Dexterous obstacle-avoidance motion control of Rope Driven Snake Manipulator based on the bionic path following

Abstract

The RDSM (Rope Driven Snake Manipulator) is a hyper-redundant robot operating in narrow and confined environments, such as aerospace, nuclear industry and etc. Inspired from the bionic following motion, an obstacle-avoidance control method of the RDSM in uncharted environments is proposed without the information of global obstacles. First, the kinematic model of a 17-DOF RDSM is established considering the mapping among the joint angle, rope length and end posture. Then, the motion path and inverse kinematic iterative algorithm of the RDSM in multi-obstacle environments is deduced using the piecewise Bézier curves. The motion control method of the RDSM is designed based on the model reference adaptive control, guaranteeing the joint error and end posture error asymptotically stable. Finally, the proposed obstacle avoidance method is verified with the simulation and experiment. As the result shown, the RDSM avoids obstacles on a path with a minimum curvature radius of 1/8 arm length, and the motion control precision of the end posture is within ±2.5mm/±0.5∘.