A model of a smart road network consisting of unsignalised intersections and smart roads connecting them is considered in this work with the aim of presenting a traffic management system for self-driving cars (or, more generally, autonomous vehicles) which travel the network. The proposed system repeatedly solves a set of mathematical programming problems (each of them relative to a single intersection or to a single road stretch of the network) within a decentralised control scheme in which each local intersection controller and each local road controller communicates with the fully autonomous vehicles in order to receive travel data from vehicles and to provide speed profiles to them once determined the optimal solution of the problem. In order to reduce the computational effort required to provide the optimal solution, a discrete-time approach is adopted so that, in each time interval, a limited number of vehicles are taken into consideration; in this way, solutions can be determined in a very short time thus making the proposed model compatible with a practical application to real traffic systems. The proposed model is general enough, and can be adapted to different scenarios of smart road networks reserved for self-driving cars.

中文翻译：

---

自动驾驶汽车专用的智能道路网络交通管理系统

在这项工作中，考虑了一种由无信号交叉口和连接它们的智能道路组成的智能道路网络模型，其目的是为行进网络的自动驾驶汽车（或更普遍的是自动驾驶汽车）提供一种交通管理系统。所提出的系统在分散控制方案中反复解决了一组数学编程问题（每个问题均与网络的单个交叉点或单个道路段有关），在该分散控制方案中，每个本地交叉路口控制器和每个本地道路控制器均与全车道通信。一旦确定了问题的最佳解决方案，自动驾驶车辆便可以从车辆接收行驶数据并为其提供速度曲线。为了减少提供最佳解决方案所需的计算量，采用离散时间方法，以便在每个时间间隔内考虑有限数量的车辆；以这种方式，可以在很短的时间内确定解决方案，从而使所提出的模型与实际交通系统的实际应用兼容。所提出的模型足够通用，并且可以适用于为自动驾驶汽车保留的智能道路网络的不同场景。