Simulation Research On Vehicle Handling Inverse Dynamics

A brief review of the developing history of vehicle handling inverse dynamics is introduced in this paper. Using radial basis function neural networks, The simulation research on vehicle handling inverse dynamics with open-loop system and driver-vehicle with closed-loop system are analyzed. The main purposes of this paper are:1. Using radial basis function neural networks, the mapping relation between lateral angular velocity and steering angle input is found. The identification results of step and sine steering angle input showed that the proposed identification method is not only practical, but also with high accuracy.2. Considered nonlinear yaw force, a two degree of freedom nonlinear vehicle model with steering angle input is introduced with Fiala tire model. Using radial basis function neural networks, the nonlinear mapping relation between lateral angular velocity and steering angle input is found with different frictional coefficient.3. According to the simulation results of a two-degree-freedom closed-loop system and the test of single lane change maneuver, the nonlinear mapping relationship between vehicle yaw velocity, lateral acceleration and steering angle can be found. The identification results show that the method is not only feasible, but also with high accuracy.In conclusion, based on Radial Basis Function neural networks, a method of identifying vehicle steering angle is proposed for further investigation of the vehicle handling inverse dynamics. The identification results show that the method is not only feasible, but also with high accuracy, little computation requirement and good stability. Meanwhile, it is also useful for automatic pilot design.