摘要
为了解决智能汽车在虚拟仿真测试过程中存在的测试模块单一化、较难提供高密度高逼真交通流场景等问题,利用智能汽车仿真软件PreScan的环境感知仿真与规划决策开发工具箱MATLAB/Simulink,以及交通仿真软件Vissim快速生成高逼真环境车流的功能,搭建了智能汽车"软件在环"一体化测试平台。该平台可快速搭建智能汽车高逼真仿真场景,可快速有效地对智能汽车的环境感知与规划决策模块进行一体化仿真测试,还能综合评估智能汽车对交通流的影响。通过快速路汇入区仿真案例,验证了本仿真测试平台的有效性。
The simplicity of testing modules and the lack of high-fidelity and high-density traffic scenes are the primary issues in the intelligent vehicle’ simulation test. To solve these problems, a "software in the loop" integration test platform is built. The environmental perception module is integrated on the basis of the PreScan, the planning decision-making module designed by MATLAB/Simulink and the virtual traffic flow environment from Vissim. Based on this platform, high-fidelity traffic flow scenes can be generated rapidly to support the test of environment perception and decision-making integratedly and effectively. Moreover, the intelligent vehicle’s influence on the traffic flow can be evaluated comprehensively. A case study of an expressway on-ramp proves the feasibility of the simulation platform.
作者
孙剑
黄润涵
李霖
刘启远
李宇迪
Sun Jian;Huang Runhan;Li Lin;Liu Qiyuan;Li Yudi(Key Laboratory of Road and Traffic Engineering of Ministry of Education,Tongji University,Shanghai 201804,China;Shanghai International Automobile City,Shanghai 201804,China)
出处
《系统仿真学报》
CAS
CSCD
北大核心
2020年第2期236-246,共11页
Journal of System Simulation
基金
国家自然科学基金(51422812)
上海市科委国际科技合作项目(16510711400)