E2Map introduces a novel approach for robotic navigation that enables agents to learn from their own experiences and emotional responses in dynamic environments.

• Real-time adaptation through a two-phase process: initial planning and experience-based replanning
• Memory integration that builds maps of environmental features and emotional responses
• Self-reflection capabilities allowing robots to identify and overcome navigation failures
• Superior performance demonstrated across various navigation tasks compared to existing methods

This research represents a significant advance for engineering applications by enabling more robust, adaptive robotic systems that can navigate unpredictable real-world environments with higher success rates.

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