide and control an autonomous glider to reach a desired target state zone. The optimal estimation technique is based on an Extended Kalman Filter that uses quaternion orientation representation and online calibration techniques in order to enhance the state estimation algorithm. The control scheme is based on a LQR tracker control law that uses a linearized time invariant representation of the plant. The optimal control law guides the glider nonlinear dynamics based on a set of continuous parameterized parabolic reference curves that allows the glider to reach the final state zone in a natural way using low energy consumption.

Commentary by Dr. Valentin Fuster
2012;():177-184. doi:10.1115/DSCC2012-MOVIC2012-8781.

We consider the following two problems to enable a quadrotor to operate a mechanical tool (e.g., screwdriver, vertical jack, etc), which is rigidly-attached on the quadrotor and whose control, different to other quadrotor motion control results, demands an integrated and simultaneous control of the quadrotor’s translation and rotation: 1) tool-tip position trajectory tracking control; and 2) tool rotating operation control with the tool-tip position regulated. We characterize some structural conditions for generating any arbitrary desired control for the tool-tip position and also for avoiding internal dynamics instability (with possible finite-time escape). Simulations are performed to support the theory.

Commentary by Dr. Valentin Fuster
2012;():185-194. doi:10.1115/DSCC2012-MOVIC2012-8817.

Motivated by the fact that regulating yaw at zero avoids complex second order aerodynamical terms coming from cross build up of thrust vector, which looses lift and increases drag, the position control problem of quadrotors has been studied generally regulating desired yaw, which limits the scope of underactuated quadrotors because yaw is a degree of freedom. In this paper, to deal with time-varying heading, we propose a second order quaternion-based sliding mode control for the tracking of the full nonlinear position-orientation dynamics, including finite time tracking. Chattering, typical from sliding modes, does not appear and knowledge of the model and its parameters is not required to implement the controller, a novelty also in sliding modes. Some applications are explored in a simulation study to show the viability of the proposed scheme, including field-of-view targeting, aerial screw driver and aerial grasp tasks.

Topics: Yaw
Commentary by Dr. Valentin Fuster
2012;():195-203. doi:10.1115/DSCC2012-MOVIC2012-8830.

This paper presents the initial study on the dynamic shape control problem of deployable mesh reflectors via feedback. To compensate the thermal distortion and reject force disturbance, the system is considered quasi-static to the temperature variation. After segmenting the orbital temperature range into narrow sections, the robust state feedback controller is proposed in each section and the controllers are properly switched between sections during the entire cycle using a supervisory controller. The control method is then implemented on a sampled deployable mesh reflector and the time-response shows much higher surface accuracy the reflector maintains during the orbiting mission comparing to the open-loop configuration.

Commentary by Dr. Valentin Fuster
2012;():205-214. doi:10.1115/DSCC2012-MOVIC2012-8869.

This paper focuses on designing control systems for the regulation of unmanned aerial vehicles (UAVs) along real-time trajectories. The reference trajectories will be generated in real-time from a library of pre-specified motion primitives. Two control approaches will be studied. The first is a hybrid control approach, in which we account for all possible connections between compatible library primitives by adding corresponding coupling conditions into the control synthesis program. The switching between primitives in this case takes place with stability and performance guarantees, but at the expense of added computational complexity. The second approach is a decoupled control approach, where the plant associated with each primitive is regarded as a system with an uncertain initial state. This approach is less computationally intensive than the hybrid approach, but comes with no theoretically established stability guarantees across switching boundaries. The two approaches are applied to regulate a nonlinear mathematical model of a 6 foot Telemaster fixed-wing UAV along a real-time trajectory in the presence of model uncertainties and atmospheric disturbances.

Commentary by Dr. Valentin Fuster

Autonomous Systems

2012;():215-224. doi:10.1115/DSCC2012-MOVIC2012-8544.

The velocity occupancy space (VOS) algorithm was developed to allow an unmanned ground vehicle (UGV) to avoid moving and stationary obstacles and navigate efficiently to a goal using only uncertain sensor data. The original VOS concept was designed for an ideal, holonomic UGV that was capable of perfect, repeatable and instantaneous velocity changes. The method presented here adapts VOS through the use of extended velocity obstacles (EVOs) so that VOS can operate on experimental UGVs that suffer from actuation error. For this research, the EVOs have been designed based on the performance of a SuperDroid ATR, but they can be easily calibrated for other velocity controlled UGVs. The proposed method is validated through numerous simulations.

Topics: Vehicles , Errors
Commentary by Dr. Valentin Fuster
2012;():225-233. doi:10.1115/DSCC2012-MOVIC2012-8605.

This paper is concerned with robust impedance shaping control for electric power steering (EPS) systems. The analysis here presents robust stability conditions for both sector-bounded and passive uncertainties. A 2-DOF controller formed of static output feedback (SOF) gains and a sector-bounded nonlinearity is proposed for the EPS closed loop system. The controller employs torque-feedback inner loop and an outer impedance shaping controller. The controller gains are synthesized via a multi-objective optimization involving nonlinear matrix inequality constraints. An iterative algorithm combining both semi-definite programming (SDP) and genetic algorithms (GA) is implemented to find the (sub)optimal controller gains.

Commentary by Dr. Valentin Fuster
2012;():235-241. doi:10.1115/DSCC2012-MOVIC2012-8620.

For racecar drivers, careful consideration is used when defining a racing line for a given track. Each driver ultimately seeks to minimize the track time by compromising between the shortest distance around a track and a line that results in the fastest speed along the path. Professional racing strategies could be used to create trajectories for obstacle avoidance in autonomous vehicles if there was a way to analytically represent their techniques.

Commentary by Dr. Valentin Fuster
2012;():243-251. doi:10.1115/DSCC2012-MOVIC2012-8640.

This paper describes the design, construction, and simulation of a prototype, teleoperated, omnidirectional robotic ground vehicle. The design of a dynamic control system to assist the human operator of the vehicle is also presented. This work sought to test the feasibility of a novel vehicle architecture and to develop a dynamic multi-body simulation tool to assist in the development of future iterations of such a vehicle. The vehicle design seeks to achieve high-speed, omnidirectional mobility, and modest off-road capability. This paper presents results from the physical operation and simulation of the vehicle as well as describing some future work to achieve improved performance of the vehicle system.

Topics: Simulation , Design , Vehicles
Commentary by Dr. Valentin Fuster
2012;():253-262. doi:10.1115/DSCC2012-MOVIC2012-8694.

Velocity Occupancy Space (VOS) is an algorithm used to perform real-time moving obstacle avoidance for Unmanned Ground Vehicles (UGVs). Pedestrian safety and comfort must be ensured by the guidance algorithm when UGVs operate near pedestrians. Studies on human interactions are used to formulate desired behavior of the UGV as it avoids people. Then, VOS is enhanced with improved avoidance behavior. We investigate three potential methods for enhancing pedestrian safety and comfort. To study the effect of these enhancements, numerical simulations of a large set of randomly generated scenarios. The number of collisions, and minimum required opening width, were used to characterize the performance of each avoidance enhancement technique. Two of the methods eliminate collisions with pedestrians; the preferred technique, which artificially enlarges the size of detected pedestrians, accomplishes this consistently.

Topics: Vehicles
Commentary by Dr. Valentin Fuster
2012;():263-271. doi:10.1115/DSCC2012-MOVIC2012-8843.

Real-time obstacle avoidance and navigation are key fields of research in the area of autonomous vehicles. The primary requirements of autonomy are to detect or sense changes and react to them without human intervention in a safe and efficient manner. The objective of this research is to develop autonomous way-point navigation and obstacle avoidance capabilities for an unmanned ground vehicle (UGV). This research consists of developing and implementing an environment mapping system capable of detecting and localizing potential obstacles using real-time sensor data. The real-time obstacle mapping system developed in this work automatically generates the Probabilistic Threat Exposure Map (PTEM). The PTEM construction algorithm successfully constructs a probabilistic obstacle map both in simulation and real-time. Autonomous waypoint navigation is also achieved for both simulation and real-time platforms. These activities are a part of a larger effort to establish a theoretical foundation and real-time implementation of autonomous and cooperative multi-UxV guidance solutions in adversarial environments.

Topics: Vehicles , Navigation
Commentary by Dr. Valentin Fuster

Battery Modeling

2012;():273-278. doi:10.1115/DSCC2012-MOVIC2012-8525.

Control oriented battery pack models can be very useful in the context of electrified vehicle development because such models can be used to develop the battery management system, which is a broad term that encompasses everything from thermal management and charge balancing to state of charge estimation. In this paper, a control oriented battery pack model is developed using the Modelica modeling environment. This model combines a Randle equivalent circuit representation with an lumped parameter thermal model for each cell to provide simultaneously the electrical and thermal variations among the cells. Modelica differs from modeling environment such as Simulink because it offers non-causal modeling, which handles both differential and algebraic equations. The resulting implementation is both scalable and flexible to accomodate any pack size, geometry, and connection. As part of the future work, supporting architectures such as the charge balancing system and thermal management system can be added to the pack model to simulate the complete pack operation.

Topics: Batteries
Commentary by Dr. Valentin Fuster
2012;():279-286. doi:10.1115/DSCC2012-MOVIC2012-8540.

State-Of-Charge (SOC) estimation for Valve-Regulated Lead-Acid (VRLA) battery is complicated by the switched linear nature of the underlying dynamics. A first principles nonlinear model is simplified to provide two switched linear models and linearized to produce charge, discharge, and averaged models. Luenberger and switched SOC estimators are developed based on these models and propogated using experimental data. The results show that estimation errors are halved by including switching in the observer design.

Commentary by Dr. Valentin Fuster
2012;():287-291. doi:10.1115/DSCC2012-MOVIC2012-8584.

Valve Regulated Lead-Acid (VRLA) batteries can degrade due to a variety of mechanisms, including corrosion, hard sulfation, water loss, shedding, and active mass degradation. Hard sulfation can be the dominant aging mechanism for many cells. In this paper, pressure feedback is used to minimize water loss during low current charging designed to break up hard sulfate and recover capacity. A VRLA battery that was cycle tested to failure is used to test the desulfation charging control scheme. One cell of this battery that was diagnosed with sulfation degradation was desulfated for 313 hrs at an average current of 0.2 A. The capacity of the cell was recovered by 41% with minimal water loss, demonstrating the effectiveness of the desulfation charge controller.

Commentary by Dr. Valentin Fuster
2012;():293-297. doi:10.1115/DSCC2012-MOVIC2012-8616.

The vanadium redox flow battery (VRFB) is an attractive grid scale energy storage option, but high operating cost impedes widespread commercialization. One way of mitigating cost is to optimize system performance, which requires an accurate model capable of predicting cell voltage under different operating conditions such as current, temperature, flow rate, and state of charge. This paper presents an isothermal VRFB model based on principles of mass transfer and electrochemical kinetics that can predict transient performance with respect to the aforementioned operating conditions. The model captures two important physical phenomena that occur at significantly different time scales as a result of vanadium crossover: (1) rapid self discharge reactions and (2) capacity loss after long term cycling. A gap metric analysis showed that a linear controller for the flow rate may be suitable near the 50% SOC range.

Commentary by Dr. Valentin Fuster
2012;():299-307. doi:10.1115/DSCC2012-MOVIC2012-8649.

Health management of Li-ion batteries depends on knowledge of certain battery internal dynamics (e.g., lithium consumption and film growth at the solid-electrolyte interface) whose inputs and outputs are not directly measurable with noninvasive methods. This presents a problem of identification of inaccessible subsystems. To address this problem, we apply the retrospective-cost subsystem identification (RCSI) method. As a first step, this paper presents a simulation-based study that assumes as the truth model of the battery an electrochemistry-based battery charge/discharge model of Doyle, Fuller, and Newman, and later augmented with a battery-health model by Ramadass. First, this truth model is used to generate the data needed for the identification study. Next, the film-growth component of the battery-health model is assumed to be unknown, and the identification of this inaccessible subsystem is performed using RCSI. The results show that the subsystem identification method can identify the film growth quite accurately when the chemical reactions leading to film growth are consequential.

Topics: Batteries
Commentary by Dr. Valentin Fuster
2012;():309-318. doi:10.1115/DSCC2012-MOVIC2012-8751.

Accurate battery health modeling allows one to make better design decisions, enables health conscious control, and allows for feed-forward State of Health estimation. However, experiments are necessary in order to obtain and validate these models. Unfortunately, battery health experiments are costly in terms of time, person-hours, and equipment. This makes it extremely important to minimize the number of experimental iterations.

This paper aims to minimize time and expense of experiments while maximizing information gathered by bridging an important gap between the Optimal Experimental Design (OED) and the battery health experimental/modeling literature. We demonstrate how to apply static OED methods to a battery aging experiment. This allows us to select a set of Constant Current Constant Voltage (CCCV) cycles that maximizes the amount of information gathered — in turn allowing us to better identify the health model parameters. The CCCV cycling is carried out in a laboratory using 14 LiFePO4 cells (10 for fitting and 4 for validation). Each of these cells undergoes 429 days of battery health cycling. Results from these experiments include: a model of battery capacity fade based on voltage and current, battery health dependence on voltage, and a lack of power fade under the cycling conditions. The use of OED to coordinate our model form and experiment helped to ensure a fruitful model resulted when processing the collected data. Based on this success we suggest a generalized framework for Optimal Battery Health Model Experiments (OBHME), which allows one to apply OED to a variety of related problems.

Commentary by Dr. Valentin Fuster

Biochemical Systems

2012;():319-326. doi:10.1115/DSCC2012-MOVIC2012-8550.

The most widespread approach for glycemic control in diabetic patients is the so-called basal-bolus insulin regimen, comprising insulin injections at meal times, correction doses in hyperglycemia and compensatory carbohydrate in case of insulin-induced hypoglycemia. The present contribution represents an attempt at implementing such a strategy on a population of 4 virtual, i.e., in-silico, T1DM patients. Low-order physiologically sound transfer function models were estimated for each of the in-silico subjects from simulated data and exploited in an optimization-based control algorithm, the objective being sustainment of glycemia in the near-normal range (70–180 [mg/dL]).

Commentary by Dr. Valentin Fuster
2012;():327-334. doi:10.1115/DSCC2012-MOVIC2012-8630.

This paper develops a control integration methodology that can be used to balance competing performance objectives. The target application for the method is blood glucose management for diabetic patients. The true system model is uncertain and changes over time. Multiple controllers have been designed with competing performance objectives in mind. In the proposed method, on-line predictors are used to estimate the future state of the system based on the current state and the different control recommendations. The system then switches between the different controllers based on the predicted outcomes from their control recommendations. Simulation results demonstrate that the integrated approach based on predicted outcomes performs better across a large range of disturbances than any of the individual controllers.

Commentary by Dr. Valentin Fuster
2012;():335-344. doi:10.1115/DSCC2012-MOVIC2012-8729.

Mathematical models in systems biology are often constructed by either Ordinary Differential Equation (ODE) modeling or logical (Boolean) modeling. We develop a Hybrid Boolean Model (ODE+Boolean) for biological signal pathways with postulated epigenomic feedback. The basic idea in this model is to combine continuous dynamical systems (an ODE model for already well-known parts of the network) with a discrete transition system (Boolean, for postulated but largely unknown components). We use the existing or well-known ODE model to “trigger” signal pathways represented by a Boolean model. This framework is easier to validate than a complete ODE model for large and complex signal pathways, for example to find unknown pathways to match the response to experimental data. The advantage of using a Boolean model for the unknown parts of the network is that relatively few parameters are needed. Thus, the framework avoids over-fitting, covers a broad range of pathways and easily represents various experimental conditions. The overall goal of the hybrid model is to predict the behavior of biological signal pathways, thus helping to understand unknown parts of the pathway between experimental results and qualitative/quantitative results. Extensions are discussed, and numerical examples in biological systems and one engineering example are provided.

Topics: Modeling , Signals
Commentary by Dr. Valentin Fuster
2012;():345-354. doi:10.1115/DSCC2012-MOVIC2012-8803.

In this paper, we propose and design an adaptive dual controller for automatic glucose control of diabetic patients. The results could be used in the development of an artificial pancreas, which, while as yet unavailable, must consist of three major components: an insulin delivery device or “pump”, a continuous glucose sensor, and a control algorithm linking insulin delivery to measured glucose concentration. For improved performance the system would also include “feed-forward” information about food intake, physical activity and other blood glucose perturbing inputs. A linear time-varying autoregressive model with exogenous inputs is constructed to characterize the kinetics of both glucose-insulin and glucose-carbohydrate interaction. Combined with a Kalman-filter based estimation scheme for online estimation of the time-varying model coefficients, we design an adaptive dual control that both excites the glucose dynamic system sufficiently to accelerate the parameter estimation and cautiously tracks the desired glucose level. Performance evaluation of the adaptive dual controller is accomplished via simulations on virtual patients constructed using clinical data from five different patients with type-1 insulin-deficient diabetes using continuous subcutaneous insulin infusion for diabetes management during observation. Simulation results show both smaller glucose excursions and a reduction in the number of hypoglycemic events for all but one of the five subjects.

Commentary by Dr. Valentin Fuster

Control Over Networks

2012;():355-362. doi:10.1115/DSCC2012-MOVIC2012-8553.

This work explores the stability properties of a class of nonlinear system, which is obtained by applying a special transformation to the vector field of a linear system. It is proved that such a nonlinear system preserves the stability of the original linear system. This property can be used for checking the stability of certain nonlinear systems, and designing novel network control protocols based on homogeneous system theory. For the network control application, finite-time convergence can be achieved in a decentralized manner by properly choosing the smooth feedback control law. The results are demonstrated by numerical examples.

Topics: Feedback
Commentary by Dr. Valentin Fuster
2012;():363-372. doi:10.1115/DSCC2012-MOVIC2012-8622.

In this paper, a modified preview control technique is proposed to compensate packet loss in a wireless tracking control system, where future reference signals over a finite horizon can be previewed. In order to utilize the future reference information for controller design, the system model is augmented with a reference generator whose states are the future reference signals. As a response to the packet loss that occurs in the wireless network, the preview control technique is modified by employing Bernoulli variables to represent packet losses in both controller-actuator and sensor-controller channels. The Bernoulli packet loss model, along with tracking errors and control inputs, is included in a quadratic cost function, and the optimal controller gain that minimizes the cost function is obtained by dynamic programming. A modified Kalman filter considering packet loss is utilized for full state estimation and state feedback control. Stability of the modified preview control system is studied using linear matrix inequalities (LMIs). Choice of preview horizon is discussed and performance of the proposed controller is verified by simulation and experimental results.

Commentary by Dr. Valentin Fuster
2012;():373-382. doi:10.1115/DSCC2012-MOVIC2012-8677.

One of the main challenges of co-simulating hardware-in-the-loop systems in real-time over the Internet is the fidelity of the simulation. The dynamics of the Internet may significantly distort the dynamics of the network-integrated system. This paper presents the development of an iterative learning control based approach to improve fidelity of such networked system integration. Towards this end, a new metric for characterizing fidelity is proposed first, which, unlike some existing metrics, does not require knowledge about the reference dynamics (i.e., dynamics that would be observed, if the system was physically connected). Next, using this metric, the problem of improving fidelity is formulated as an iterative learning control problem. Finally, the proposed approach is illustrated on a purely simulation-based case study. The conclusion is that the proposed approach holds significant potential for achieving high fidelity levels.

Commentary by Dr. Valentin Fuster
2012;():383-391. doi:10.1115/DSCC2012-MOVIC2012-8814.

A procedure to modularly and recursively assemble port-based hybrid models is introduced. This work extends the conventional Modular Assembly Method for linear and affine models. Hybrid models offer a unifying framework to characterize and predict processes or phenomena that exhibit both continuous and discrete dynamic behavior. The main features of the proposed method are portability and protection of content. The connection of a three-gear transmission model on an electric motor model illustrates the assembly, condensation and simulation of a hybrid model without revealing the connection topology of the modeling components.

Topics: Manufacturing
Commentary by Dr. Valentin Fuster
2012;():393-402. doi:10.1115/DSCC2012-MOVIC2012-8837.

In this article, we address the problem of synthesizing UAV communication networks in the presence of resource constraints. UAVs can be deployed as backbone nodes in ad-hoc networks that can be central to civilian and military applications. The cost of operation of the network depends on the resources that are used such as the total power consumption associated with the network and the number of communication links in the network. The objective of the problem is to synthesize a communication network that maximizes connectivity subject to cost of operation being within the specified resource. We choose algebraic connectivity as a measure of connectivity of the network as it is known to be a measure of robust connectivity to random node failures in the network. We pose the network synthesis problem as a mixed-integer semi-definite program and provide (1) an algorithm for computing optimal solutions using cutting plane methods, and (2) construct feasible solutions using heuristics and estimate their quality. The network synthesis problem is a NP-hard problem and there are no guarantees on the running time of the algorithm that computes an optimal solution. We provide some computational results to corroborate the performance of the proposed algorithms.

Commentary by Dr. Valentin Fuster

Control Systems Design

2012;():403-411. doi:10.1115/DSCC2012-MOVIC2012-8564.

Two first order nonlinear controllers are introduced: partial state setting and error modulated state setting. Both controllers smoothly combine feedforward steady state predictions with integral control. In both controllers the mixing is controlled though the use of a single gain that determines how aggressively the steady state predictions are used. Steady state error bounds are presented for each controller. For error modulated state setting control steady state errors can be eliminated if the maximum expected error in the steady state predictions is known. Both controllers are demonstrated to improve the transient response of a simulated nonlinear second order variable gain plant when compared to a well tuned proportional plus integral controller.

Topics: Steady state
Commentary by Dr. Valentin Fuster
2012;():413-421. doi:10.1115/DSCC2012-MOVIC2012-8704.

This paper presents a model-based automated controller tuning algorithm for repetitive references. A key challenge in automated tuning is guaranteeing stability of the closed loop system during the tuning process. In order to guarantee the system stability while tuning, we reformulate the traditional controller tuning problem into the tuning of the Youla parameterized version of the controller. Thus at each iteration of the tuning process, the Youla parameter is iteratively tuned to minimize a given quadratic cost function. This makes the optimization problem affine in the parameters and at the same time, eliminates the need for checking stability at each iteration. The proposed algorithm can be used for (1) auto-tuning for optimizing a known controller, or (2) auto-tuning a controller from some arbitrary controller candidates. These capabilities of the model-based tuning algorithm are finally demonstrated experimentally on a precision linear stage. Criteria for convergence of the tuning law has also been presented.

Commentary by Dr. Valentin Fuster
2012;():423-432. doi:10.1115/DSCC2012-MOVIC2012-8732.

In this paper theoretical modeling, identification and model validation of a hybrid magnetic ball suspension system incorporating both permanent and electric magnets are investigated. Furthermore, the deviation of the theoretical model has been investigated using Comsol Multiphysics® modeling. Simple PD, PID and PID with pre-filter controllers have been designed based on the linearized model of the system. A nonlinear simulation using Simulink® has been carried out using the derived model and has been verified by experimental results. The suggested controller has been applied to the nonlinear simulation and the real-time maglev ball experiment to manage the ball position.

Commentary by Dr. Valentin Fuster
2012;():433-440. doi:10.1115/DSCC2012-MOVIC2012-8768.

In this paper, a method is provided for determining all fractional-order (FO) proportional–integral-derivative (PID) controllers that robustly stabilize a single-input and single-output (SISO) transfer function of arbitrary order with a time delay. The parameters of the FO PID controllers are determined in the frequency domain and are given in terms of the proportional gain Kp, integral gain Ki, and derivative gain Kd. In this paper, they will be plotted on the (Kp, Ki), (Kp, Kd), and (Ki, Kd) planes. For a robust stability condition, a multiplicative weight is selected to bound all multiplicative errors of the closed-loop system. In particular, this technique can be used even when the transfer function of a system is not available, as long as the system frequency response can be obtained. An example is given to illustrate the applicability and effectiveness of the method presented.

Commentary by Dr. Valentin Fuster
2012;():441-447. doi:10.1115/DSCC2012-MOVIC2012-8807.

Time delays in the control loop complicate the analysis and design of controllers. The use of a Smith predictor is a well-known technique for dealing with delays in stable systems. This paper presents a method for finding all stabilizing Proportional Integral Derivative (PID) controllers when the Smith predictor is used. This paper introduces a general method to determine all achievable PID controllers even when there is a mismatch between the plant and Smith predictor model. A key advantage of this approach is that it requires only the frequency response of the plant.

Commentary by Dr. Valentin Fuster

Cooperative and Decentralized Control

2012;():449-457. doi:10.1115/DSCC2012-MOVIC2012-8688.

This paper revisits an earlier investigation on a leader-follower consensus problem [1] affected by multiple time delays. The new perspective is in the use of a novel analysis technique which provides exact, exhaustive and explicit stability margins in the domain of the time delay. This methodology resides within the Cluster Treatment of Characteristic Roots (CTCR) paradigm, which is applied after a block diagonalization of the system takes place. Further novelty is described in the first stage of CTCR. At this step, the determination of the potential stability switching curves is achieved utilizing the concept of Spectral Delay Space, SDS. Example cases are provided in order to show the effectiveness of this procedure, and its advantages with respect to the peer treatments.

Topics: Stability , Delays
Commentary by Dr. Valentin Fuster
2012;():459-466. doi:10.1115/DSCC2012-MOVIC2012-8733.

A procedure to analyze interaction in an experimental roll-to-roll system that uses a decentralized control strategy is presented in this paper. A Perron root based interaction metric is employed for the analysis. Experiments conducted on a roll-to-roll system are used to evaluate the interaction between different subsystems of the roll-to-roll system. To minimize interaction between subsystems of the roll-to-roll system, a procedure for designing pre-filters based on the Perron root of the system is also discussed in the paper. Experimental results with and without pre-filter clearly indicate the effectiveness of the pre-filter in minimizing interaction. Discussions regarding the roll-to-roll application, stability considerations and insights on using the Perron root based interaction measure for decentralized control applications are also given.

Commentary by Dr. Valentin Fuster
2012;():467-474. doi:10.1115/DSCC2012-MOVIC2012-8772.

Set-point control problem for robotic manipulators when signals are exchanged via a delayed communication channel is studied in this paper. The gravitational effects, which were not considered or were assumed to be pre-compensated in the previous research, are considered in this paper. Using an appropriately defined controller with gravity compensation, it is first shown that simply utilizing the scattering variables can stabilize the closed-loop system in the presence of constant delays; however, position regulation cannot be guaranteed. Therefore, we study a new control algorithm where explicit position feedback, in conjunction with scattering variables is used, to guarantee both stability and tracking performance. Moreover, the efficacy of this architecture to handle time-varying input/output delays is also demonstrated. The proposed algorithm is numerically validated on a two-degree-of-freedom manipulator for both constant and time-varying delays.

Commentary by Dr. Valentin Fuster
2012;():475-481. doi:10.1115/DSCC2012-MOVIC2012-8789.

In this paper, we propose a fully distributed, scalable method of controlling agents with nonholonomic constraints using a Morse potential function. This method successfully controls a swarm of differential-drive (unicycle-type) agents to stable and predictable formations whose structures are not defined a priori. The system achieves a stable, minimal energy state.

We consider the effect of numerosity constraints, as observed in birds and fish in their shoaling and flocking behavior as a mechanism of reducing complexity, in the interest of achieving fully distributed control over a swarm of any size. The application of numerosity constraints to a swarm system allows the swarm to grow without bound and with no increase in required processing capability of the individual agents. We explore this parameter as a method of minimizing processing and storage requirements while still achieving the qualitative swarm performance. Results from simulations are given for swarms ranging in size over N = {6,…,100} acting under our proposed controller as applied to differential-drive (unicycle-type) robots.

Commentary by Dr. Valentin Fuster
2012;():483-489. doi:10.1115/DSCC2012-MOVIC2012-8868.

Synchronization of coupled laser arrays is required in many applications of high-power laser systems. While the problem is approached by numerical or experimental methods traditionally, we propose a new approach to rigorously characterize the synchronization condition inspired by recent advances in cooperative control. We study synchronization of an array of coupled solid state lasers where each individual laser is modeled by a second-order nonlinear oscillators. We analyze synchronization conditions over a mean-field model for all-to-all coupling configuration, and prove that the coupled lasers with identical frequencies can be stabilized on the synchronization state for any positive coupling strength. We then extend the all-to-all coupling to the limited communication case, and similar synchronization conditions are proved for undirected connected graphs. Our analysis is conducted using tools from algebraic graph theory and Lyapunov dynamic system theory. Simulation examples are given to illustrate the results.

Commentary by Dr. Valentin Fuster

Dynamic System Modeling

2012;():491-498. doi:10.1115/DSCC2012-MOVIC2012-8700.

In this research, color-depth (RGB-D) camera pairs, like the Microsoft® Kinect™, are used to enable one-point visual odometry. In the proposed algorithm, features are detected in the color image using the speeded-up robust features (SURF) algorithm and converted into 3-dimensional (3D) feature locations using information from the depth image. These 3D features allow feature tracking between successive images based on spatial proximity. An inverse kinematic solution is used to calculate the visual odometry between frames based on the 3D feature matches. The proposed method supports visual odometry measurement with a single feature correspondence between frames. The proposed algorithm is implemented on a small, wheeled mobile robot (WMR) and evaluated experimentally in a series of representative, indoor, operational contexts. The preliminary results demonstrate that the proposed approach accurately tracks the motion of the robot to within 4–6% error, which is comparable to other more computationally expensive algorithms.

Commentary by Dr. Valentin Fuster
2012;():499-508. doi:10.1115/DSCC2012-MOVIC2012-8703.

In this paper, we have considered the application of Stackelberg game theory to driver behavior modeling in highway driving. Our investigation focuses on developing a 3-person Stackelberg game that simulates driver’s reasoning with provisions to incorporate drivers’ dispositions so that the driver’s reactions can result in different interactions. In order to include uncertainty, the drivers’ intentions are also limited or changed by the driver’s dispositions. We present two-vehicle and five-vehicle simulations that show different drivers’ behaviors according to the dispositions and their interactions.

Topics: Highways
Commentary by Dr. Valentin Fuster
2012;():509-518. doi:10.1115/DSCC2012-MOVIC2012-8748.

Many small size field robots experience heat dissipation problems due to their compact construction, the sealing of compartments to protect against dirt and moisture, and the lack of space and/or power to support an active cooling system. Overheating may cause early failure of their components or even necessitate a shutdown of the robot mid-mission. In this paper we introduce the concept of temperature aware operations in which a robotic system uses thermal models to predict, and thus avoid, overheating events (by either warning the operator or adjusting its behavior autonomously). In this work we develop a model of a small robot’s thermal dynamics based on gray box system identification techniques. We show that this model can predict system temperatures over a 20 minute horizon with a typical accuracy of a few degrees Celsius, thus demonstrating the feasibility of temperature aware operations.

Commentary by Dr. Valentin Fuster
2012;():519-528. doi:10.1115/DSCC2012-MOVIC2012-8755.

Parameter tuning of air conditioning system models is a critical step in constructing accurate dynamic models for use in control design and optimization. However, traditional manual or simulation-based methods of parameter tuning are tedious, time-consuming, or simply infeasible due to the large number of parameters. In this paper, an approach to tune multiple parameters of HVAC&R systems is proposed and shown to be effective in practice. An accurate and computationally efficient model is derived, and a wavelet decomposition approach is adopted in formulating the cost function that seeks to minimize the error between the predicted and measured data. Wavelets are advantageous in handling the multi-domain signals and capturing large and small dynamic features. In order to reduce the optimization time, a hybrid parameter tuning method is proposed. This method first uses a stochastic global tuning method (genetic algorithm) to find a set of estimated parameters, and these parameters serve as the initial values of a gradient search method. The results show that the proposed method can effectively tune many simultaneous parameters using large data sets.

Commentary by Dr. Valentin Fuster
2012;():529-534. doi:10.1115/DSCC2012-MOVIC2012-8867.

The aim of this effort is to develop a model of an actual unmanned ground vehicle system for computer simulations in order to evaluate guidance algorithms developed for autonomous waypoint navigation and obstacle avoidance. Simulation is a vital tool for the development of autonomous systems. Simulating individual parts and units of the system can help identify flaws in its design or implementation. In the Matlab-Simulink environment, a kinematic based model of an skid-steer ground vehicle is designed. Furthermore, a model of quadrature encoders for position estimation, and a laser range finder (LRF) sensor model for obstacle detection are also created. Two different groups of experiments are performed to test the performance of the proposed models. Experimental results indicate that the models can adequately simulate the actual vehicle behaviors. This effort is part of an ongoing research to create fully autonomous UxVs capable of waypoint navigation and obstacle avoidance.

Commentary by Dr. Valentin Fuster
2012;():535-541. doi:10.1115/DSCC2012-MOVIC2012-8870.

Motivated by the interest to increase production throughputs of immersion lithography machines, wafers are scanned at increasingly high velocities and accelerations, which may result in liquid loss at the receding contact line. The dynamic characteristics of the immersion fluid with free boundary play an important role for fluid management system, and are concerned in various potential immersion unit designs. To offer intuitive insights into the dynamic effects of the immersion fluid due to scan speeds, a lumped-parameter model is developed to characterize the hydrodynamics of the immersion flow process. To validate the model, meniscus behavior information under dynamic conditions is extracted experimentally and analyzed using image processing techniques. The reduced model agrees qualitatively well with the experimental data revealing that parts of the surface tension have an effect on the dynamic response of the menisci similar to that due to a pure time delay in the system.

Commentary by Dr. Valentin Fuster

Dynamical Modeling and Diagnostics in Biomedical Systems

2012;():543-551. doi:10.1115/DSCC2012-MOVIC2012-8579.

This paper presents a novel active non-intrusive system identification (SYSID) approach to cardiovascular monitoring. The proposed approach uses a dual blood pressure cuff as an actuator as well as a pressure transducer for cardiovascular SYSID. In this paradigm, the dual blood pressure cuff excites the cardiovascular system to create rich trans-mural pressure waves traveling in the cardiovascular system, which are in turn measured via cuff pressure oscillations. Mathematical model was developed to describe the propagation of arterial and cuff excitation pressure waves in the cardiovascular system, which was subsequently used to study the effect of cuff maneuvers on trans-mural pressures and also to develop a methodological framework to reconstruct trans-mural pressure waveforms from cuff pressure oscillation measurements. This paper successfully demonstrated that 1) cardiovascular system can be excited non-intrusively via active cuff maneuvers, and 2) arterial and trans-mural pressure waveforms can be reconstructed accurately by judiciously processing cuff pressure oscillations.

Commentary by Dr. Valentin Fuster
2012;():553-560. doi:10.1115/DSCC2012-MOVIC2012-8591.

This work is motivated by the Physionet/Computing in Cardiology Challenge 2011, “Improving the quality of ECGs collected using mobile phones”. The advancement of cell phone technology makes it possible to collect, analyze, and transmit vital physiological signals in real time, promising to a new era of tele-health care. However, noises and artifacts can lead to false readings and thus misdiagnosis. Unlike common methods based on time series analysis techniques, we analyze the quality of the 12-lead ECG using image processing techniques. Various image patterns are used as features to distinguish between low- and high-quality signals. When tested on a data set from the Physionet Challenge 2011, the analyses yield up to 94.36% accuracy. The work here provides an interesting alternative for ECG quality evaluation. The technique will have particular use for ECGs scanned from paper recordings.

Commentary by Dr. Valentin Fuster
2012;():561-566. doi:10.1115/DSCC2012-MOVIC2012-8601.

This paper is concerned with the prediction of the occurrence and severity of Periventricular Leukomalacia (PVL), a form of white-matter brain injury that occurs often in neonates after heart surgery. The data which is collected over a period of twelve hours after the cardiac surgery contains vital measurements. The fact that the exact cause of the PVL have still not been clearly understood renders a mathematical modeling approach for fault diagnosis impractical, if not impossible. Hence, the decision tree classification technique has been selected for its capacity for discovering rules and novel associations in the data. It classifies groups based on reducing uncertainty in the classified data. From a physiological point of view we know that there are several regulatory mechanisms responsible for fluctuation of the hemodynamic variables at different time scales. To discover the most important active physiological components which might lead to the occurrence of PVL and possibly affect its severity, we focus on the variation in the data in one minute, twenty minute and two hour periods. We calculate the energy of continuous wavelet transform coefficients of vital data at these time scales as a measure of variation in the different time frames. The results obtained from developing decision tree classifiers show that among all variations in all the variables, 2 hour and 20 minute variations in the heart rate, 1 minute and 20 minute variations in Oxygen saturation, and 2 hour variations in the mean arterial pressure are the most important parameters to be able to predict PVL occurrence.

Commentary by Dr. Valentin Fuster
2012;():567-572. doi:10.1115/DSCC2012-MOVIC2012-8684.

In this study, we applied the continuous wavelet transform (CWT) to determine electroencephalogram (EEG) discriminating features of Alzheimer’s Disease (AD) patients compared to control subjects. The EEG was recorded from 24 subjects including 10 AD and 14 age-matched control during six sequential resting eyes-closed (EC) and eyes-open (EO) states followed by cognitive tasks and auditory stimulation. We computed the absolute and relative geometric mean powers of Morlet wavelet coefficients at different scale ranges corresponding to the major brain frequency bands. Kruskal-Wallis statistical testing method was then employed to determine the statistically significant features of the cohort geometric means. The results show that there are many discriminating features of AD patients at several different brain major frequency bands, particularly during the second and third EC and EO states. Since many features were identified, a decision tree algorithm was employed to classify the most significant one(s). The algorithm found the absolute power of θ frequency band during the second EO state to be higher for all AD patients when compared to control subjects and identified it as the most significant discriminating feature.

Commentary by Dr. Valentin Fuster
2012;():573-578. doi:10.1115/DSCC2012-MOVIC2012-8804.

This paper presents analysis of a one degree of freedom human postural control model. The central nervous system is captured in the model as an intermittent control law. Parametric and stability analysis of this single degree-of-freedom model are presented. The L2 Norm of the response is used as a measure of stability. This study shows that joint stiffness and associated time delay parameter are the most sensitive towards the response dynamics. In addition, a comparison of parametric linear stability boundary with the L2 Norm estimates is presented.

Topics: Stability
Commentary by Dr. Valentin Fuster

Dynamics and Control in Medicine and Biology

2012;():579-583. doi:10.1115/DSCC2012-MOVIC2012-8631.

This paper presents a modeling framework for an intracellular signaling network based on formalisms derived from the fundamental concepts in probability theory. Cellular behavior is mediated by a network of intracellular protein activations that originate at the membrane in response to stimulation of cell surface receptors. Multiple protein signal transductions occur concurrently through diverse pathways triggered by different extracellular cues. Through crosstalk, these pathways intersect at various node proteins. The state of a particular node protein is dependent on the binding order of molecules from various pathways. The probability of a particular binding order is evaluated using state dependent transduction time probabilities associated with each pathway. In this way, the probability of the cell to be in a given internal state is tracked and used to gain insight into the cell’s phenotypic behavior. A simulation example illustrates the approach. Future work will incorporate the proposed method into the development of a feedback control strategy for the development of an in silico control design of endothelial cell migration during angiogenesis.

Topics: Modeling
Commentary by Dr. Valentin Fuster
2012;():585-589. doi:10.1115/DSCC2012-MOVIC2012-8646.

A beat-to-beat alternation in the duration of the action potential (APD) of myocytes, i.e. alternans, is believed to be a direct precursor of ventricular fibrillation (VF) in the whole heart. A common technique for the prediction of alternans uses the restitution curve, the nonlinear functional relationship between the APD and the preceding diastolic interval (DI). It was proposed that alternans appears when the magnitude of the slope of the restitution curve exceeds one, known as the restitution hypothesis. However, this restitution hypothesis was derived in the presence of feedback, i.e. the partial dependence of the DI on the immediate preceding APD. Physiologically, the heart rate exhibits substantial variations, known as heart rate variability (HRV), which might affect the feedback relationship. In this manuscript, we aim to investigate the effect of feedback on alternans formation in the heart.

Topics: Feedback
Commentary by Dr. Valentin Fuster
2012;():591-595. doi:10.1115/DSCC2012-MOVIC2012-8657.

Toxoplasma gondii is a protozoan capable of replicating sexually in cats and asexually in other warm-blooded animals. When the parasites first enter a host, the replication process begins very quickly as the free parasites invade healthy host cells, until a host-mediated immune response suppresses replication of the rapidly replicating parasites. To study the effects of this process, a data-driven model is developed where the number of parasites and the IFN-γ level are selected as state variables. The reproduction number, R0 is calculated to determine the criteria for sustaining infection within the host. Stability analyses are carried out for the endemic equilibriums. Parameter fitting is performed using maximum likelihood to match existing data to the model. Finally, an optimal control problem is formulated to investigate the optimal immune strategy when minimizing the negative host reaction to increased IFN-γ levels. The results shed light into the role the immune response plays in suppressing acute infection of Toxoplasma gondii.

Commentary by Dr. Valentin Fuster
2012;():597-603. doi:10.1115/DSCC2012-MOVIC2012-8718.

A challenge in computational neuroscience is to develop neuronal models with minimal complexity to enable the development of large networks of brain regions to understand the functions they might implement. Biologically realistic neuronal models aim to replicate the electrical functioning of the cell at the level of ionic channels, and sometimes even include complex dendritic trees and intracellular molecular cascades. Such models are computationally intensive and not suitable for implementation in networks. Reduced order models, such as the one proposed by Izhikevich (2007), aim to preserve the key neurocomputational properties and so form an attractive alternative for implementation in large network models.

A systematic methodology is proposed to convert a biologically realistic neuronal model to an equivalent reduced order Izhikevich model, given the key morphological features that impact network structure. For instance, multiple compartments may be required, afferents may be distributed differently between basal and apical dendrites, and the synaptic plasticity mechanisms may be different at different dendritic sites. This will require careful design of the morphology of the reduced order model. By current mapping and phase portrait analysis, we suggest how such a biologically realistic model can be converted to a reduced order Izhikevich model that preserves the key neurocomputational and morphological features.

Commentary by Dr. Valentin Fuster
2012;():605-613. doi:10.1115/DSCC2012-MOVIC2012-8738.

Synaptic plasticity plays an important role in mediating the behavior of neuronal networks including learning and memory. However, not much is understood about how it is implemented in vivo, especially in the complex dendritic trees of neurons in the mammalian fear circuit. Neuromodulation is very important in such circuits and its impact on plasticity has not been characterized adequately. We will consider the calcium-based plasticity mechanisms implemented in recently reported models, and study how neuromodulation and dendritic morphology impact plasticity.

Norepinephrine and dopamine are the key neurotransmitters that modulate the neurons in the mammalian amygdala during Pavlovian fear conditioning. We will consider their impact on long-term potentiation (LTP) and long-term depression (LTD) using the NMDAR-based calcium learning rule implemented in the post-synapse. The calcium based learning rule requires depolarization of local membrane potentials in dendrites to relieve the Mg2+ block of NMDA receptors. A high level depolarization produces LTP, while LTD is caused by lower levels of depolarization. Hebbian pairing typically requires both dendritic spiking and back-propagating action potentials to achieve LTP. However, biological reports show that a certain level of dendritic spiking may by itself be sufficient to cause LTP/LTD. We will investigate all these mechanisms comprehensively using compartmental models of different complexity.

Topics: Plasticity
Commentary by Dr. Valentin Fuster

Estimation and Fault Detection

2012;():615-624. doi:10.1115/DSCC2012-MOVIC2012-8533.

The problem of state estimation for a spherical pendulum is studied by comparing the performance of a set-bounded estimation scheme designed for dynamics evolving on the sphere with that of an extended Kalman filter. The set-bounded estimator uses a global, coordinate-free description of the dynamics of a spherical pendulum, while the extended Kalman filter uses a local description of the motion using spherical coordinates. The dynamics model of the pendulum is assumed to be known; however, measurements of the pendulum state have unknown errors. The extended Kalman filter is known to be optimal for the case of zero-mean white Gaussian noise, while the set-bounded filter assumes that the measurement noise is bounded by known ellipsoidal set. Results are presented for numerical simulations in which the pendulum system has measurement errors with various statistical distributions, allowing for a direct comparison between the performances of the two state estimation strategies.

Commentary by Dr. Valentin Fuster
2012;():625-632. doi:10.1115/DSCC2012-MOVIC2012-8600.

A novel robust control technique for discrete time nonlinear systems with random actuator failures is proposed in this paper. This controller is optimally robust for actuator failures in achieving general performance criteria ranging from quadratic optimality with inherent asymptotic stability property to various forms of quadratic dissipative type of disturbance reduction. By solving a state dependent linear matrix inequality at each time instant, the control solution is found which satisfies these general performance criteria. The effectiveness of the proposed technique is demonstrated by simulations of the control of the inverted pendulum on a cart.

Commentary by Dr. Valentin Fuster
2012;():633-640. doi:10.1115/DSCC2012-MOVIC2012-8650.

We report a hybrid particle filter for measurement of specific heat, cp, and thermal conductivity, κ, of a micro- or nanowire using the well-known 3ω method. In the 3ω method, current at frequency ω is passed through the sample, and the 3ω component of the voltage response is measured. The data analysis approach used by previous authors neglects time-varying and higher-order terms in a series expansion of the 1D transient heat equation. This approximation is inaccurate at high currents and high frequencies. We remove this source of estimation error with a transient electrothermal finite element model. A Kalman filter estimates the temperature distribution in the wire, while a particle filter estimates κ and cp. Experiments on a ∼ 30 μm diameter platinum wire confirm that current and frequency sensitivity are reduced using our approach. Furthermore, our method is applicable to compensation of other geometric and material effects that cannot be handled by the previous formulation.

Commentary by Dr. Valentin Fuster
2012;():641-648. doi:10.1115/DSCC2012-MOVIC2012-8743.

In this paper, a robust nonlinear observer is proposed to estimate the State of Charge (SOC) of a Li-ion battery, a problem which is critical in designing efficient Li-ion battery management systems and energy management systems in battery-powered applications. An equivalent circuit is used to model the battery behavior. The advantage of this model is that a straightforward identification process can be utilized for parameter identification. Although this model can capture battery dynamics very well for various operating conditions, modeling errors and also unknown disturbances will still be present; therefore, the battery management system should be able to take these uncertainties into consideration. To this end, the proposed estimation algorithm is designed to be robust against uncertainties. Furthermore, the observer does not impose any constraints on the battery current or the SOC relationship with Open Circuit Voltage (OCV). In other words, this algorithm does not require the battery current to be constant or the SOC-OCV relationship to be linear. Global asymptotic convergence of the estimated SOC to its true value is proved via the Lyapanov Stability Theorem. Simulation and experimental results demonstrate the effectiveness of the proposed method.

Commentary by Dr. Valentin Fuster
2012;():649-655. doi:10.1115/DSCC2012-MOVIC2012-8838.

For linear and well-defined estimation problems with Gaussian noise, the Kalman filter (KF) yields the best result in terms of estimation accuracy. However, the KF performance degrades and can fail in cases involving large uncertainties such as modeling errors in the estimation process. The smooth variable structure filter (SVSF) is a model-based estimation method built on sliding mode theory with excellent robustness to modeling uncertainties. Wavelet theory has attracted interest as a powerful tool for signal and image processing, and can be used to further improve estimation accuracy. In this paper, a new filtering strategy based on stationary wavelet theory and the smooth variable structure filter is proposed. This strategy, referred to as W-SVSF, is applied on an electrohydrostatic actuator (EHA) for the purposes of state estimation. The results of the W-SVSF are compared with the standard KF, SVSF, and combined W-KF.

Topics: Filters , Wavelets
Commentary by Dr. Valentin Fuster

Estimation and Fault Detection for Vehicle Applications

2012;():657-664. doi:10.1115/DSCC2012-MOVIC2012-8528.

This paper investigates the impact of fuel property variations on the rail pressure fluctuations in high pressure common rail (HPCR) systems and explores the possibility of indentifying fuel physical property based on the measurement of a rail pressure sensor. Fluid transients, particularly the water hammer effect, in a HPCR system are discussed and the 1D-governing equations are given. A HPCR model is developed in GT-Suite. The injectors, a three-plunger high pressure pump, and a pressure control valve are modeled in a relatively high level of detail. Five different fuels are modeled and their properties including density, bulk modulus, and acoustic wave speed are validated. Simulation results are obtained under different conditions with variable rail pressures and injection durations. The results show that natural frequency of the common rail varies with the fuel type filled in it. By applying the Fast Fourier Transform to the pressure signal, the differences of fuel properties can be revealed in the frequency domain. Since the rail pressure natural frequency is affected by the acoustic wave speed in the fuel, it can be concluded that this approach not only works for biodiesel blend level estimation, but also universally applies to the identification of various fuels and their blends as long as the acoustic wave speed in the fuel is known and the difference comparing to regular diesel is discernable.

Commentary by Dr. Valentin Fuster
2012;():665-672. doi:10.1115/DSCC2012-MOVIC2012-8530.

In this paper, a real-time estimator, based on an extended Kalman filter (EKF), for the position of vehicle center of gravity (CG) is proposed. Accurate knowledge of the CG longitudinal location and the CG height in the vehicle frame is helpful to the control of vehicle motions, especially for lightweight vehicles (LWVs), whose CG positions can be substantially varied by freight goods or passengers onboard. The proposed estimation method, unlike many existing ones, extracts signals only from vehicle longitudinal maneuvers in which road course elevation may exist. A three-state vehicle dynamic model, including the longitudinal velocity, the front-wheel angular speed, and the rear-wheel angular speed of the vehicle, is employed in the EKF formulation. With the help of the GPS altitude measurement, the road grade, which provides excitation for the estimation of the CG height, can also be obtained using a typical Kalman filter. Simulation studies based on a CarSim® vehicle model show that the proposed estimator is capable of accurately estimating both the CG longitudinal location and the CG height without a priori knowledge of the tire-road contact condition. Moreover, though the performance of the CG height estimation largely depends on the road grade variations, the CG longitudinal location can always be accurately estimated, even on a horizontal road.

Commentary by Dr. Valentin Fuster
2012;():673-680. doi:10.1115/DSCC2012-MOVIC2012-8609.

Vibration based diagnosis to detect gear tooth damage in gearboxes has been studied widely and it can assist in scheduling maintenance and reducing capital losses that may result from gearbox failures. However, such vibration based techniques are difficult to implement in planetary gearboxes due to the complex nature of measured vibration spectrum resulting from rotating planets with respect to the stationary transducer mounted on the gearbox housing. Motor current signal analysis (MCSA) provides an alternative and non-intrusive way to detect mechanical faults through electrical signatures. So far, no investigation has been reported in literature to monitor a planetary gearbox in an electromechanical drive-train using MCSA because of the difficulties in modeling the planetary gear-set such as a large number of degrees of freedom and nonlinearity associated with tooth separations. In this paper, a lumped parameter model of an electro-mechanical drive-train has been developed, which consists of a permanent magnet synchronous machine (PMSM) connected to a load through a planetary gearbox. Afterwards, a seeded tooth defect is introduced into the electro-mechanical model to show that MCSA can successfully provide valuable diagnostic information regarding the planetary gearbox failure. Finally, the time waveform, as well as, the Fourier transform and Morlet wavelet transform of the PMSM stator current are presented to demonstrate the capability to detect the gear tooth fault and its severity in planetary gearbox using MCSA.

Commentary by Dr. Valentin Fuster
2012;():681-688. doi:10.1115/DSCC2012-MOVIC2012-8633.

This project focuses on the use of magnetoresistive and sonar sensors for imminent collision detection in cars. The magnetoresistive sensors are used to measure the magnetic field from another vehicle in close proximity, so as to estimate relative position, velocity and orientation of the vehicle from the measurement.

An analytical formulation is presented for the planar variation of the magnetic field from a car as a function of two dimensional position and orientation. While this relationship itself can be used to estimate position and orientation, a challenge is posed by the fact that the parameters in the analytical function vary with the type and model of the encountered car. Since the type of vehicle encountered is not known apriori, the parameters in the magnetic field function are unknown. The use of both sonar and magnetoresisitive sensors and an adaptive estimator is shown to address this problem. While the sonar sensors do not work at very small inter-vehicle distance and have low refresh rates, their use during a short initial time duration leads to a reliable estimator. Experimental results are presented for a laboratory wheeled car door and show that planar position, relative angular position and orientation can be accurately estimated for a range of relative motions at different oblique angles.

Commentary by Dr. Valentin Fuster
2012;():689-695. doi:10.1115/DSCC2012-MOVIC2012-8839.

Biodiesel is a renewable alternative fuel that produces lower exhaust emissions with the exception of nitrogen oxides (NOx) when compared to conventional diesel fuel. Fuel blend information is useful during engine operation for optimizing emissions and performance. Therefore, online estimation of biofuel content is a critical step in allowing diesel engines to maintain performance while simultaneously meeting emission requirements when operating on biodiesel blends. Presented in this paper is a model-based biodiesel blend estimation strategies using crankshaft torsionals. A sensitivity analysis investigation is conducted for the method to quantify robustness of the proposed fuel blend estimation methods.

Topics: Sensors , Biodiesel
Commentary by Dr. Valentin Fuster

Fluid Power Systems

2012;():697-706. doi:10.1115/DSCC2012-MOVIC2012-8543.

Developed is a model based fault detection, isolation and estimation study of a diesel engine air handling system. Diagnostic is based on an air path model, wherein the coefficients of a healthy system model are compared with that of adapted online coefficients during steady state operating conditions. Fault estimation is realized by analyzing the residual between the adapted coefficient and those of the original (healthy) model. This approach is shown to be robust to modeling errors, sensor noise and process variability. The proposed algorithm is experimentally validated on the intake air path of a Cummins 5.9 L diesel engine and is shown to effectively perform intake leakage FDIE.

Topics: Diesel engines
Commentary by Dr. Valentin Fuster
2012;():707-714. doi:10.1115/DSCC2012-MOVIC2012-8707.

A system capable of simulating the dynamic behavior of a single DOF joint of the human arm was developed. The inertia, stiffness and damping produced by the passive tissue, as well as the muscle torque and its variable stiffness and damping specific to human shoulder, elbow, wrist, and finger joints can be accurately reproduced. The torque created by external loads and by active assistive devices for people with disabilities, can be emulated. This allows the system to be used for safety and performance evaluation of control algorithms implemented in active orthotic, prosthetic, and rehabilitation devices in a safe and highly reproducible environment, before testing on actual human subjects.

Commentary by Dr. Valentin Fuster
2012;():715-721. doi:10.1115/DSCC2012-MOVIC2012-8784.

To increase the efficiency of hydraulic systems by eliminating valve throttling losses, a direct displacement open circuit is proposed to control a single rod hydraulic actuator. The circuit provides three control inputs, including the displacement of a variable displacement pump, the opening area of a proportional valve, and the position of a directional valve. Pump control has a low bandwidth, but the efficiency is high due to the lack of throttling losses. Valve control has a high bandwidth, but the throttling loss is high. A novel approach has been proposed to distribute the control efforts between the pump and the proportional valve considering both control bandwidth balancing and throttling loss reduction. The proportional valve will follow a high frequency opening profile, while the nominal valve opening is large, and the pump output flow will follow a low frequency demand. Experimental results validate the effectiveness of the proposed approach.

Commentary by Dr. Valentin Fuster
2012;():723-732. doi:10.1115/DSCC2012-MOVIC2012-8836.

In this study, a controller is designed for a hydraulic actuator of a single axis durability test rig. The mathematical equations of the system are derived and a MATLAB Simulink® model is developed. The non-linear equations defining the system dynamics are linearized and the system is represented in state space. For the position control of the hydraulic actuator, a combined feedforward and feedback controller is designed. The feedforward controller estimates the force transmitted between the vehicle and the hydraulic actuator. Considering this force as a disturbance, a feedforward valve command signal is generated for compensation. In the design of feedback controller, the hydraulic actuator is considered independent of the load. Neglecting the valve dynamics and assuming the actuator chamber pressures are linearly dependent, a third order actuation system model is derived and a LQR controller is designed. In the design of LQR controller, the matrix penalizing the states is formed by approximating the impulse response of the closed loop system to an ideal second order response profile rather than by using iteration. The closed loop state space representation of the complete system is derived and its response is compared with the non-linear model.

Commentary by Dr. Valentin Fuster
2012;():733-741. doi:10.1115/DSCC2012-MOVIC2012-8845.

In this work, a method is developed for modeling uncertainty in the frequency domain which can be used to predict, or design systems with a specified, probability of failure to meet performance objectives. The work is an application of a probability stability/performance analysis technique being developed by the authors. An example of this technique is presented using a pilot operated proportional control valve (POPCV) system. Thirty replications of the pilot stage of a proportional control valve system were obtained by the University of Missouri and tested with one main stage valve. A model of the system is developed and used in Monte Carlo simulations based on distributions of the physical variations of the pilot valve. A mixed sensitivity H-infinity control system is developed using a frequency domain uncertainty model that only bounds a fraction of specified plants. It is shown that when the controller is implemented in a closed-loop system, only the fraction of the plants bounded in the uncertainty model are able to meet specified performance objectives. This technique allows a control designer to design higher performance control systems for mass produced systems with model uncertainty at the expense of having a specified fraction of systems not achieve a performance objective.

Commentary by Dr. Valentin Fuster

Human Assistive Systems and Wearable Robots

2012;():743-752. doi:10.1115/DSCC2012-MOVIC2012-8651.

Powered ankle-foot orthoses have significant potential as both assistance and rehabilitation devices for individuals with lower limb muscle impairments. Recently, we developed an untethered pneumatically powered ankle-foot orthosis (PPAFO) for outside-the-lab walking assistance or therapy. It is critical to recognize gait modes (i.e. level walking, stair ascend/descend) because improper actuation can dramatically increase fall risk. Gait mode recognition is a challenging task for the PPAFO because the sensor array is very limited and a new mode must be recognized at the earliest possible time to prevent inappropriate actuation and decrease fall potential. While manual mode switching is implemented in most powered orthotic/prosthetic device control algorithms, we propose an automatic gait mode recognition scheme by tracking the 3D position of the PPAFO from an inertial measurement unit (IMU). The experiment results showed that, with an optimized threshold, the controller was able to identify the position and gait mode at the very beginning of the mode change, to allow for proper actuation control.

Commentary by Dr. Valentin Fuster
2012;():753-758. doi:10.1115/DSCC2012-MOVIC2012-8674.

This paper presents a new method to investigate the multivariable time-varying behavior of the ankle during human walking, and provides the first experimental results from treadmill walking. A wearable ankle robot with an ensemble-based linear time-varying system identification method enabled identification of transient ankle mechanical impedance in 2 degrees of freedom, both in the sagittal and frontal planes. Several important issues of the ensemble-based identification method in practical measurements are discussed, especially a strategy to solve the limitation of the method which assumes that the system undergoes the same time-varying behavior on every stride. The suggested method was successfully applied to 15 minutes of human walking on a treadmill. Experiments with 10 young healthy subjects showed clear time-varying behavior of ankle impedance across the gait cycle, except the mid-stance phase. Interestingly, most subjects increased ankle impedance just before heel strike in both degrees of freedom. Interpretation of impedance changes was consistent with analysis of electromyographic signals from major muscles related to ankle movements.

Commentary by Dr. Valentin Fuster
2012;():759-768. doi:10.1115/DSCC2012-MOVIC2012-8706.

Multiple Sclerosis (MS) is a complex neurological disease that destroys the myelin sheath of the nerves, consequently affecting the motor control ability in numerous ways. It has been shown that people with MS have the potential to improve their functional ability by interacting with robotic training devices through assistive forces. In order to provide a complete and task specific therapy for people with complex neurological impairments, such as those with MS, it is important to take advantage of the robots’ ability to provide measurement and force feedback for more complex and realistic 3D motions. The complexity and randomness of natural task-oriented upper-limb motions can be highly preserved in Peg-in-Hole assessment and training methods, such as the widely used Nine-Hole Pegboard Test (NHPT). In the authors’ previous work, a virtual NHPT was developed and tested. This paper presents the augmentation of the virtual NHPT with assistive forces in order to become a physiotherapy and rehabilitation system. This system includes target trajectories that are based on the motion of healthy users and adaptive assistive forces that can maximize the benefits from rehabilitation. After development and validation, the system is evaluated on three people with MS. Each participant carried out the NHPT exercise nine times (trials): Initially two trials without application of assistive forces, then five robot-assisted trials (therapy session with application of forces), and two more trials at the end without forces. The results suggest that the system can be effectively employed for rehabilitation in complex movements; nevertheless, its effectiveness must be better grounded with extensive clinical trials.

Topics: Haptics , Robotics
Commentary by Dr. Valentin Fuster
2012;():769-777. doi:10.1115/DSCC2012-MOVIC2012-8753.

The power density and variable compliance in pneumatic actuators makes them an attractive option for actuation in human assistive devices. Interaction safety in these devices can be robustly achieved through energetically passive controllers. Efficacy of these controllers depends on appropriate definition of actuator energy function. In previous works, the energy function was defined by assuming the thermodynamic process in the actuator to be either isothermal or adiabatic. In the current paper an estimate of work potential suitable for passivity analysis of a single chambered pneumatic actuator with finite heat transfer is reported. The energy function is developed by maximizing the work done on the actuator to reach an equilibrium position. Optimal conditions show that the maximal solution is attained if the thermodynamic process is a combination of adiabatic and isothermal processes. Through this storage function it is shown that the heat transfer has dissipative affect on the power flow in the pneumatic actuator, irrespective of the chamber air temperature.

Commentary by Dr. Valentin Fuster
2012;():779-786. doi:10.1115/DSCC2012-MOVIC2012-8771.

Providing powered joint actuation has been the latest trend in transfemoral prosthesis research. The capability of actively powering the joints enables the prosthesis to meet the energetic requirements of locomotion, and thus provides higher performance in restoring the lost lower limb functions in comparison with traditional passive prostheses. In this paper, a powered transfemoral prosthesis is presented, in which the knee and ankle are powered with pneumatic artificial muscle actuators, leveraging the multiple advantages of this unique actuator (such as large force output and high power density). To address the issue of uneven force capacity, a new variable-radius pulley-based mechanism is utilized, which enables the adjustment of the available actuation torque curve to better match the desired torque curve as dictated by the locomotive requirements. The design details of the prosthesis are presented, and the prosthesis is expected to provide sufficient torque for an 85 kg user in various locomotion modes, including level walking at slow/normal cadence and stair ascent/descent.

Topics: Prostheses , Muscle
Commentary by Dr. Valentin Fuster
2012;():787-793. doi:10.1115/DSCC2012-MOVIC2012-8790.

A new type of wearable robot that provides a third and fourth arm for performing manipulative tasks with the wearer’s own arms is presented. These Supernumerary Robotic Limbs (SRL) work so closely with the human that he/she can potentially perceive them to be his/her own. The SRL consist of two independently acting robotic limbs that can function as either arms or legs to help the user position objects, lift weights, and maintain balance. These wearable robots are aimed to augment not only the strength and the precision of the human users, but also their range of skills and interactions with the environment. The guiding principles of the robotic design are safety, transparency and user comfort. Series viscoelastic actuators provide suitable joint torques while ensuring compliance and robust torque sensing. A Bowden cable transmission actuates the elbow joint, minimizing the robotic arms’ weight. A tuned elastic human-robot coupling ensures wearability and comfort. To quantify the mechanical advantage the SRL offers to the operator during use, joint torques generated in the human while performing static manipulation tasks have been reconstructed experimentally.

Commentary by Dr. Valentin Fuster

Human-in-the-Loop Systems

2012;():795-802. doi:10.1115/DSCC2012-MOVIC2012-8578.

Personal mobility vehicles (PMVs) as new individual transportation vehicles have been proposed around the world. It is important to ensure the safe operation of a PMV, especially when a PMV shares the space with a pedestrian. In this paper, in order to evaluate the influence of the size of a PMV on pedestrians, we measured the personal space of the pedestrians who walked against the PMV user. From the experimental results, it was shown that the width of the PMV and height of the step of the PMV significantly affected the personal space of pedestrians. Especially when the width and height becomes larger, the front personal space of the pedestrians expanded. In the simulation introducing the idea of personal space, the affinity between the PMV and pedestrians was quantitatively analyzed. It was found that the size of PMV affects to the affinity toward pedestrians.

Topics: Vehicles
Commentary by Dr. Valentin Fuster
2012;():803-807. doi:10.1115/DSCC2012-MOVIC2012-8623.

In this paper, stability of a two-wheeled inverted pendulum vehicle was evaluated using rider–vehicle modeling, electromyography of leg’s muscle of rider, and subjective evaluation. A rider and the vehicle are synthetically modeled as a series type double inverted pendulum. Utilizing auto-regressive exogenous (ARX) model methods, correlative relations were found between control gains and stability of the vehicle that achieves better ride comfort. The experimental results show that the higher the control gains, the smaller the activities of leg’s muscles. However, according to subjective evaluation of ride comfort, higher gains did not always achieve better ride comfort. Moreover, the rider-vehicle model has no unstable poles, even when the vehicle model has an unstable pole. The poles of the rider-vehicle model had tendency to have bigger negative real numbers, which suggests stability of the vehicle can be evaluated by the positions of the poles of the rider-vehicle model. Through identification of the rider-vehicle model, the control gains of the rider’s posture were calculated. It was found that the control gain against the rider’s posture is dominant, and the higher the control gains of vehicle, the smaller the gain against the rider’s posture. The results show stability of the vehicle can be evaluated by the control gains against the rider’s posture.

Commentary by Dr. Valentin Fuster
2012;():809-815. doi:10.1115/DSCC2012-MOVIC2012-8629.

An inverted pendulum-type PMV (personal mobility vehicle) has been attracting attention as a low-carbon vehicle. For many people who like to use the PMVs, ride comfort is important. However, the problem of ride comfort has not attracted much attention in previous studies. The vibration is one of the important indicators for evaluating ride comfort. The PMV is unstable system. Therefore, the vibration may be generated when the PMV is stabilized. This study investigates the horizontal and vertical vibration of the head of the occupant when the PMV runs on a road with disturbances in numerical simulations. Frequency characteristics of the inverted pendulum-type PMV is analyzed to verify what vibrational factors that worsens ride comfort are. To consider human vibration sensitivity, the frequency weighting proposed in ISO2631-1 is used as the evaluation standard. The improvement methods are proposed from both software and hardware, and it is confirmed that the proposed method can improve ride comfort.

Topics: Vibration , Pendulums
Commentary by Dr. Valentin Fuster
2012;():817-821. doi:10.1115/DSCC2012-MOVIC2012-8728.

The purpose of this paper is to develop human following control method of a porter robot in order to support workers in carrying heavy loads. This paper describes how to control human following method by calculating velocity vectors using the data obtained from ultrasonic sensors and by applying the inverse kinematics to wheeled skid-steering mobile robot. We adopt control method in constructing porter robot to support repair operations of asphalt pavements. As repair operations of asphalt pavements are performed in limited space, it will be necessary to restrict the movements of the robot as well. In this study, we adopted the self-localization by using odometory in order to limit the moving area of the robot. The experimental results indicated that the proposed method were effective in following humans even with a limitation on the moving area.

Topics: Robots
Commentary by Dr. Valentin Fuster
2012;():823-827. doi:10.1115/DSCC2012-MOVIC2012-8791.

This article deals with the mode analysis of the kinematic structure of human locomotion. We investigated human locomotion using singular value decomposition. From motioncaptured data of human locomotion, we extracted common basic movements and residual modes, and analyzed the kinematical structures. The results show the joint synergy that is derived by the residual modes (expresses strong dependencies on walking condition, and has bifurcation properties according to walking speed.

Commentary by Dr. Valentin Fuster
2012;():829-835. doi:10.1115/DSCC2012-MOVIC2012-8864.

Avoiding falls require fast and appropriate step responses, which has been assessed by only stepping speed as an indicator of fall risk in older adults. We develop a new measurement system that applies a laser range finder for convenient assessment of stepping performance including temporal and spatial parameters such as reaction time, step velocity, step length, and accuracy. The measurement system for step tracking has a large advantage in terms of portability, cost, and the number of temporal and spatial parameters that we can measure. The aim of this study is to verify an efficacy of the measurement system for step tracking. We developed the system that applied a laser range finder for convenient assessment of stepping performance. In the test using a force platform and the developed measurement system simultaneously, based on reliability and validity, its effectiveness is confirmed.

Commentary by Dr. Valentin Fuster

Intelligent Transportation Systems

2012;():837-846. doi:10.1115/DSCC2012-MOVIC2012-8535.

Today’s driving patterns consume significant amount of useless energy, especially, when the fuel consumptions while braking, idling and re-accelerating at each traffic light are considered for millions of vehicles. This makes a high level management of driving profile crucial. In this paper, an analytical solution to the fuel consumption minimization problem with the existence of a single traffic light is investigated. The analytical solution is important for on-line implementation and sharing the information of the estimated fuel consumption of the road ahead with other vehicles. Pontryagin’s minimum principle is used to calculate the optimal velocity profile. Prior to the calculations, it is assumed that we have the knowledge of starting and ending points of the trip, the position and the operation sequence of the traffic light. In order to make the problem solvable, a simplified vehicle model is used. Furthermore, Willans approximation is utilized for fuel consumption calculations with addition of certain amount of idle speed fuel cost. The vehicle is forced to operate between a feasible torque and speed range. The optimization problem is simulated for an SUV vehicle first on a level road, then on a level road with the traffic light and finally on a road with grade. The results have shown that in addition to operating the vehicle close to its optimal point, it is possible to avoid the consumption of useless fuel due to the braking, idling and re-acceleration phases of a traffic light.

Commentary by Dr. Valentin Fuster
2012;():847-856. doi:10.1115/DSCC2012-MOVIC2012-8538.

The aim of the paper is to improve drivability of passenger cars under uncertain driving conditions by adding a proper rear steering strategy. This strategy is optimized in the frequency domain where the objectives concern lateral acceleration and yaw velocity. Due to the uncertainties the objectives are random variables. Therefore, mean values are minimized to obtain best mean behavior, and variances are minimized to achieve best robustness against change of car parameters. Comparison with a reference car shows improved results for both criterion values and probability distributions, and analysis of results in time domain validates the eligibility of the objectives.

Topics: Design , Automobiles , Wheels
Commentary by Dr. Valentin Fuster
2012;():857-866. doi:10.1115/DSCC2012-MOVIC2012-8603.

Hybrid electric vehicles offer many advantages over a traditional internal combustion engine powertrain in the area of ride handling and vehicle stability. An electric powertrain improves control system flexibility due an increased availably of measurement sensing and hardware actuation. It also avoids the limits imposed on vehicle handling by mechanical linkage connections alone. The purpose of the current research is to develop a rear-wheel steer assist control algorithm for vehicle stability under unsafe conditions. It is important that the driver is still part of the control, but is treated as a measurable disturbance. The rear-wheel stability control is an adaptive pole-placement approach utilizing an on-line recursive least squares algorithm with a variable forgetting factor to account for long drive times when vehicle stability is not in question.

Topics: Vehicles , Wheels
Commentary by Dr. Valentin Fuster
2012;():867-874. doi:10.1115/DSCC2012-MOVIC2012-8668.

This paper presents a predictive vehicle directional stability control structure that has integrated energy-loss reduction benefits during transient handling maneuvers. The method is based on the idea of balancing longitudinal and lateral tire force saturation levels using a cascade model predictive structure for the optimal distribution of tractive or braking torques. Balancing saturation levels also has the added benefit of reducing and evening-out tire wear. To demonstrate the energy-loss reduction benefits, we consider nonlinear simulations of a nominally unstable truck featuring an independent drive system. Comparisons against a commonly cited brake-based yaw stability control strategy with similar directional control performance shows that the proposed predictive saturation management approach provides energy-loss reductions of more than 60%. This energy efficiency benefits are retained whether or not the drive system has regenerative/energy recovery capabilities.

Commentary by Dr. Valentin Fuster
2012;():875-880. doi:10.1115/DSCC2012-MOVIC2012-8766.

Vehicle electrification is an increasingly popular design strategy for improving efficiency and reducing operating costs. This study uses an experimental and model-based approach to quickly and easily predict and optimize the efficiency of an electric bicycle system based on selection of critical parameters, including motor efficiency curves, rider behavior, mass, aerodynamics, and tire performance. The model is used to guide the construction of an electrified bicycle system to achieve the highest energy consumption performance among design alternatives. Tests of power consumption in actual usage show good agreement between the design predictions and measured performance.

Topics: Simulation , Design , Bicycles
Commentary by Dr. Valentin Fuster
2012;():881-887. doi:10.1115/DSCC2012-MOVIC2012-8858.

This paper presents a development of a robust steering controller that can be used for backing up maneuver of tractor-trailer vehicles. Tractor-trailer vehicles are naturally unstable when backing up maneuvers are performed. It is even more challenging for an inexperience driver to backing a trailer along a straight line since small errors in steering are amplified and the vehicle often departs from the desired path. Therefore it is desirable to develop a control system that can be used for such a scenario. In this paper, a robust controller is developed by using Lyapunov method based on a kinematic vehicle model for an application of tractor-trailer vehicles operating at a low speed in which state and input constraints are explicitly considered. The simulation results show the effectiveness of the proposed controller.

Commentary by Dr. Valentin Fuster

Learning Control

2012;():889-896. doi:10.1115/DSCC2012-MOVIC2012-8589.

This article presents a new identification method for discrete-time, single-input single-output, linear time-varying (LTV) plant utilizing the input and output data. This method applies specifically to plants that are able to repeat their trajectories. The identification learning law is determined through an optimization framework, which is similar in nature to the norm optimal design for Iterative Learning Control (ILC). Conditions for stability and convergence are obtained in the study. The design is verified using a numerical example.

Commentary by Dr. Valentin Fuster
2012;():897-905. doi:10.1115/DSCC2012-MOVIC2012-8676.

This paper presents the control system design and tracking performance of a large range single-axis nanopositioning system that is based on a moving magnet actuator and flexure bearing. While the physical system is designed to be free of friction and backlash, the nonlinearities in the electromagnetic actuator as well as the harmonic distortion in the drive amplifier restrict the achievable tracking performance for dynamic command following. It is shown that linear feedback proves to be inadequate due to limitations arising from the low open-loop bandwidth of the physical system. For periodic commands, like those used in scanning applications, the component of the tracking error due to the nonlinearities is deterministic and repeats every period. Therefore, an iterative learning controller (ILC) is designed and implemented in conjunction with linear feedback to reduce this periodic tracking error by more than three orders of magnitude. Experimental results demonstrate the effectiveness of this ILC in achieving 18nm RMS tracking error over 6mm range in response to a 2Hz band-limited triangular command. This corresponds to a dynamic range of 105.

Commentary by Dr. Valentin Fuster
2012;():907-915. doi:10.1115/DSCC2012-MOVIC2012-8721.

The appropriate choice of sensing and how to obtain the desired state information from available sensing for feedback or learning process are essential for most control schemes, including iterative learning control (ILC), to achieve their performance objective. In the multi-joint robots with joint elasticity, the load side joint space measurements are usually not available, even though the load side (end-effector) performance is of ultimate interest. This is termed as mismatched sensing problem. Furthermore, the mismatched uncertainty and mismatched real-time feedback signals in the robots with joint elasticity set further difficulty in achieving high performance. In this paper, a hybrid two-stage model based iterative learning control (ILC) scheme is proposed to deal with the mismatched dynamics. Also, to tackle the mismatched sensing issue, a sensor fusion scheme is developed. An optimization based inverse differential kinematics algorithm and decoupled adaptive kinematic Kalman filter (KKF) are integrated to obtain load side joint space information from the insufficient end-effector measurements. The proposed ILC scheme together with the load side state estimation algorithm is validated through the experimental study on a 6-DOF industrial robot.

Commentary by Dr. Valentin Fuster
2012;():917-925. doi:10.1115/DSCC2012-MOVIC2012-8726.

Learning feedforward control based on the available dynamic/kinematic system model and sensor information is generally effective for reducing the repeatable errors of a learned trajectory. For new trajectories, however, the system cannot benefit from previous learning data and it has to go through the learning process again to regain its performance. In industrial applications, this means production line has to stop for learning, and the overall productivity of the process is compromised. To solve this problem, this paper proposes a learning control scheme based on neural network (NN) prediction. Learning/training is performed for the neural networks for a set of trajectories in advance. Then the feedforward compensation torque for any trajectory in the set can be calculated according to the predicted error from multiple neural networks managed with expert logic. Experimental study on a 6-DOF industrial robot has shown the superior performance of the proposed NN based learning scheme in the position tracking as well as the residual vibration reduction, without any further learning or end-effector sensors during operation after completion learning/training of motion trajectories in advance.

Commentary by Dr. Valentin Fuster
2012;():927-932. doi:10.1115/DSCC2012-MOVIC2012-8834.

In this paper, a position regulation control strategy is developed for a platform mounted laser operating in the presence of a bounded, periodic disturbance. Specifically, the proposed control method utilizes an iterative learning estimator that is designed in order to promote system stability and asymptotically regulate the laser’s target position to a desired location. The learning estimator requires that the disturbance be bounded and its period to be known. Simulation and experimental results are provided to demonstrate the efficacy of the proposed method.

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