Path tracking of eight in-wheel-driving autonomous vehicle: controller design and experimental results

This paper presents the design of an integrated path tracking controller that will drive an eight in-wheel-driving autonomous vehicle with rear-wheel-steering (8WD-RWS) through rapidly varying off-road terrain. This work treats the 8WD-RWS’s path tracking in a new manner, by integrating a control allocation algorithm and torque-vectoring control. Particular focus is paid on compound steering gain that significantly influence the obtained controllability and stability. A tracking control law is designed using the state-space equations of motion, for which global asymptotic stability is proven. It was implemented on a newly designed all-terrain vehicle, "Mission-Terminator", the NOVERI Racing Team’s entry in the Chinese Unmanned System Challenge 2019, a task-based autonomous off-road race. Experimental results from Mission-Terminator demonstrate the ability of the controller to track paths between obstacles, over rough dirt and wavy surface, and through steep slope. It shows a typical cross-track error of less than 0.5 m. In the National Challenge, Mission-Terminator had the fastest task completion time and highest tracking accuracy.