A fundamental problem in traffic networks is driving under safety and limited physical space constraints. In this paper, we design longitudinal vehicle controllers and study the dynamics of a system of homogeneous vehicles on a single-lane ring road in order to understand the interplay of limited space, speed, and safety. Each vehicle in the system either operates in the cruise control mode or follows a vehicle ahead by keeping a safe time headway. We show that if the number of vehicles is less than a certain critical threshold, vehicles can occupy the limited space in many different configurations, i.e., different platoons of different sizes, and they converge to a uniform maximum speed while attenuating errors in the relative spacing upstream a platoon. If the number of vehicles exceeds the threshold, vehicles converge to a unique symmetric configuration and the equilibrium speed decreases as the number of vehicles increases. Next, we consider vehicle-to-vehicle (V2V) communication and show that it increases the critical number of vehicles that can travel with the maximum speed. Finally, we consider central coordination and show that the proposed controllers can force vehicles to converge to a desired configuration specified by the coordinator while maintaining safety and comfort. We demonstrate the performance of the proposed controllers via simulation.
Robust Geospatial Coordination of Multi-Agent Communications Networks Under Attrition
Advanced Technology Center, Lockheed Martin Space（洛克希德·马丁航天公司先进技术中心）；Fu Foundation School of Engineering and Applied Science, Columbia University（哥伦比亚大学傅氏基金工程与应用科学学院）；Lockheed Martin AI Center, Lockheed Martin Corporation（洛克希德·马丁公司人工智能中心）；Barnard College, Columbia University（哥伦比亚大学巴纳德学院）；Dartmouth College（达特茅斯学院）
