EvolveNav: Proactive Preflection and Self-Evolving Memory for Zero-Shot Object Goal Navigation
2026-06-16 • Artificial Intelligence
Artificial Intelligence
AI summaryⓘ
The authors focus on teaching robots to find objects they've never seen before without prior training, a task called Zero-Shot Object-Goal Navigation. They created a system that learns from its past experiences during testing to get better over time. This system uses a memory of previous actions and smartly picks rules to guide the robot, helping it avoid repeated mistakes and unnecessary searching. Their approach led to better success rates and more efficient exploration compared to earlier methods.
Zero-Shot LearningObject-Goal NavigationEmbodied AgentsFoundation ModelsTest-Time AdaptationRule MemoryUpper Confidence BoundPreflection ModuleExploration Efficiency
Authors
Qi Chai, Wenhao Shen, Nanjie Yao, Yue Xia, Kaiyong Zhao, Jie Ma, Guosheng Lin, Hao Wang
Abstract
Zero-Shot Object-Goal Navigation (ZS-OGN) requires embodied agents to explore and locate target objects without any prior training. To this end, recent methods leverage foundation models. But they typically rely on static priors and lack adaptation, which leads to repeated errors and costly trial and error. In this paper, we propose a self-evolving ZS-OGN framework that enables continuous test-time improvement. Specifically, we build an agentic rule memory by extracting actionable knowledge from past trajectories. Then, we propose a retrieval strategy based on upper confidence bound, selecting effective rules by balancing semantic relevance and historical success. In addition, we introduce a memory-guided preflection module that forecasts potential outcomes before action, reducing inefficient exploration. Extensive experiments show that our method outperforms existing zero-shot baselines, achieving a 10.1\% improvement in success rate with fewer unnecessary steps.