Learning to Arbitrate Human and Robot Control using Disagreement between Sub-Policies

TL;DR

This paper introduces a reinforcement learning approach for optimal arbitration in teleoperation, dynamically balancing human and robot control at decision points to improve performance and safety.

Contribution

It presents a novel RL-based arbitration method that detects decision points via mixture distribution modality and adapts control blending accordingly.

Findings

Outperforms direct control in teleoperation tasks.

Enhances safety and accuracy in human-robot control.

Flexible control blending improves user performance.

Abstract

In the context of teleoperation, arbitration refers to deciding how to blend between human and autonomous robot commands. We present a reinforcement learning solution that learns an optimal arbitration strategy that allocates more control authority to the human when the robot comes across a decision point in the task. A decision point is where the robot encounters multiple options (sub-policies), such as having multiple paths to get around an obstacle or deciding between two candidate goals. By expressing each directional sub-policy as a von Mises distribution, we identify the decision points by observing the modality of the mixture distribution. Our reward function reasons on this modality and prioritizes to match its learned policy to either the user or the robot accordingly. We report teleoperation experiments on reach-and-grasping objects using a robot manipulator arm with different simulated human controllers. Results indicate that our shared control agent outperforms direct control and improves the teleoperation performance among different users. Using our reward term enables flexible blending between human and robot commands while maintaining safe and accurate teleoperation.