Perspective-Shifted Neuro-Symbolic World Models: A Framework for Socially-Aware Robot Navigation
This work addresses social navigation for robots in human-populated environments, presenting an incremental approach by combining existing methods like Theory of Mind and Influence-based Abstractions.
The paper tackles the problem of social robot navigation by addressing the challenge of belief tracking in partially observable environments, proposing a neuro-symbolic model-based reinforcement learning architecture and a perspective-shift operator for belief estimation.
Navigating in environments alongside humans requires agents to reason under uncertainty and account for the beliefs and intentions of those around them. Under a sequential decision-making framework, egocentric navigation can naturally be represented as a Markov Decision Process (MDP). However, social navigation additionally requires reasoning about the hidden beliefs of others, inherently leading to a Partially Observable Markov Decision Process (POMDP), where agents lack direct access to others' mental states. Inspired by Theory of Mind and Epistemic Planning, we propose (1) a neuro-symbolic model-based reinforcement learning architecture for social navigation, addressing the challenge of belief tracking in partially observable environments; and (2) a perspective-shift operator for belief estimation, leveraging recent work on Influence-based Abstractions (IBA) in structured multi-agent settings.