Controle adaptativo auto-ajustável para controle de trajetórias de um robô móvel com rodas

Abstract

This work presents the implementation and experimental tests of an adaptive control strategy for tracking path of wheeled mobile robots (WMR) with differential drive and nonholonomic constraints. The proposed control system consists of two control loops cascaded, which the first is a PID controller responsible for the speed control of the drive wheels, and the second is a nonlinear controller which has the purpose of allowing the robot tracking a reference trajectory. To ensure the asymptotic convergence of the robot even when subject to uncertainty, it was proposed to apply a nonlinear adaptive controller with time varying parameters. This kind of controller allows the robot to adjust to different reference trajectories, and overcome problems related to disturbances and uncertainties in system parameters. In this work, the robot kinematic model was used to estimate the robot posture and to calculate the error between the real and the desired trajectory. Results of computational simulations and experimental tests carried out on a real mobile robot show advantages and disadvantages of the proposed adaptive controller.