A trajectory tracking controller with dynamic gains for mobile robots

This paper presents a new fuzzy controller for trajectory tracking for unicycle-like mobile robots, represented by a Takagi-Sugeno fuzzy model. Such controller allows setting different gains for the controller, according to the values of the velocity of the trajectory and the values of the tracking error. The noise inherent to the robot odometry is treated by the fuzzy controller itself, not requiring the use of filters. The stability of the developed controller is proven based on the theory of Lyapunov. Experimental results are also presented, and show that the proposed controller is effectively able to guide the robot during a trajectory tracking task.