典型交通场景下智能车的定位与协作

提出并设计了基于多智能体的多智能车协作系统平台,包括基于嵌入式多车协作系统的硬件、软件及通信协议,该系统可作为多车协作算法的仿真实验平台。针对典型城市交通场景,提出了基于拓扑结构的两层地图模型,并采用视觉与里程结合的方法实现智能车在地图中的准确定位。在此基础上,设计了基于冲突表的路口协调算法、车队跟随算法,并完成路口、车队、超车等协作任务,通过实验结果证明了多车协作系统定位与协作的可行性和可靠性。

Localization and Cooperation of Intelligent Vehicles in Typical Transportation

This paper builds and designs a distributed cooperative system of intelligent vehicles based on multi-agents. The embedded system includes multi-vehicle cooperative system’s hardware, software and communication protocol. This system can be used as simulation platform for research on coordination algorithm of multi-vehicle cooperative system. Considering typical urban traffic situation, this paper proposes two layers’ cartographic model based on topology structure, and implements vehicle’s precise localization with the combination of visual system and odometer. On this basis, the paper designs coordinating algorithm for intersection based on conflict table, and platoon algorithm. This article implements intersection, platoon and overtaking with this system. Experiment results prove the feasibility and reliability of multi-vehicle system’s localization and cooperation.