用常识世界+可信度机制，让自动驾驶轨迹规划更安全可靠。
C-TRAIL: A Commonsense World Framework for Trajectory Planning in Autonomous Driving
- 将大模型常识与可信度评估结合，动态调整决策权重。
- 在多个场景中平均降低40.2%的平均位移误差，提升16.9个百分点成功率。
- 适合关注自动驾驶安全性与大模型可信应用的研究者。
自动驾驶轨迹规划越来越多地依赖大语言模型（LLMs）进行常识推理，但其输出固有不可靠，危及安全关键应用。我们提出C-TRAIL框架，基于常识世界，融合LLM生成的常识与可信度机制，指导轨迹规划。该框架采用闭环的召回、规划、更新循环：召回模块通过双重可信度机制量化语义关系的可靠性；规划模块将可信度加权的常识注入蒙特卡洛树搜索（MCTS），采用狄利克雷可信度策略；更新模块则根据环境反馈自适应优化可信度评分与策略参数。在Highway-env四个仿真场景及真实数据集highD、rounD上的实验表明，C-TRAIL显著优于现有基线，平均降低ADE 40.2%、FDE 51.7%，并提升成功率达16.9个百分点。源代码已公开于https://github.com/ZhihongCui/CTRAIL。
原文摘要 · Abstract (English)
Trajectory planning for autonomous driving increasingly leverages large language models (LLMs) for commonsense reasoning, yet LLM outputs are inherently unreliable, posing risks in safety-critical applications. We propose C-TRAIL, a framework built on a Commonsense World that couples LLM-derived commonsense with a trust mechanism to guide trajectory planning. C-TRAIL operates through a closed-loop Recall, Plan, and Update cycle: the Recall module queries an LLM for semantic relations and quantifies their reliability via a dual-trust mechanism; the Plan module injects trust-weighted commonsense into Monte Carlo Tree Search (MCTS) through a Dirichlet trust policy; and the Update module adaptively refines trust scores and policy parameters from environmental feedback. Experiments on four simulated scenarios in Highway-env and two real-world levelXData datasets (highD, rounD) show that C-TRAIL consistently outperforms state-of-the-art baselines, reducing ADE by 40.2%, FDE by 51.7%, and improving SR by 16.9 percentage points on average. The source code is available at https://github.com/ZhihongCui/CTRAIL.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。