GÁSPÁR, Péter
Research Activity
The modeling of mechanical systems
The dynamics of mechanical systems is described by linear and nonlinear models. The approximation of nonlinear models with linear models is based on the quasi Linear Parameter Varying (quasi-LPV or LPV) description. The advantage of LPV models is that the entire operational interval of nonlinear systems can be defined and a well-developed linear system theory to analyze and design nonlinear system can be used.
The models are augmented with performance specifications and uncertainties. Weighting functions are applied to the performance signals to meet performance specifications and guarantee a tradeoff between performances. The uncertainties are modeled by both unmodelled dynamics and parametric uncertainties.
Gray-box identification methods are proposed for the estimation of unknown parameters in continuous-time nonlinear models. The advantage of the gray-box identification is that the knowledge about the model structure, the parameters and the components can also be used. In the gray-box identification an observer design is also proposed since this method is less sensitive to the initial conditions of the nonlinear modelling. This identification method is applied in several vehicle problems linked to modelling and control design.
References
[1] Iterative design of structured uncertainty model and robust controller based on closed-loop data, IET Control Theory & Applications, Vol. 4, No. 12, 2823-2836, 2010.
[2] System identification with Generalized Orthonormal Basis Functions: An Application to flexible structures, Control Engineering Practice, Vol. 11, No. 3, pp. 245-259, 2003.
Control design methods based on linear and nonlinear models
Several linear control design methods, i.e. Hinf/mu and H2/Hinf methods, are proposed for linear or linearized models. The Hinf/mu method guarantees robust stability and robust performance. The H2/Hinf method reduces the conservatism of the robust controller, however, robustness is guaranteed only for some of the performance demands.
A mechanical system is usually nonlinear in nature. Since the solution of the output-feedback nonlinear control problem usually results in highly nonlinear partial differential equations and a large number of theoretical and practical difficulties, it is difficult to solve it in practice. LPV methods, which are able to handle the nonlinear problems, are applied. The advantage of an LPV model based controller is that it meets robust stability and nominal performance demands in the entire operational interval, since it is able to adapt to the current operational conditions.
References
[1] Active suspension design using linear parameter varying control, International Journal of Vehicle Autonomous Systems, Vol. 1, No. 2, 206-221, 2003. ISSN 1471-0226.
[2] Design of robust controllers for active vehicle suspension using the mixed mu synthesis, Vehicle System Dynamics, Vol. 40, No. 4, 193-228, 2003.
Vehicle Oriented Research Activity
Components of vehicle control
The lateral and longitudinal dynamics are combined in a trajectory tracking system which is able to track road geometry with a predefined reference velocity. The control system manipulates front steering, braking and driving. Although the selection of the actuator is usually performed by using practical considerations, a theory-based method is developed in a weighting strategy.
References
[1] Reconfigurable control structure to prevent the rollover of heavy vehicles, Control Engineering Practice, Vol. 13, No. 6, 699-711, 2005
[2] The design of a combined control structure to prevent the rollover of heavy vehicles, European Journal of Control, No. 2, 2004
[3] Design of actuator interventions in the trajectory tracking for road vehicles, Conference on Decision and Control, Orlando, Florida, 2011.
[4] Integrated vehicle dynamics control via coordination of active front steering and rear braking, European Journal of Control, in print, 2011.
Variable-geometry suspension system
A method is developed for the design of the construction of a variable-geometry suspension system, which affects the design of the robust suspension control. In the control solution a predefined road trajectory required by the driver with a steering command is followed. The variable-geometry suspension system allows to modify the mechanical geometry and set up an efficient structure.
References
[1] Mechanical Analysis and Control Design of McPherson Suspension, International Journal of Vehicle Systems Modelling and Testing, Vol. 7, No. 2, pp. 173-193, 2012.
[2] Integration of Control Design and Variable Geometry Suspension Construction for Vehicle Stability Enhancement, Conference on Decision and Control, Orlando, Florida, 2011.
Suspension control for global chassis technology
The global chassis control is a cooperative control involving several vehicle subsystems. This control strategy must guarantee greater driving safety and comfort. The control of the suspension subsystem allows to improve vehicle road holding (safety) and passenger comfort but not at the same time. In order to achieve them in every driving situation, an adaptive two-degree-of-freedom controller for active suspensions is proposed. This control strategy ensures on the one hand robustness with respect to the performances, and on the other hand the trade-off between road holding and comfort. The proposed design is based on the LPV/Hinf theory.
References
[1] Attitude and Handling Improvements Through Gain-scheduled Suspensions and Brakes Control, Control Engineering Practice, Vol. 19, No. 3, 252-263, 2011.
[2] An LPV/Hinf Active Suspension Control for Global Chassis Technology:Design and Performance Analysis, Vehicle System Dynamics, Vol. 46, No. 10, 889-912, 2008.
[3] A New Semi-active Suspension Control Strategy Through LPV Technique, Control Engineering Practice, Vol. 16, 1519-1534, 2008.
Cooperative control of a vehicle platoon system
The focus of the research is the design of a vehicle platoon system. In the approach the platoon has a non-autonomous leader vehicle, i.e., this first vehicle is driven by a human driver, and some subordinate vehicles following the leader vehicle at some small pre-specified distances between them are driven automatically by driver assistance systems. Usually, two performance goals are set for a platoon system: a) to improve safety compared to that of the individual transports, b) to reduce the total fuel consumption. The longitudinal controller is responsible for the stabilization of the platoon, which is characterized formally by the string stability theorem.
In the control design the variation of slope along the route can also be taken into consideration. By choosing the velocity of the platoon fitting in with the inclinations of the road the number of unnecessary accelerations and brakings can be reduced. Since the velocity of the leader vehicle determines the velocity of all the members in the platoon, it must be selected as close as possible to the optimal velocities of the members.
References
[1] Analysis and experimental verification of faulty network modes in an autonomous vehicle string, 20th Mediterranean Conference on Control and Automation, Barcelona, 2012.
[2] Design and analysis of an automated heavy vehicle platoon, 9th Conference on Informatics in Control, Automation and Robotics, Rome, 2012.
[3] LPV-based control design of vehicle platoon considering road inclinations, IFAC World Congress, Milan, 2011.
|
Integrated control systems for vehicles
Several individual active control mechanisms are applied in road vehicles to solve different control problems. In the operation of autonomous control systems interactions or conflicts may occur. These difficulties can be handled in such a way that the effects of a control system on other vehicle functions are taken into consideration in the design process.
Research into the functional integration possibilities of active vehicle components is aimed at integrating the suspension, steering, braking and driveline systems in order to guarantee performance demands, achieve better management of different resources and eliminate harmful interference between them. In the design of integrated systems several difficulties must be handled, e.g. uncertainties caused by physical or communication components. Using integrated control the operation of the partial control system is also extended with a fault-tolerant and reconfigurable structure.
References
[1] LPV design of fault-tolerant control for road vehicles, International Journal of Applied Mathematics & Computer Science , Vol. 22, No. 1. 2012.
[2] Design of Integrated Vehicle Chassis Control Based on LPV Methods, Control of Linear Parameter Varying Systems with Applications, Eds: C. Scherer, J. Mohammadpour, Springer, 2012.
[3] Active Suspension in Integrated Vehicle Control, in "Switched Systems", Ed. Janusz Kleban, ISBN 978-953-307-018-6, IN-TECH Education and Publishing, 1-22, 2009.
Supervisory decentralized architecture
The research also focuses on a multi-layer supervisory decentralized architecture. The performance specifications are guaranteed by the local controllers, while the coordination of these components is provided by the supervisor. The supervisor has information about the current operational mode of the vehicle and it is able to make decisions about the necessary interventions into the vehicle components. These decisions are propagated to the lower layers through predefined interfaces encoded as suitable scheduling signals. Local control components are designed by LPV (Linear Parameter Varying) methods by taking into consideration the additional scheduling variables received from the supervisor. These methods handle trade-off between the different conflicting performance demands and integrate control methods and the validity of the theoretical methods are analyzed in several case studies.
References
[1] Design of a two-level controller for an active suspension system, Asian Journal of Control, Vol. 15, No. 3, 664-678, 2012.
[2] LPV design of reconfigurable and integrated control for road vehicles, Conference on Decision and Control, Orlando, Florida, 2011.
[3] Design of a supervisory integrated control for driver assistance systems, Conference on Decision and Control, Maui, Hawai, US, 2012.
Industrial Activity
Automatic visual detection system
In this project, an automatic visual detection system is developed to prevent unintended lane-departures. This system is important because a large number of road accidents occur due to lapses in the driver's attention. The detection system contains several correlated parts. The vehicle is equipped with digital cameras for monitoring lane geometry in real-time and detecting obstacles in front of the vehicle. The vehicle's motion is predicted by using velocity and steering angle signals from on-board sensors. Decision logic uses both the detected lane geometry and the predicted motion of the vehicle. The supervisor can intervene in the vehicle’s motion by using the so-called drive stability control system (DSC) if there is no driver’s activity (steering, accelerating, or breaking) and the duration of the lane departure is less than the reaction time of the driver. The intervention is performed by using the brake system.
References
[1] Visual lane and obstacles detection system for commercial vehicles, Safeprocess, Budapest, Hungary, 2000.
[2] Automatic Detection of Lane Departure of Vehicles, Proc. of the 8th IFAC Symp. on Transportation Systems, Chania, pp. 1096-1101, 1997.
Fleet management system
The main objectives of the project are the elaboration of the theory and methods of intelligent supervision, control and communication systems installed on the vehicles and an associated information service system for fleet management. The development of the information service as part of the fleet management system that collects, analyses and evaluates databases sent by the individual vehicles extends the utility of the installed vehicle systems and represents a significant added value to its application. This system integrates the capabilities of recent mobile telecommunication and the sensory and data measurement systems on the vehicle into a unified framework. The system can also supervise and control the transportation processes, can assist the driver in decision making, support the coordination of transport activities and provides an information database for all participants possessing the particular unit on the vehicle.
State-of-Charge recalibration of lead-acid batteries
The aim of the research is to analyze the problem of State-of-Charge (SOC) recalibration during vehicle key-off mode from battery current, voltage and temperature measurements. A quiescent voltage battery model is formalized and two different observers are proposed, i.e., a Multiple Model Adaptive Estimator (MMAE) and an Unscented Kalman Filter (UKF). Both approaches are capable of robust quiescent voltage estimation in a computationally inexpensive way. The performance of the observers is numerically quantified by exhaustive simulations within application relevant boundary conditions.
References
[1] State-of-Charge recalibration in automotive applications, Proceedings of the Institution of Mechanical Engineers, Part D, Journal of Automobile Engineering, Vol. 226, pp. 1-10. IF: 0.441, 2012.
[2] Quiescent voltage determination of lead-acid batteries in automotive applications, 12th European Lead Battery Conference, Istanbul, Turkey, 2010.
Observer-based brake control for railways
Since in a braking operation the shortest possible brake distance is required at all times an efficient and robust slip prevention control must be developed. A control strategy based on an estimation method for the actual wheel–rail friction coefficient is proposed. A logic-based scheme that estimates a set point that prevents wheel slip is proposed. Having this estimation a conventional control algorithm maintains the system at the prescribed set point. If the external environment changes a new set point corresponding to the current condition is estimated. The estimation method is based on an adaptive observer design. The proposed control procedure does not rely on measured values of the slip ratio.
References
[1] Discussion on: "Combining Slip and Deceleration Control for Brake-by-Wire Control Systems: a Sliding-Mode Approach", European Journal of Control, 2007.
[2] Observer-Based Brake Control for Railways, Nonlinear Control Systems, Pretoria, South Africa, 2007.
Control of a pressurizer at the NPP
A pressure controlling tank located in the primary circuit of a Nuclear Power Plant is designed. All steps from modeling through control design to implementation are carried out. Based on first engineering principles a Wiener model is formulated and its unknown parameters are identified. The control design is based on a dynamic inversion method. The performance of the designed closed-loop system is tuned by an error feedback. The implemented controller is of a distributed structure including measurement and control PLCs, a continuous power controller and a special supervisor module. The nominal stability of the controller in the networked environment is analyzed by using the maximum allowable transmit interval. The hardware and software design and implementation meet the safety-critical requirements imposed by the special nature of the plant.
References
[1] Identification and Dynamic Inversion-Based Control of a Pressurizer at the Paks NPP, Control Engineering Practice, Vol. 18, No. 5, pp. 554-565, 2010.
[2] Networked control approaches for a nuclear power plant pressurizer subsystem, IFAC Workshop on Estimation and Control of Networked Systems, Venice, Italy, 2009.
[3] Tracking Design for Wiener Systems Based on Dynamic Inversion, Conference CCA/CACSD/ISIC, Munich, 2006.
|
