Team Xiaomi EV-AD VLA: Learning to Navigate Socially Through Proactive Risk Perception - Technical Report for IROS 2025 RoboSense Challenge Social Navigation Track

2025-10-09

AI Summary

This work addresses egocentric social navigation for mobile robots in dynamic indoor environments without global maps, relying solely on onboard RGB-D and odometry. The proposed method introduces an end-to-end reinforcement learning framework featuring an active risk awareness module that computes a collision risk score based on inter-agent distance and motion states, enabling proactive risk anticipation and socially compliant avoidance. It jointly optimizes navigation policy and risk estimation by integrating the Falcon vision architecture with a dedicated risk prediction network—without requiring prior maps or privileged information. The approach simultaneously ensures goal-directedness, safety (collision avoidance), and social norm compliance (personal space preservation). Evaluated on the Social-HM3D benchmark, the method ranks second among 16 participating teams, demonstrating significant improvements in real-time social navigation performance within dense, dynamic scenes.

Technology Category

• Natural Language Processing: Safety and Robustness
• Intelligent Robots: Localization, Mapping, and Navigation
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Economics, Online Markets and Human Computation: Social networks and social learning
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

In this report, we describe the technical details of our submission to the IROS 2025 RoboSense Challenge Social Navigation Track. This track focuses on developing RGBD-based perception and navigation systems that enable autonomous agents to navigate safely, efficiently, and socially compliantly in dynamic human-populated indoor environments. The challenge requires agents to operate from an egocentric perspective using only onboard sensors including RGB-D observations and odometry, without access to global maps or privileged information, while maintaining social norm compliance such as safe distances and collision avoidance. Building upon the Falcon model, we introduce a Proactive Risk Perception Module to enhance social navigation performance. Our approach augments Falcon with collision risk understanding that learns to predict distance-based collision risk scores for surrounding humans, which enables the agent to develop more robust spatial awareness and proactive collision avoidance behaviors. The evaluation on the Social-HM3D benchmark demonstrates that our method improves the agent's ability to maintain personal space compliance while navigating toward goals in crowded indoor scenes with dynamic human agents, achieving 2nd place among 16 participating teams in the challenge.

Problem

• Developing socially compliant navigation systems for human-populated indoor environments
• Enabling autonomous agents to navigate safely using only onboard RGB-D sensors
• Enhancing collision risk perception for proactive avoidance in crowded dynamic scenes

Innovation

• Proactive Risk Perception Module enhances social navigation
• Collision risk understanding predicts distance-based risk scores
• Augments Falcon model for robust spatial awareness