Guided locomotion in 3D for snake robots based on contact force optimization

Abstract

The versatility of snake robots has led to their use in a wide variety of settings, including archaeological sites, natural disaster zones, and nuclear power plants. Currently, snake robots locomote through these rugged environments using repeatable pre-programmed motions, often with underwhelming performance. This paper presents the novel design of a control architecture that addresses the limitations of preprogrammed motions by using contact information from the environment. The controller uses contact force optimization, a concept taken from the field of manipulation, in combination with gain scheduling, to perturb existing gait controllers to perform better in three-dimensional environments. The efficacy of the controller is tested in a simulation of a snake robot on rugged terrain.