Autonomous Pose Detection and Alignment of Suction Modules of a Biped Wall-Climbing Robot

摘要

Vacuum adsorption is a simple but effective attaching method widely used in many fields including robotic wall climbing. It is required that the sucker is aligned well with the target surface to form airtight chamber for vacuum generation. For applications in biped wall-climbing robots, automatically aligning the sucker with the wall is beneficial and important to enhance the efficiency and effectiveness of vacuum adsorption. Especially, such a function is essential for autonomous intelligent climbing. To this end, we propose a novel and low-cost approach to perform autonomous alignment of a sucker (suction module) based on noncontact sensors. We first develop a sensing system to detect the configuration of the swinging suction module with respect to the target surface, and then present an algorithm to compute the configuration transformation and control the robot to drive the suction cups toward the target surface with a well-aligned configuration for adherence. In this paper, the basic theory for autonomous pose detection and alignment of the suction module for wall climbing with a biped robot is presented. Specifically, the key configuration of the swinging suction module for adsorption is analyzed, and the pose detection model, the conditions for forming airtight chamber, and the autonomous alignment algorithm are introduced. Calibration of the sensing system and experiments with our biped wall-climbing robot W-Climbot is conducted. The results have verified the feasibility, effectiveness and applicability of the proposed sensing system, theoretical analysis and the algorithm for autonomous pose detection and alignment of the suction modules.