Robust Path Tracking Control for Autonomous Heavy Vehicles 2018-01-1082

With high maneuverability and heavy-duty load capacity, articulated steer vehicles (ASV) are widely used in construction, forestry and mining sectors. However, the steering process of ASV is much different from wheeled steer vehicles and tractor-trailer vehicles. Unsuitable steering control in path following could easily give rise to the “snaking” behaviour, which greatly reduces the safety and stability of ASV. In order to achieve precise control for ASV, a novel path tracking control method is proposed by virtual terrain field (VTF) method. A virtual U-shaped terrain field is assumed to exist along the reference path. The virtual terrain altitude depends on the lateral error, heading error, preview distance and road curvature. If the vehicle deviates from the reference line, it will be pulled back to the lowest position under the influence of additional lateral tire forces which are caused by the virtual banked road. By converting the lateral tire forces to the controlled steering wheel angle, the path following control based on VTF method can be realized. The influence of sideslip angle is considered during curved path tracking. The expected lateral error for ASV under the VTF method could be controlled by adjusting the road altitude coefficient. High coefficient could get more precise results but need high frequency moved steering wheel and shows aggressive operation, and low coefficient corresponds to leisurely operation. In order to evaluate the performance of the proposed VTF method, simulations of path tracking control for a nonlinear ASV model under annular road at different speed are conducted. Results show that even for high speed steering with the lateral acceleration of 4.0 m/s², the maximum lateral error and heading error are less than 0.2 meter and 1.0 degree, respectively.