LPV Torque Vectoring for an Electric Vehicle with Experimental Validation

In this paper, we propose a torque vectoring controller for an electric vehicle with two electric machines which drive the front wheels independently. The torque vectoring system includes a vehicle dynamics controller and a motor torque and wheel slip limiter. The vehicle dynamics controller is designed as a polytopic, linear parameter-varying (LPV) gain-scheduled controller. The LPV controller tracks the longitudinal velocity and the yaw rate of the vehicle. A torque and slip limiter (TSL) is developed to deal with physical saturation of the electric motors and wheel slip limitations which are related to the traction of the tires. The TSL deals with actuator limitations and prevents the wheel from spinning or blocking. Following the design, the controller is converted into fixed-point representation and is implemented on an automotivequalified microcontroller. As part of the European project eFuture, test drives are performed to obtain experimental data. For comparison the automotive prototype is driven with both, equal torque and the designed torque vectoring controller. The differences between the two approaches are illustrated with experimental results for a double lane change maneuver.