Development and Evaluation of Integrated Chassis Control Systems.

Abstract

Integrated chassis control (ICC) systems can be used to reduce the economic and social costs of road accidents. If these systems are to achieve their full potential for improved safety, however, two critical issues must be resolved: (i) the design of a controller integrating all sub-control systems, and (ii) rigorous evaluation to ensure their functionalities. A decentralized design that coordinates the commands from sub-chassis control systems is achieved under the current business practice, in which suppliers provide OEMs with proprietary controllers. For effective coordination of sub-control commands and for avoidance of liability, the coordination strategy of saturating sub-control commands is used. A coordinator based on a hybrid approach--an offline model predictive control and an online fixed-point control allocation method--is designed, which has superior computational efficiency and flexibility. The effectiveness of the decentralized ICC system is verified via commercial software, CarSim. The simulation results show that ICC can resolve conflicts among subsystems and achieve improved stability. Reconfiguration in the control, for dealing with actuator failure in sub-control systems and robust control under uncertainties is presented. For the evaluation of ICC, the worst-case scenario evaluation (WCSE) method is enhanced and applied to find the worst possible scenarios, for rigorous evaluation of vehicles, especially vehicles with chassis control systems. Two optimization methods (Sequential Quadratic Programming and Mesh Adaptive Direct Search) are used because of their convergence and computation efficiency. The worst allowable persistent bounded disturbance input generation method is applied to populate the initial points for the optimization problem. The effectiveness of the proposed WCSE method was shown through a rollover prevention case study.