Influence-Based Reward Modulation for Implicit Communication in Human-Robot Interaction

TL;DR

This paper presents a novel influence-based reward modulation method using Transfer Entropy to enhance implicit communication in human-robot interaction, improving collaboration and social independence without explicit intention modeling.

Contribution

It introduces a new influence modulation approach integrated into reward functions for agents, validated through simulations and real-world experiments in social navigation and autonomous driving.

Findings

Boosting influence improves collaboration performance.

Resisting influence promotes social independence.

Method validated in both simulations and real-world tests.

Abstract

Communication is essential for successful interaction. In human-robot interaction, implicit communication holds the potential to enhance robots' understanding of human needs, emotions, and intentions. This paper introduces a method to foster implicit communication in HRI without explicitly modelling human intentions or relying on pre-existing knowledge. Leveraging Transfer Entropy, we modulate influence between agents in social interactions in scenarios involving either collaboration or competition. By integrating influence into agents' rewards within a partially observable Markov decision process, we demonstrate that boosting influence enhances collaboration and interaction, while resisting influence promotes social independence and diminishes performance in certain scenarios. Our findings are validated through simulations and real-world experiments with human participants in social navigation and autonomous driving settings.