A robust GPI controller for trajectory tracking tasks in the limbs of a walking robot

Abstract

The study and development of walking robots implies solving several problems related with the design, dynamics, gait patterns, control and other topics inherent to their complex mechanism. One of the most important problems is the trajectory tracking tasks made by the degrees of freedom of each limb. Once the path planner has established the correct trajectory to follow, is necessary to track it with the least possible error. This important task is made by a controller, and helps the robot to maintain a stable walking. In this work we have focused on proposing a GPI control law for a limb of a hexapod walking robot, in order to track smooth trajectories and reject disturbances present in the closed-loop system. We also present the dynamic model of the limb, the designed reference trajectories, control simulations and experimental results.