Constant Force Self-Tuning Control of Bathing Assistance Robots Based on GA-Fuzzy-PID Algorithm

This study addresses the issue of constant force control for a bathing assistance robot's end-effector during direct contact with human skin. A novel constant force control algorithm for the robot is proposed. Initially, based on establishing a contact model between the robot's end-effector cleaning tool and the skin, a GA-Fuzzy-PID controller was designed. Genetic algorithms were employed for iterative optimization to find appropriate PID parameters; subsequently, fuzzy control was utilized to adjust the PID coefficients in real-time. Finally, the control system was simulated using MATLAB/Simulink, comparing the effects of four control algorithms on the control of the constant force system during cleaning, and both constant and variable force tracking experiments were conducted. Simulation results demonstrated that compared to traditional PID, GA-PID and Fuzzy PID, the GA-Fuzzy-PID significantly improved the accuracy and robustness of constant force control, reducing peak error to 0.03 and decreasing overshoot by 41.97%, 42.09% and 10.74<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup>, respectively.