Design of autonomous driving controls for multi-trailer articulated heavy vehicles

This article proposes a method for devising autonomous driving controls for multi-trailer articulated heavy vehicles (MTAHVs). This design is formulated as an optimization problem for improving ride quality and path-following performance. To implement the multi-objective design, a nonlinear model predictive control (NLMPC) technique is used to devise a tracking-controller for a MTAHV. For the NLMPC controller design, a nonlinear model is generated as the prediction model, and the respective TruckSim model is developed as the virtual plant. The weighting matrices of the NLMPC controller are chosen as the design variables, and a metaheuristic search algorithm is used to optimize these variables. By offline tuning these matrices automatically, the lateral-displacement error for the tractor decreases by 53%. Simulations demonstrate the reliability of the proposed design approach and the robustness of the NLMPC tracking-controller.