Adaptation of a Six-legged Walking Robot to Its Local Environment

Introduction

Walking in rough and unstructured terrain with a legged robot remains to be a challenging task, not only regarding the construction of a robust walking machine, but also looking at the control of such a complex system. Walking robots have the ability to climb over obstacles and can also walk fast in flat terrain. This is possible because the control parameters are adapted to the surrounding terrain. For example, the swing height has to correspond with the obstacle height, elsewise the robot cannot walk over these obstacles. An adaptation of the control parameters according to the environment can be realised in different ways. If there is no environment model present, the robot can only rely on its sensor systems trying to react to influences from the environment. Only the use of an environment model can enable the robot to interact more intelligently with the environment and adapt its parameters in advance. In this way, the robot can avoid many collisions, which speeds up the walking process. In the example, this would mean to increase the swing height and therefore modify the footpoint trajectory with the aim to overstep the obstacle. Because the swing height is only one parameter among many others, it is obvious how expedient an environment model is for a complex walking robot like LAURON.