Learning Human-aware Robot Policies for Adaptive Assistance

Stony Brook University1 Peking University2
MY ALT TEXT

In our task scenario (illustrated by the feeding example), the robot's primary goal is to achieve the basic "feeding the food" task, represented by the Task Reward. However, as shown on the Left, the human user has more nuanced Preference Rewards, unknown to the robot, causing a misalignment between their reward functions. To address this, we propose a novel framework (Right) that goes beyond basic task fulfillment by introducing two modules: Motion Anticipation for predicting human motion and Utility Inference for estimating user preferences.

Abstract

Developing robots that can assist humans efficiently, safely, and adaptively is crucial for real-world applications such as healthcare. While previous work often assumes a centralized system for co-optimizing human-robot interactions, we argue that real-world scenarios are much more complicated, as humans have individual preferences regarding how tasks are performed. Robots typically lack direct access to these implicit preferences. However, to provide effective assistance, robots must still be able to recognize and adapt to the individual needs and preferences of different users. To address these challenges, we propose a novel framework in which robots infer human intentions and reason about human utilities through interaction. Our approach features two critical modules: the anticipation module is a motion predictor which captures spatial-temporal relationship between the robot agent and user agent, which contributes in predicting human behavior; the utility module infers the underlying human utility functions through progressive task demonstration sampling. Extensive experiments across various robot types and assistive tasks demonstrate that the proposed framework not only enhances task success and efficiency but also significantly improves user satisfaction, paving the way for more personalized and adaptive assistive robotic systems.

Framework

Video Presentation

Visualizations

BibTeX

@misc{qin2024learninghumanawarerobotpolicies,
        title={Learning Human-Aware Robot Policies for Adaptive Assistance}, 
        author={Jason Qin and Shikun Ban and Wentao Zhu and Yizhou Wang and Dimitris Samaras},
        year={2024},
        eprint={2412.11913},
        archivePrefix={arXiv},
        primaryClass={cs.RO},
        url={https://arxiv.org/abs/2412.11913}}