Utilizing Human Feedback for Primitive Optimization in Wheelchair Tennis

TL;DR

This paper develops a human feedback-based method to refine primitive motions for an agile wheelchair tennis robot, enhancing safety and adaptability in high-speed manipulation tasks.

Contribution

It introduces an online primitive refinement process using human evaluative feedback within a probabilistic movement primitive framework for agile robots.

Findings

Safe execution of learned primitives demonstrated

Human feedback improves primitive adaptation

Enhanced agility and safety in robot tennis tasks

Abstract

Agile robotics presents a difficult challenge with robots moving at high speeds requiring precise and low-latency sensing and control. Creating agile motion that accomplishes the task at hand while being safe to execute is a key requirement for agile robots to gain human trust. This requires designing new approaches that are flexible and maintain knowledge over world constraints. In this paper, we consider the problem of building a flexible and adaptive controller for a challenging agile mobile manipulation task of hitting ground strokes on a wheelchair tennis robot. We propose and evaluate an extension to work done on learning striking behaviors using a probabilistic movement primitive (ProMP) framework by (1) demonstrating the safe execution of learned primitives on an agile mobile manipulator setup, and (2) proposing an online primitive refinement procedure that utilizes evaluative feedback from humans on the executed trajectories.