Research on trajectory tracking of welding-cutting integrated robot based on fuzzy control

Abstract

To address the issues of low trajectory tracking accuracy of welding-cutting integrated robots in complex working environments and the difficulty of traditional PID control in adapting to system nonlinearities and strong disturbances, a two-dimensional Mamdani-type fuzzy controller is designed. The controller takes the deviation between the torch tip and the preset trajectory center, as well as the deviation change rate, as inputs, and outputs the speed adjustment amount for each joint motor. By establishing a fuzzy rule base and using the center of gravity method for defuzzification, precise control of the welding or cutting torch trajectory is achieved. Step signals and sinusoidal trajectories are respectively set in simulations for comparison with a PID controller. The results show that the fuzzy controller achieves a root mean square error and mean absolute error of 0.778 mm and 0.716 mm respectively in sinusoidal trajectory tracking, representing a reduction of approximately 65% compared to the PID controller; the step response has no overshoot, and the total variation of the control output is reduced by 54.7%. The proposed fuzzy controller does not require precise modeling and offers advantages such as smooth response, strong anti-interference ability, and smooth control, making it suitable for high-precision trajectory tracking control of welding-cutting integrated robots.