SanDRA: Safe Large-Language-Model-Based Decision Making for Automated Vehicles Using Reachability Analysis
SanDRA:基于可达性分析的自动驾驶车辆安全大语言模型决策
机构 * Department of Computer Engineering, Technical University of Munich(计算机工程系,慕尼黑技术大学) ; Munich Center for Machine Learning(慕尼黑机器学习中心)
专题命中 规划推理 :reasoning(abstract)
AI总结 针对大语言模型用于自动驾驶决策时安全性无法保证的问题,提出SanDRA框架,先描述驾驶场景促使模型生成排序动作,转化为时序逻辑公式并结合可达性分析消除不安全动作,经实验验证能提供安全合规驾驶动作,代码和实验设置公开。
Comments @2026 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works
Journal ref T-ITS, 2026