Explicit World Models for Reliable Human-Robot Collaboration

TL;DR

This paper proposes building and updating explicit world models in embodied AI to improve reliability and alignment with human expectations during social, multimodal interactions, addressing robustness challenges.

Contribution

It introduces a novel approach focusing on explicit world models to enhance reliability and interpretability in human-robot collaboration, diverging from traditional formal verification methods.

Findings

Explicit world models improve alignment with human expectations.

The approach enhances robustness in ambiguous and noisy environments.

It facilitates more comprehensible and predictable robot behaviors.

Abstract

This paper addresses the topic of robustness under sensing noise, ambiguous instructions, and human-robot interaction. We take a radically different tack to the issue of reliable embodied AI: instead of focusing on formal verification methods aimed at achieving model predictability and robustness, we emphasise the dynamic, ambiguous and subjective nature of human-robot interactions that requires embodied AI systems to perceive, interpret, and respond to human intentions in a manner that is consistent, comprehensible and aligned with human expectations. We argue that when embodied agents operate in human environments that are inherently social, multimodal, and fluid, reliability is contextually determined and only has meaning in relation to the goals and expectations of humans involved in the interaction. This calls for a fundamentally different approach to achieving reliable embodied AI that is centred on building and updating an accessible "explicit world model" representing the common ground between human and AI, that is used to align robot behaviours with human expectations.