A Telemanipulation-Based Human–Robot Collaboration Method to Teach Aerospace Masking Skills

Abstract

Traditional offline programming or teach pendant-based methods limit the collaboration capabilities of users and robots to cope with complex and changing industrial tasks. To address efficient robot manipulation in high-mix and low-volume tasks, especially for skillful tasks involving both trajectory and force control requirements, fast robot teaching and skill transferability are critical. Compared to manually dragging a heavy robot, or programming a trajectory by complex calculations, we believe that robots can efficiently learn skills from direct teaching through telemanipulation, and improve the skills based on optimization with sensory feedback. In aerospace engine, maintenance, repair, and operations, the surfaces of aerospace components are required to be masked by tapes. We propose a fast and intuitive telemanipulation-based method to teach a robot these masking skills and compare the performance of the proposed method with teach pendant-based methods among several users. This study aims to prove the efficiency and intuitiveness of the telemanipulation-based method proposed herein for enabling a robot to learn skillful and complex manipulation tasks.