Integrating the effect of driver to improve handling performance of vehicle

Abstract

Abstract: Considering the effect of human driver in the stability of vehicle has always been one of the most important nuances in vehicle literature. Yet, the lack of a standard framework for modeling the human behavior coerces the companies to mostly entrust to the drivers - as an unknown external input - for ultimate control tasks. This evinces that a key problem still to work out is a reliable control technique incorporating effect of human to serve driver’s request better. In this paper a simple model of human path following behavior is used that capture the most important parameters of a driver as well as the cumulative driver’s observation and the action delay. Integrating the driver model with a linear vehicle model, a linear parameter varying formulation is adopted. Assuming certain ranges for the parameters and their respective rate of changes, a controller robust to time delay is proposed and the corresponding disturbance attenuation in the sense of truncated l 2 norm is estimated. From an implementation point of view, the important advantage of the proposed controller is that only uses the IMU sensor and the vehicle velocity estimation, without requiring extra information about the desired path. The control action is based on the current vehicle states along with the estimation of driver status. In distributed wheel drive implementation, an optimal torque distributor transfers the torque to the wheels to achieve the best overall performance. The simulation results show that the proposed technique improves the vehicle overall performance compared to the case that the controller is designed separately.