重庆理工大学学报2025，Vol.39Issue(5)：24-33,10.DOI:10.3969/j.issn.1674-8425(z).2025.03.004
优化采样点选择策略的自动驾驶轨迹规划研究
Research on autonomous driving trajectory planning with optimized sampling point selection strategy
摘要
Abstract
To improve the timeliness,safety and comfort of trajectory planning for autonomous driving,we propose an autonomous driving trajectory planning method that optimizes the sampling point selection strategy.It selects the optimal trajectory by means of a cost function.We thoroughly consider the road risks and obstacle risks to construct the risk cost function based on trigonometric function and two-dimensional Gaussian function.Meanwhile,the obstacle screening function is introduced,improving the ability to deal with the complex environment.To address the redundant computation in traditional dynamic programming algorithms,the sampling point selection strategies of maximum lateral constraints and non-reversible constraints are designed respectively based on the vehicle kinematics,dynamics properties and driving characteristics.Thus,the efficiency of dynamic programming algorithms is improved.We build a speed optimization model to ensure that the speed profile meets the speed limit,safety and comfort criteria.Our joint simulation experiments demonstrate the algorithm markedly improves its computational efficiency and meets the safety requirements in different road environments while ensuring the comfort level.关键词
自动驾驶/轨迹规划/驾驶风险评估/动态规划Key words
automated driving/trajectory planning/driving risk assessment/dynamic programming分类
交通运输引用本文复制引用
李文超,张一政,盘朝奉..优化采样点选择策略的自动驾驶轨迹规划研究[J].重庆理工大学学报,2025,39(5):24-33,10.基金项目
国家自然科学基金项目(52272367) （52272367）
江苏大学京江学院一流课程建设项目(2023ylkc005) （2023ylkc005）
