基于数字孪生的异常天气下车辆路径规划建模
Vehicle path planning modeling under abnormal weather based on digital twins
徐瑞萍 1郭昌鑫2
作者信息
- 1. 佛山大学马克思主义学院,广东佛山 528000
- 2. 佛山大学机电工程与自动化学院,广东佛山 528225
- 折叠
摘要
为解决当前自动驾驶系统在异常天气下传感器数据的采集具有不确定性,导致车辆路径规划中存在训练样本稀缺问题,提出了将数字孪生技术应用于自动驾驶系统复杂路况的车辆规划与控制中.通过搭建自动驾驶规划与控制系统的数字孪生框架,将物理实体数据与虚拟数据融合,扩充了异常天气下的孪生数据,采用数值优化的规划算法对异常天气下复杂路况进行规划,实现模型预测控制算法(MPC)模拟实现车辆对最优路径的跟踪,展现了应用数字孪生技术创建模拟异常天气的高效性和便捷性,验证了应用孪生数据进行车辆轨迹规划和跟踪的可行性.
Abstract
In order to solve the problem of shortage of training samples in vehicle path planning due to the uncertainty of sensor data collection in the current automatic driving system under abnormal weather,the digital twin technology was applied to vehicle planning and control in complex road conditions of the automatic driving system.By building the digital twin framework of the automatic driving planning and control system,the physical entity data and virtual data are fused to expand the twin data under abnormal weather,and the numerical optimization planning algorithm is adopted to plan the complex road conditions under abnormal weather,and the model predictive control algorithm(MPC)is simulated to realize the vehicle's tracking of the optimal path.It shows the high efficiency and convenience of using digital twin technology to create simulated abnormal weather,and verifies the feasibility of using twin data for vehicle trajectory planning and tracking.
关键词
异常天气/自动驾驶/数字孪生/数值优化/轨迹跟踪Key words
abnormal weather/autonomous driving/digital twins/numerical optimization/trajectory tracking引用本文复制引用
基金项目
教育部人文社会科学研究规划基金项目(23YJAZH170)
出版年
2024
国家科技期刊平台
NSTL
万方数据