Trajectory Tracking Control Strategy Based on Model-Free Adaptive Sliding Mode Control Approach for Robot Manipulator

This paper proposes the trajectory tracking control strategy based on the model-free adaptive sliding mode control (MFAC-SMC) approach for a robot manipulator. The MFAC-SMC controller is designed based on partial form dynamic linearization (PFDL), model-free adaptive control, and discrete sliding mode control methods. Afterwards, the stability of MFAC-SMC is proven, and its superiority in robotic arm trajectory tracking control is demonstrated by comparing it with other control methods. To improve the trajectory tracking performance of the robotic manipulator, the control strategy of MFAC-SMC controller parameters is explored, which provides possibilities to find the optimal parameter configuration combination under a given torque constraint. From parameter experiment results, the values of control system response, overshoot, and steady-state error are reduced with the increase of values of exponential reaching law coefficient, and the decrease of values of control input variation penalty factor, but the control input torque is raised. Finally, two cases are designed to verify the correctness of the parameter effect patterns. A cosine trajectory scenario is designed to demonstrate the optimal control strategy of parameters combination under a torque constraint and a quintic polynomial trajectory scenario is supplemented to verify the effectiveness of the parameter patterns in different scenarios. The results show that the proposed control strategy obtains the optimal trajectory tracking performances including the overshoot, rise time, and steady-state error.