Deductive Chain-of-Thought Augmented Socially-aware Robot Navigation World Model

TL;DR

This paper introduces NaviWM, a novel socially-aware robot navigation model that combines large language models with a structured world model and logic-based reasoning to improve safety and social compliance in dynamic environments.

Contribution

NaviWM uniquely integrates a structured world model with LLMs using a deductive reasoning process, enhancing interpretability and safety in social robot navigation.

Findings

Improves success rates in crowded environments

Reduces social violations during navigation

Enhances interpretability of navigation decisions

Abstract

Social robot navigation increasingly relies on large language models for reasoning, path planning, and enabling movement in dynamic human spaces. However, relying solely on LLMs for planning often leads to unpredictable and unsafe behaviors, especially in dynamic human spaces, due to limited physical grounding and weak logical consistency. In this work, we introduce NaviWM, a socially-aware robot Navigation World Model that augments LLM reasoning with a structured world model and a logic-driven chain-of-thought process. NaviWM consists of two main components: (1) a spatial-temporal world model that captures the positions, velocities, and activities of agents in the environment, and (2) a deductive reasoning module that guides LLMs through a multi-step, logic-based inference process. This integration enables the robot to generate navigation decisions that are both socially compliant and physically safe, under well-defined constraints such as personal space, collision avoidance, and timing. Unlike previous methods based on prompting or fine-tuning, NaviWM encodes social norms as first-order logic, enabling interpretable and verifiable reasoning. Experiments show that NaviWM improves success rates and reduces social violations, particularly in crowded environments. These results demonstrate the benefit of combining formal reasoning with LLMs for robust social navigation. Additional experimental details and demo videos for this work can be found at: https://sites.google.com/view/NaviWM.