Research on Motion Control of Wheel-Legged Robot Based on ASMC and LQR

Abstract

Wheel-legged robots combine the high-speed movement of wheeled robots with the superior obstacle-navigation abilities of legged robots. To enhance the robot's resilience against disturbances and improve its ability to overcome obstacles, we developed a control scheme that decouples the control into two parts: posture control and balance control. First, the robot's dynamics were modeled using the Euler-Lagrange method. To manage external disturbances, an Adaptive Sliding Mode Controller (ASMC) was designed to handle posture control. For balance control, a Linear Quadratic Regulator (LQR) was implemented to solve the Riccati equation in real-time, generating the necessary control torques to stabilize the robot. By combining these two control strategies, we present a framework that significantly improves the robot's mobility. Furthermore, tests conducted on a physical prototype demonstrated the effectiveness of the algorithm and the robot's disturbance rejection capability. The results confirm that the control scheme, which integrates ASMC and LQR, allows the wheel-legged robot to execute complex maneuvers and traverse uneven terrain with ease.