迈向统一机器人学习：联结表征、视觉-语言-动作（VLA）与世界模型
Toward Unified Robot Learning: Bridging Representation, Vision-Language-Action, and World Models
- Fujitsu Research of America（美国富士通研究院）
- Carnegie Mellon University（卡内基梅隆大学）
- Fujitsu Limited（富士通株式会社）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本综述提出统一机器人学习视角，梳理表征、VLA模型与世界模型三类方法，分析其交互与局限，展望未来集成化、物理接地的概率化机器人学习方向。
AI中文摘要：
机器人要在真实环境中可靠运行，需感知周围环境、执行动作并推理动作后果。表征学习、视觉-语言-动作（VLA）模型与世界模型领域的快速进展，大幅提升了机器人学习系统的能力，使其能在日益复杂的环境中工作。但这些范式通常独立开发，导致系统碎片化，难以实现泛化、长时序推理与规划，也难以在非结构化环境中部署。本综述从统一视角出发，沿三个互补轴梳理现有机器人学习方法：通过表征学习实现环境理解，通过VLA模型实现动作执行，通过世界模型实现推理。我们提出结构化分类法，涵盖环境表征、策略学习与预测建模的关键设计选择，并总结这些领域的最新进展。除对现有研究分类外，我们还分析各组件的交互方式，探讨共同局限性，并指出更集成化系统的新兴趋势。通过该视角，我们识别出机器人学习领域的挑战，包括不确定性量化、分布外泛化、跨 embodiment（ embodiments ）迁移、长上下文理解与长时序规划。我们认为这些挑战不仅源于各组件内部的局限，还源于感知、动作与推理之间缺乏集成。基于此分析，我们展望未来方向，即迈向统一、物理接地与概率化的机器人学习，以开发能维持一致内部表征、支持真实环境中长时间交互决策的稳健机器人系统。
英文摘要：
For robots to operate reliably in real-world environments, they need to perceive their surroundings, act, and reason about the consequences of those actions. Rapid progress in the domains of representation learning, VLA models, and world models has significantly enhanced the capabilities of robot learning systems, enabling robots to work in increasingly complex environments. However, these paradigms are typically developed in isolation, resulting in fragmented systems that struggle with generalization, long-horizon temporal reasoning and planning, and deployment in unstructured environments. In this survey, we present a unified perspective on robot learning by organizing the existing methods along three complementary axes: understanding through representation learning, acting through VLA models, and reasoning through world models. We introduce a structured taxonomy that captures key design choices in environment representation, policy learning, and predictive modeling, and summarize the recent progress in these domains. Beyond classifying the existing works, we analyze how these components interact, discuss common limitations, and highlight emerging trends towards more integrated systems. Through this lens, we identify the challenges in the domain of robot learning, including uncertainty quantification, out-of-distribution generalization, cross-embodiment transfer, long-context understanding, and long-horizon planning. We argue that these challenges arise not only from limitations within individual components but also from the lack of integration across perception, action, and reasoning. Building on this analysis, we outline future directions towards unified, physically grounded, and probabilistic robot learning to develop robust robotic systems that maintain consistent internal representations and support decision making over extended interactions in real-world environments.
