Robust control of farm tractors on variable sloped terrains based on sliding-mode observers

This paper considers the problem of motion control of farm tractors on sloped terrains. In reality, farm tractors often operate on sloped agricultural fields, especially in China's hilly north-eastern provinces. In such scenarios, many tractors using GNSS navigation systems have shown deterioration in control accuracy. Existing researches mostly studied the control problem of tractor motion under the assumption of flat surface and often with a kinematic model. However, when the lateral gravitational force kicks in, these assumptions become no longer valid. It is necessary to deal with the additional lateral force at force-level, and to achieve this. The main contribution of this paper is to establish a dynamic tractor model on sloped terrains and develop a robust control framework based on sliding-mode observers. The lateral gravitational force is treated as a slowly varying disturbance at the vehicle's lateral direction. Then a sliding-mode observer is designed to reconstruct the lateral disturbances. Finally, the H∞ robust controller based on linear matrix inequality (LMI) approach is designed to attenuate the influence from unknown sloped terrains and improve the lateral offset of the tractor motion. Simulation results verified the effectiveness of the proposed method and showed superior performance than LQR controllers.