The sliding mode controller with improved reaching law for harvesting robots

The chattering problem is caused by the switching function of sliding mode controller, which makes the accuracy of autonomous driving limited. To solve this problem, a steering sliding mode controller with improved nonlinear reaching law is proposed. The exponential function is used in the new reaching law to eliminate chattering problem, and the initial value of switching function is employed to avoid output saturation in the case of large initial errors. In the new reaching law, the large slope of sign function coefficient ensures the anti-disturbance ability of the controller after the system converges. A disturbance observer is constructed for real-time observation of the disturbance signals, which makes the controller get rid of the dependence on the prior knowledge of the disturbance. This paper introduces the tuning principle of controller parameters in detail and constructs the simulations and experiments of autonomous driving system. The experiments demonstrate that the proposed controller restrains the path-tracking mean error by 38% and restrains the standard error by 18%, the accuracy and efficiency of autonomous driving system are improved. Furthermore, the proposed reaching law can be used in other systems besides autonomous harvesting robots.