Self-evolving Embodied AI
This work addresses the problem of enabling embodied AI to operate in variable, in-the-wild settings, which is a foundational step toward general artificial intelligence.
The paper tackles the limitation of existing embodied AI systems that rely on human-crafted settings by proposing a new paradigm called self-evolving embodied AI, which enables agents to autonomously adapt through memory self-updating, task self-switching, and model self-evolution for dynamic environments.
Embodied Artificial Intelligence (AI) is an intelligent system formed by agents and their environment through active perception, embodied cognition, and action interaction. Existing embodied AI remains confined to human-crafted setting, in which agents are trained on given memory and construct models for given tasks, enabling fixed embodiments to interact with relatively static environments. Such methods fail in in-the-wild setting characterized by variable embodiments and dynamic open environments. This paper introduces self-evolving embodied AI, a new paradigm in which agents operate based on their changing state and environment with memory self-updating, task self-switching, environment self-prediction, embodiment self-adaptation, and model self-evolution, aiming to achieve continually adaptive intelligence with autonomous evolution. Specifically, we present the definition, framework, components, and mechanisms of self-evolving embodied AI, systematically review state-of-the-art works for realized components, discuss practical applications, and point out future research directions. We believe that self-evolving embodied AI enables agents to autonomously learn and interact with environments in a human-like manner and provide a new perspective toward general artificial intelligence.