E2Map: Experience-and-Emotion Map for Self-Reflective Robot Navigation with Language Models

TL;DR

E2Map enhances robot navigation by integrating language model knowledge with real-world experiences, allowing for adaptive behavior in unpredictable environments, demonstrated through improved performance in stochastic navigation tasks.

Contribution

Introduces E2Map, a novel framework combining LLM knowledge and experiential data for self-reflective, adaptive robot navigation in stochastic environments.

Findings

Significantly improves navigation success in stochastic environments.

Enables one-shot behavior adjustment based on real experiences.

Validated in both simulation and real-world scenarios.

Abstract

Large language models (LLMs) have shown significant potential in guiding embodied agents to execute language instructions across a range of tasks, including robotic manipulation and navigation. However, existing methods are primarily designed for static environments and do not leverage the agent's own experiences to refine its initial plans. Given that real-world environments are inherently stochastic, initial plans based solely on LLMs' general knowledge may fail to achieve their objectives, unlike in static scenarios. To address this limitation, this study introduces the Experience-and-Emotion Map (E2Map), which integrates not only LLM knowledge but also the agent's real-world experiences, drawing inspiration from human emotional responses. The proposed methodology enables one-shot behavior adjustments by updating the E2Map based on the agent's experiences. Our evaluation in stochastic navigation environments, including both simulations and real-world scenarios, demonstrates that the proposed method significantly enhances performance in stochastic environments compared to existing LLM-based approaches. Code and supplementary materials are available at https://e2map.github.io/.