Human-in-the-loop Learning and Control for Robot Teleoperation

Chapter Seven: Task learning of teleoperation robot systems

Abstract

This chapter presents several frameworks for task learning of teleoperation robots to explore the relationship between teleoperation robots, operators, and tasks. First, we propose a novel task learning framework that utilizes a teleoperation approach and GMM to encode the trajectory of the teleoperation robot end-effector. Based on this, a novel haptic myoelectric perception mechanism and a robot learning framework based on a hidden semi-Markov model coupled with Gaussian hybrid theory are proposed. Integration of the DTW method and the LWR method enabled the robots to learn the task trajectories and human muscle stiffness simultaneously after human demonstrations using a teleoperation ...