Design of Automatic Steering Controller for Trajectory Tracking of Unmanned Vehicles Using Genetic Algorithms

In this paper, an efficient strategy is proposed to design the automatic steering controller for trajectory tracking of unmanned vehicles, which is robust with respect to the inherent nonlinearities and uncertainties of vehicles. The proposed automatic steering controller consists of a feedback part and a feedforward part. First, a fuzzy controller is proposed as the feedback part, and the parameters of membership functions and rules are optimized by genetic algorithms (GAs) to guarantee high performance. Then, a feedforward controller is designed to assist the controller when the vehicle is engaged in a curved section of trajectory, which utilizes preview information regarding upcoming curvature of reference trajectory to calculate a preview steering angle so that it offsets the disturbance of curvature. Both simulation and experimental results show that the proposed strategy can robustly track the reference trajectories under various conditions with high accuracy.