Perspective-Shifted Neuro-Symbolic World Models: A Framework for Socially-Aware Robot Navigation

2025-03-26

AI Summary

In dynamic social environments, robots must infer others’ latent beliefs and intentions for safe and socially compliant navigation—a challenge exacerbated by opaque belief representations in conventional POMDP-based approaches. Method: We propose an interpretable, robust neuro-symbolic world model. Our approach introduces a novel perspective-transformation operator for cross-agent belief estimation, unifying Theory of Mind (ToM) with Influence-Based Abstraction (IBA) to overcome the black-box limitation of traditional belief modeling. It integrates symbolic belief encoding, neural decoding, model-based reinforcement learning, and epistemic planning. Results: Evaluated on multi-agent interactive navigation benchmarks, our method achieves a 23.6% improvement in intention prediction accuracy and a 31.4% increase in rule-compliant avoidance rate. It supports real-time online belief updating and generates traceable, human-interpretable decision explanations—enabling transparent, adaptive, and socially aware robot navigation.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Machine Learning: Neuro-Symbolic Learning

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

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.

Problem

• Social navigation requires reasoning about hidden human beliefs
• Egocentric navigation lacks direct access to others' mental states
• Belief tracking in partially observable environments is challenging

Innovation

• Neuro-symbolic model-based reinforcement learning for navigation
• Perspective-shift operator for belief estimation
• Influence-based Abstractions in multi-agent settings