ASHA: Assistive Teleoperation via Human-in-the-Loop Reinforcement Learning

TL;DR

This paper introduces ASHA, a hierarchical reinforcement learning approach that efficiently learns assistive robot control from sparse user feedback and high-dimensional inputs like eye gaze, enabling rapid adaptation to individual users.

Contribution

The paper presents a novel hierarchical RL framework with offline pre-training and hindsight relabeling, improving learning efficiency from sparse feedback in assistive teleoperation tasks.

Findings

Learned to map gaze features to robot actions in under 10 minutes

Effectively adapted to different gaze strategies and environmental shifts

Achieved successful task completion with sparse rewards

Abstract

Building assistive interfaces for controlling robots through arbitrary, high-dimensional, noisy inputs (e.g., webcam images of eye gaze) can be challenging, especially when it involves inferring the user's desired action in the absence of a natural 'default' interface. Reinforcement learning from online user feedback on the system's performance presents a natural solution to this problem, and enables the interface to adapt to individual users. However, this approach tends to require a large amount of human-in-the-loop training data, especially when feedback is sparse. We propose a hierarchical solution that learns efficiently from sparse user feedback: we use offline pre-training to acquire a latent embedding space of useful, high-level robot behaviors, which, in turn, enables the system to focus on using online user feedback to learn a mapping from user inputs to desired high-level behaviors. The key insight is that access to a pre-trained policy enables the system to learn more from sparse rewards than a na\"ive RL algorithm: using the pre-trained policy, the system can make use of successful task executions to relabel, in hindsight, what the user actually meant to do during unsuccessful executions. We evaluate our method primarily through a user study with 12 participants who perform tasks in three simulated robotic manipulation domains using a webcam and their eye gaze: flipping light switches, opening a shelf door to reach objects inside, and rotating a valve. The results show that our method successfully learns to map 128-dimensional gaze features to 7-dimensional joint torques from sparse rewards in under 10 minutes of online training, and seamlessly helps users who employ different gaze strategies, while adapting to distributional shift in webcam inputs, tasks, and environments.