Probabilistic Multimodal Modeling for Human-Robot Interaction Tasks

TL;DR

This paper introduces a reformulated Interaction Primitives approach that enhances multimodal human-robot interaction by enabling more accurate, robust, and faster inference from demonstrations, addressing nonlinearities in complex scenarios.

Contribution

The paper presents a novel reformulation of Interaction Primitives that improves inference efficiency and robustness in multimodal human-robot interaction tasks.

Findings

More accurate inference compared to standard methods

Enhanced robustness in challenging scenarios

Faster inference in multimodal HRI tasks

Abstract

Human-robot interaction benefits greatly from multimodal sensor inputs as they enable increased robustness and generalization accuracy. Despite this observation, few HRI methods are capable of efficiently performing inference for multimodal systems. In this work, we introduce a reformulation of Interaction Primitives which allows for learning from demonstration of interaction tasks, while also gracefully handling nonlinearities inherent to multimodal inference in such scenarios. We also empirically show that our method results in more accurate, more robust, and faster inference than standard Interaction Primitives and other common methods in challenging HRI scenarios.