Learning Personalized Human-Aware Robot Navigation Using Virtual Reality Demonstrations from a User Study

TL;DR

This paper introduces a reinforcement learning framework that personalizes human-aware robot navigation through virtual reality demonstrations, leading to more comfortable human-robot interactions validated by a user study.

Contribution

It presents a novel personalized navigation controller trained with VR demonstrations, outperforming classical methods and generalizing well to new states.

Findings

Personalized approach outperforms classical methods in user comfort

Learned controller generalizes to unseen states

Successful transfer of controller to real robot

Abstract

For the most comfortable, human-aware robot navigation, subjective user preferences need to be taken into account. This paper presents a novel reinforcement learning framework to train a personalized navigation controller along with an intuitive virtual reality demonstration interface. The conducted user study provides evidence that our personalized approach significantly outperforms classical approaches with more comfortable human-robot experiences. We achieve these results using only a few demonstration trajectories from non-expert users, who predominantly appreciate the intuitive demonstration setup. As we show in the experiments, the learned controller generalizes well to states not covered in the demonstration data, while still reflecting user preferences during navigation. Finally, we transfer the navigation controller without loss in performance to a real robot.