Agentic Self-Evolutionary Replanning for Embodied Navigation

TL;DR

This paper introduces SERP, a novel approach for embodied navigation that enables robots to self-evolve their action models during run-time, significantly improving success rates and efficiency in complex environments.

Contribution

SERP pioneers a paradigm shift by allowing models to evolve dynamically through in-context learning and graph-based replanning, unlike static model approaches.

Findings

Higher success rates in navigation tasks.

Lower token expenditure during replanning.

Robust performance across diverse environments.

Abstract

Failure is inevitable for embodied navigation in complex environments. To enhance the resilience, replanning (RP) is a viable option, where the robot is allowed to fail, but is capable of adjusting plan until success. However, existing RP approaches freeze the ego action model and miss the opportunities to explore better plans by upgrading the robot itself. To address this limitation, we propose Self-Evolutionary RePlanning, or SERP for short, which leads to a paradigm shift from frozen models towards evolving models by run-time learning from recent experiences. In contrast to existing model evolution approaches that often get stuck at predefined static parameters, we introduce agentic self-evolving action model that uses in-context learning with auto-differentiation (ILAD) for adaptive function adjustment and global parameter reset. To achieve token-efficient replanning for SERP, we also propose graph chain-of-thought (GCOT) replanning with large language model (LLM) inference over distilled graphs. Extensive simulation and real-world experiments demonstrate that SERP achieves higher success rate with lower token expenditure over various benchmarks, validating its superior robustness and efficiency across diverse environments.