论文/arXiv 2610.21984·13 天前

GRID-Planner：用于扩散规划的图结构关系奖励塑形

针对复杂多智能体交互下的自动驾驶闭环规划难题，GRID-Planner用图关系奖励整合扩散规划，并在nuPlan和interPlan验证。

AI 整理

• 硬件零部件

原文链接：GRID-Planner: Graph-Structured Relational Reward Shaping for Diffusion Planning（arXiv 2610.21984 · 发布日期：2026-10-15）

GRID-Planner: Graph-Structured Relational Reward Shaping for Diffusion Planning

针对复杂多智能体交互下的自动驾驶闭环规划难题，GRID-Planner用图关系奖励整合扩散规划，并在nuPlan和interPlan验证。

核心要点

• 复杂多智能体场景中，安全、合规、舒适的轨迹选择与闭环决策仍具挑战。
• 方法结合图感知重要性加权、时间索引关系奖励替代和优势加权扩散目标。
• 通过在策略自蒸馏形成rollout-return-advantage-update循环，无需显式似然。
• 相较仅扩散基线，碰撞率、车道偏离、jerk分别降35.8%、60.8%、27.4%，规划分数升3.4%。

验证范围与局限：验证范围为nuPlan和interPlan上的实验，并与diffusion-only基线比较。

英文摘要 (Abstract)

Autonomous driving planning requires selecting trajectories that remain safe, rule-compliant, and comfortable under complex multi-agent interactions, yet robust closed-loop decision making in planning-critical scenes remains challenging. Existing diffusion- and RL-based planners often rely on generic rewards and use relational structure mainly for representation, limiting how planning-critical rollouts shape policy improvement. We present GRID-Planner, a graph--reward integrated diffusion planning framework that aligns generative planning with structured relational signals. Specifically, GRID-Planner combines graph-aware importance weighting, a time-indexed relational reward surrogate, and an advantage-weighted diffusion objective with on-policy self-distillation, forming a rollout-return-advantage-update loop without explicit likelihoods. Experiments on nuPlan and interPlan show that, compared with the diffusion-only baseline, GRID-Planner reduces collision rate, lane deviation, and jerk by 35.8%, 60.8%, and 27.4%, respectively, and improves the planning score by 3.4%. Project website: https://github.com/yangzhaojason/GRID-Planner.

作者：Zhao Yang, Zezhong Qian, Yinan Shi, Longjun Liu, Yawei Jueluo

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.21984：GRID-Planner: Graph-Structured Relational Reward Shaping for Diffusion Planning访问 2026-10-21