Entity · paper

Learning Adaptive Safety Margins for Visual Navigation

paperactiveprovisionallearning-adaptive-safety-margins-for-visual-navigation-1a4bd4cc·1 events·first seen 6d ago

Aliases: Learning Adaptive Safety Margins for Visual Navigation

Co-occurring entities

Control-Barrier-Function MP3D HM3D Unitree G1

More like this (12)

Self-Refining Agentic Reinforcement Learning for Vision-Conditioned UAV Navigation EvolveNav: Proactive Preflection and Self-Evolving Memory for Zero-Shot Object Goal Navigation Visual Verification Enables Inference-time Steering and Autonomous Policy Improvement Visually Grounded Self-Reflection for Vision-Language Models via Reinforcement Learning InSight: Self-Guided Skill Acquisition via Steerable VLAs InSight: Self-Guided Skill Acquisition via Steerable VLAs TempoVLA: Learning Speed-Controllable Vision-Language-Action Policies Safety Usage Dashboard Paved with True Intents: Intent-Aware Training Improves LLM Safety Classification Across Training Regimes Where Does the Answer Come From? Benchmarking View-Level Visual Evidence Identification in Multi-View MLLMs for Autonomous Driving Neural Safety Filters Harmonizing AI Safety Thresholds

Recent events (1)

4arXiv · cs.AI·6d ago·source ↗

Adaptive safety margins for visual robot navigation via context-conditioned safety critic

A new arXiv paper proposes a context-conditioned safety critic that learns adaptive clearance preferences for ranking diffusion-based trajectory proposals in indoor robot navigation. The critic decomposes into safety, efficiency, and distance-constraint matching terms, trained with privileged ESDF geometry in simulation and distilled into a perception-only selector via teacher-student learning. The method achieves top success rate and SPL on PointGoal navigation benchmarks (HM3D, MP3D) and transfers zero-shot to a Unitree G1 humanoid robot without task-specific tuning.

AI Safety Research Control-Barrier-Function Learning Adaptive Safety Margins for Visual Navigation MP3D +2 more