Vehicle Dynamic Control And Simulation Based On Lateral And Longitudinal Coupling

The paper is set in project of national natural science foundation (No.50475064) and project of Chongqing natural science foundation (CSTC,2006BA6017). In this paper, aimed at the control problem of the intelligent vehicle dynamic system, a lateral and longitudinal dynamic coupling vehicle control system is designed for improving the dynamic performance of the vehicle. A vehicle model is presented considering the coupling effects, and a coupled controller is designed based on the model. To keep the vehicle centered in the desired road curve and meanwhile keep the desired safe distance from the leading vehicle, simulation is conducted on the coupling vehicle control system to control the vehicle more precisely, enhance the vehicle driving safety and harmony, and offer a theoretical basis for the intelligent vehicle system. The main research works are presented as follows:Through the analysis of the vehicle dynamic system, three coupling effects including dynamic coupling, tire force coupling and weight shift coupling are identified and described. Then, a vehicle dynamic model combined the longitudinal and the lateral dynamic coupling effects is proposed taking the longitudinal velocity, lateral velocity, yaw rate, wheel speed and slip ratio as the states.Based on the control purpose of the vehicle control system, the spacing control strategy and preview control strategy are adopted after the discussion of the strategy for longitudinal and lateral control. Sliding mode control and Dynamic Surface Control methods are used to design a coupled controller, which compensate for the coupling effects.The coupled control system is compared against the decoupled control system through simulation; and is tested under different vehicle mass, roll coefficient and road curvature to identify its robust performance.At the end of the paper, since there is no lateral velocity measurement available, a lateral velocity estimation method is designed combined the multi-sensor information fusion and the vehicle dynamic model. By using the sensors'signals, the system state equation and measurement equations are implemented. The velocities are estimated by the Kalman Filter technology and used as controller inputs.