Search for: All records

In this paper, an adjustable autonomy framework is proposed for Human-Robot Collaboration (HRC) in which a robot uses a reinforcement learning mechanism guided by a human operator's rewards in an initially unknown workspace. Within the proposed framework, the robot can adjust its autonomy level in an HRC setting that is represented by a Markov Decision Process. A novel Q-learning mechanism with an integrated greedy approach is implemented for robot learning to capture the correct actions and the robot's mistakes for adjusting its autonomy level. The proposed HRC framework can adapt to changes in the workspace, and can adjust the autonomy level, provided consistent human operator's reward. The developed algorithm is applied to a realistic HRC setting, involving a Baxter humanoid robot. The experimental results confirm the capability of the developed framework to successfully adjust the robot's autonomy level in response to changes in the human operator's commands or the workspace.