Path Tracking of a Four-Wheel Independently Driven Skid Steer Robotic Vehicle Through a Cascaded NTSM-PID Control Method

Abstract

Four-wheel independently driven skid steer robotic vehicles (IDSSRVs) have been widely applied in many fields. However, limited by small size and low power consumption, many sensors and high-performance processors cannot be equipped into four-wheel IDSSRVs, which makes path tracking more complex. The nonsingular terminal sliding mode (NTSM) is a good choice for small robotic vehicles because of its strong robustness, fast convergence, and low complexity. However, the performance of the NTSM is easily affected by vehicular weight and road friction because a four-wheel IDSSRV has no steering devices and its driving behavior is completed by adjusting the velocity of the four wheels based on the wheel–ground interaction. Furthermore, a minor instability caused by the output difference of the motors can bring extra torque on the vehicle and ultimately lead to a tracking deviation for the NTSM controller. To resolve these problems, a dynamic model of a four-wheel IDSSRV is analyzed first. Subsequently, a cascaded NTSM-proportional–integral–derivative (PID) method is proposed, in which a conversion function based on compensating the sideslip angle is raised to improve the NTSM controller performance of antidisturbance and tracking accuracy with the limitations of the sensors and processors, and a joint PID controller simultaneously optimizing the motor output error and vehicular yaw error is designed to eliminate the imbalance of motor output and to reduce the deviation of the NTSM controller. Finally, comparative experiments are conducted, and the results highlight the benefits of the proposed approach by reducing the tracking error by approximately 80% compared with the dynamic model predictive control (MPC) method and reducing the tracking error by over 30% compared with the pure NTSM method.