FocusNav: Spatial Selective Attention with Waypoint Guidance for Humanoid Local Navigation

2026-01-19

AI Summary

This work addresses the challenge of balancing long-range navigation objectives with immediate motion stability for humanoid robots operating in unstructured dynamic environments. To this end, the authors propose a spatial selective attention framework that dynamically reconciles task-oriented perception with urgent safety responses through a waypoint-guided spatial cross-attention mechanism (WGSCA) and a stability-aware selective gating module (SASG), enabling adaptive adjustment of perceptual range. By integrating collision-free waypoint prediction with environmental feature aggregation, the method significantly improves navigation success rates in complex dynamic scenarios on the Unitree G1 platform, outperforming existing baselines in both obstacle avoidance capability and motion stability.

Technology Category

• Intelligent Robots: Motion and Path Planning
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Natural Language Processing: Safety and Robustness

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Search and Retrieval-Augmented AI: Agentic search
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Robust local navigation in unstructured and dynamic environments remains a significant challenge for humanoid robots, requiring a delicate balance between long-range navigation targets and immediate motion stability. In this paper, we propose FocusNav, a spatial selective attention framework that adaptively modulates the robot's perceptual field based on navigational intent and real-time stability. FocusNav features a Waypoint-Guided Spatial Cross-Attention (WGSCA) mechanism that anchors environmental feature aggregation to a sequence of predicted collision-free waypoints, ensuring task-relevant perception along the planned trajectory. To enhance robustness in complex terrains, the Stability-Aware Selective Gating (SASG) module autonomously truncates distal information when detecting instability, compelling the policy to prioritize immediate foothold safety. Extensive experiments on the Unitree G1 humanoid robot demonstrate that FocusNav significantly improves navigation success rates in challenging scenarios, outperforming baselines in both collision avoidance and motion stability, achieving robust navigation in dynamic and complex environments.

Problem

• humanoid navigation
• local navigation
• dynamic environments
• motion stability
• unstructured environments

Innovation

• spatial selective attention
• waypoint-guided cross-attention
• stability-aware gating
• humanoid navigation
• local navigation