Toward the Cognitive--Physical Limits of Embodied Intelligence through a World-Model-Centric Autonomous Racing Agent
作者: Zitong Shan, Baichuan Lou, Yanxin Zhou, Shuge Wu, Xianqi He, Bolin Zhao, Sheng Zhao, Zhouheng Li, Chee Kiong Ong, King Ho Holden Li, Chen Lv
分类: cs.RO
发布日期: 2026-08-11
💡 一句话要点
提出基于世界模型的自主赛车代理以探索具身智能的极限
🎯 匹配领域: 支柱二：RL算法与架构 (RL & Architecture)
关键词: 具身智能 自主赛车 世界模型 闭环优化 认知极限 物理极限 高频定位 动态交互
📋 核心要点
- 现有的具身智能系统在极端条件下的能力边界尚不清楚，缺乏对认知与物理极限的联合建模与优化。
- 论文提出了一种基于世界模型的自主赛车代理，通过学习预测模型来捕捉动态交互和运动边界。
- 实验结果表明，该代理在全规模模拟赛车中实现了88.3%的交互成功率，并在多种条件下表现出更好的泛化能力。
📝 摘要（中文）
具身人工智能旨在开发能够通过与物理世界的持续互动进行感知、推理和行动的代理。然而，大多数具身系统仍在保守的安全边际或适度的交互模式下进行评估，导致其在极端条件下的能力边界尚不清楚。自主赛车提供了一个严格的测试平台，结合了高频定位与感知、对抗性互动、近饱和的车辆动态和严格的安全约束。本文展示了一种基于世界模型的自主赛车代理，提供了探索这些耦合极限的具体步骤。该框架从近极限的成功与失败中学习预测世界模型，以捕捉互动演变、个体动态和可行运动边界，结合世界状态构建、未来感知推理和近极限控制，形成闭环优化过程。
🔬 方法详解
问题定义：本文旨在解决现有具身智能系统在极端条件下能力边界不明确的问题，现有方法往往未能同时建模认知与物理极限。
核心思路：提出基于世界模型的自主赛车代理，通过从成功与失败中学习，捕捉动态交互和运动边界，以实现更高效的决策与控制。
技术框架：整体架构包括数据收集、世界模型学习、闭环优化和策略调整等模块。数据通过真实车辆在高速下的自主赛车收集，模型通过预测未来状态进行优化。
关键创新：最重要的创新在于将世界模型与认知-物理极限的联合优化相结合，形成了一种新的边界感知方法，与传统方法相比，能够更好地适应复杂环境。
关键设计：在模型训练中，采用了多种损失函数以平衡预测准确性与控制稳定性，同时设计了适应高速动态的网络结构以处理复杂的车辆行为。
🖼️ 关键图片
📊 实验亮点
实验结果显示，经过充分训练的基于世界模型的代理在全规模模拟赛车中实现了88.3%的交互成功率，相较于传统方法显著提升了认知-物理极限的利用效率，并在多种复杂场景中表现出更强的泛化能力。
🎯 应用场景
该研究的潜在应用领域包括自动驾驶、机器人控制和智能交通系统等。通过提升具身智能系统在极端条件下的表现，能够为安全的自动驾驶技术和高效的交通管理提供重要支持，推动智能交通的实际应用与发展。
📄 摘要（原文）
Embodied artificial intelligence aims to develop agents that perceive, reason, and act through continuous interaction with the physical world. However, most embodied systems are still evaluated within conservative safety margins or moderate interaction regimes, leaving their capability boundaries under extreme conditions insufficiently understood. Autonomous racing provides a stringent testbed by combining high-frequency localization and perception, adversarial interaction, near-saturated vehicle dynamics, and strict safety constraints. Existing systems push high-speed performance but rarely model and refine cognitive and physical limits jointly. Here we show that a world-model-centric autonomous racing agent provides a concrete step toward exploring these coupled limits. The framework learns predictive world models from near-limit successes and failures to capture interaction evolution, ego dynamics, and feasible-motion boundaries, coupling world-state construction, future-aware reasoning, and near-limit control in a closed-loop refinement process. Training data were collected from real-vehicle autonomous racing, where the onboard system maintained robust localization and perception at speeds up to 256.3 km/h and peak lateral acceleration of 26.8 m/s$^2$. In full-scale simulated racing, the well trained world-model-centric agent achieves an 88.3% interaction success rate across various challenging simulated racing scenarios. Closed-loop refinement of the world model and policy further improved utilization of cognitive-physical limits, recovery from failure modes, and generalization across varying conditions and unseen circuits. These results suggest a boundary-aware methodology in which world models help embodied agents represent, predict, and continually refine their capability boundaries for safer real-world deployment.