ReWAM：用于交互式自动驾驶的互惠世界动作模型
ReWAM: Reciprocal World Action Models for Interactive Autonomous Driving
查看机构详情
- The Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州））
- Leapmotor（零跑汽车）
机构由 AI 辅助整理，请以论文原文为准。
浏览论文内容
中文总结 AI 辅助
针对交互式自动驾驶，提出互惠世界动作模型（ReWAM），通过博弈论框架和Level-k响应层次建模自车与他车的互惠影响，在NAVSIM数据集上取得最先进性能，尤其显著提升交互场景表现。
中文摘要 AI 辅助
在交互式场景中，自动驾驶系统需要在其他智能体行为的影响下生成自车动作。现有的世界动作模型（WAMs）通常将其他智能体建模为世界模型的组成部分，而非从根本上塑造自车动作的决策者，这损害了它们在密集交互场景中的性能。我们提出了互惠世界动作模型（ReWAM），一种博弈论世界动作建模框架，通过将自车和其他智能体表示为条件响应者（其动作相互影响）来捕捉它们之间的互惠影响。我们使用Level-$k$响应层次结构实例化该框架，其中特定角色的自车和其他动作DiTs通过跨智能体注意力交换紧凑的策略令牌，同时保持基于未来驾驶世界的共享表示。为了从示范中学习自车的响应策略，我们将专家动作表述为最佳响应分布的样本，并使用条件流匹配联合优化整个层次结构。我们的框架在NAVSIM数据集上进行了评估，与基线相比达到了最先进的性能。在交互式场景中改进尤为显著，验证了建模互惠响应为交互感知的世界动作生成提供了更有效的基础。
英文摘要
In interactive scenarios, an autonomous driving system is required to generate ego actions under the influence of other agents' behaviors. Existing World Action Models (WAMs) typically model other agents as components of the world model rather than as decision-makers that fundamentally shape the action of the ego agent, which impairs their performance in dense interaction scenarios. We introduce Reciprocal World Action Models (ReWAM), a game-theoretic world action modeling framework that captures the reciprocal influence between the ego agent and other agents by representing them as conditional responders whose actions are mutually influenced. We instantiate this framework with a Level-$k$ response hierarchy, where role-specific ego and other action DiTs exchange compact strategy tokens through cross-agent attention while remaining grounded in a shared representation of the future driving world. To learn the response policy of the ego agent from demonstrations, we formulate expert actions as samples from the best response distribution and jointly optimize the entire hierarchy using conditional flow matching. Our framework is evaluated on the NAVSIM dataset and achieves state-of-the-art performance compared to baselines. The improvement is particularly significant in interactive scenarios, validating that modeling reciprocal responses provides a more effective foundation for interaction-aware world action generation.
