Dynamic identification and control of a nonholonomic mobile robot

Abstract

Identifies dynamic equations, for a mobile robot with nonholonomic constraints, and uses them to implement a computed torque control law. The model is linear in relation to a minimum set of dynamical parameters which have to be identified. These parameters are identified using least squares techniques as this is commonly done for robot manipulators. Using the identification dynamic model the authors then implement a computed torque control law in the joint space in order to track a desired cartesian trajectory. The results are compared with a classical PID control law.