Text2Interaction: Establishing Safe and Preferable Human-Robot Interaction

TL;DR

Text2Interaction leverages large language models to enable zero-shot customization of robot behavior, ensuring safety and user satisfaction without extensive human feedback, outperforming baseline methods.

Contribution

The paper introduces a novel framework that uses language models to generate task plans and safety parameters, allowing zero-shot adaptation to user preferences in human-robot interaction.

Findings

83% of users agree it integrates their preferences

94% of users prefer it over baseline methods

Better alignment with unseen preferences in ablation studies

Abstract

Adjusting robot behavior to human preferences can require intensive human feedback, preventing quick adaptation to new users and changing circumstances. Moreover, current approaches typically treat user preferences as a reward, which requires a manual balance between task success and user satisfaction. To integrate new user preferences in a zero-shot manner, our proposed Text2Interaction framework invokes large language models to generate a task plan, motion preferences as Python code, and parameters of a safety controller. By maximizing the combined probability of task completion and user satisfaction instead of a weighted sum of rewards, we can reliably find plans that fulfill both requirements. We find that 83 % of users working with Text2Interaction agree that it integrates their preferences into the plan of the robot, and 94 % prefer Text2Interaction over the baseline. Our ablation study shows that Text2Interaction aligns better with unseen preferences than other baselines while maintaining a high success rate. Real-world demonstrations and code are made available at sites.google.com/view/text2interaction.