Event
RetroNav: A Training-Free Self-Evolving Navigation Framework
1 reports · 1 sources
Diantao Tu and other authors propose RetroNav, a training-free self-evolving navigation framework that integrates navigation, memory accumulation, and failure reflection. The authors claim that the framework achieves state-of-the-art performance on five navigation benchmarks. RetroNav uses a unified memory system to associate geometric and semantic scene knowledge with historical navigation experience, and distinguishes objective observations from task-dependent interpretations. After each navigation, an independent evaluator determines whether a retry is needed; when needed, a retrospective agent combines scene evidence to generate a retry briefing. The authors state that the method will be open-sourced.
Report timeline
- RetroNav: A Self-Evolving Lifelong Navigation Framework Combining Scene Retrospection and Retry
RetroNav is a training-free lifelong navigation framework that integrates navigation, memory accumulation, and failure reflection, achieving state-of-the-art performance on five navigation benchmarks. It uses a unified memory system to link geometric and semantic scene knowledge with historical navigation experience, distinguishing objective observations from task-dependent interpretations. After each navigation attempt, an independent evaluator determines whether a retry is needed; when it is, a retrospection agent uses scene evidence to generate a retry brief. The authors say the method will be open-sourced.
arXiv cs.RO ·