Improving the Navigability of a Hexapod Robot Using a Fault-Tolerant Adaptive Gait

Abstract

This paper encompasses a study on the development of a walking gait for fault tolerant locomotion in unstructured environments. The fault tolerant gait for adaptive locomotion fulfills stability conditions in opposition to a fault (locked joints or sensor failure) event preventing a robot to realize stable locomotion over uneven terrains. To accomplish this feat, a fault tolerant gait based on force-position control is proposed in this paper for a hexapod robot to enable stable walking with a joint failure. Furthermore, we extend our proposed fault detection and diagnosis (FDD) method to deal with the critical failure of the angular rate sensors responsible for the attitude control of the robot over uneven terrains. A performance analysis of straight-line walking is carried out which shows that the proposed FDD-based gait is capable of generating an adaptive walking pattern during joint or sensor failures. The performance of the proposed control is established using dynamic simulations and real-world experiments on a prototype hexapod robot.