G2-Nav: Grounded and Guarded Vision-Language Costmaps for Robot Social Navigation

2026-07-18

AI Summary

This work addresses the challenge of achieving interpretable, safe, and socially aware robot navigation in complex real-world environments. The authors propose an autonomous navigation framework that integrates vision-language models to generate semantically interpretable cost maps through open-set perception, embeds social context understanding into trajectory planning, and incorporates a high-frequency safety verification mechanism to ensure operational reliability. By synergistically coupling multimodal semantic reasoning with motion planning, the approach demonstrates robust, efficient, and socially compliant navigation performance in unstructured real-world scenarios, effectively balancing explainability, safety, and adherence to human social norms.

Abstract

Social navigation requires the robot to reason and respond in complex real-world environments. While recent works attempt to incorporate human-level intelligence into robot planning using large Vision-Language Models (VLMs), end-to-end frameworks often create an unpredictable black-box, and existing instruction-following methods are not designed for full autonomy. To bridge this gap, we present G2-Nav, a novel framework that grounds abstract social reasoning and guards safe real-world deployment. Instead of asking the VLM for direct planning decisions, G2-Nav translates its semantic reasoning into a vision-language costmap with reliability and interpretability. The VLM evaluates traversable regions and social agents from open-set perception, mapping social context into the costmap. To improve real-world robustness, the VLM performs semantic verification on upstream tracking, and we introduce a high-frequency safety check to guard against system latency prior to trajectory generation. We demonstrate through real-world experiments that G2-Nav delivers safe, efficient, and socially compliant autonomous navigation in unstructured environments. Code will be made publicly available.

Problem

• social navigation
• vision-language models
• autonomous navigation
• costmap
• real-world robustness

Innovation

• Vision-Language Models
• Costmap
• Social Navigation
• Semantic Verification
• Safety Guard