考虑车车与车路通讯并存的路侧单元优化布设方法

路侧单元(Road Side Unit,RSU)作为车路协同系统(Cooperative Vehicle Infrastructure System,CVIS)中路侧感知控制网与网联化车辆自组网的重要节点，其布设方案对车路交互效能、系统建设成本与协同服务水平均产生较大影响。现有RSU优化布设研究主要基于车路(Vehicle to Infrastructure,V2I)通讯，忽视了车车(Vehicle to Vehicle,V2V)通讯的转发和中继作用，也缺乏对RSU设备及线网布设成本的相关考量。针对上述问题，面向V2V和V2I通讯并存环境，考虑信息覆盖正向收益与设备/线网布设负向收益，构建面向综合收益最优的RSU布设方案优化模型。首先结合信息流-交通流耦合模型获取传播时间矩阵，提出权重约束可达性算法，构建布设方案与信息覆盖正向收益之间的关系方程。然后针对动态最优布设方案条件下的最优线网布设问题，提出改进的最小生成树模型，并以此设计综合收益最优的RSU优化布设模型。最后，基于井字形算例路网，进行模型验证和关键系数敏感性分析；并基于Sioux Falls路网，将所提出的模型与现有未考虑信息流-交通流耦合特性的RSU优化布设模型进行对比。研究结果表明：所构建的布设方法可较好地反映实际信息流传播状态，覆盖率误差在5%以内；RSU最优布设方案与交通流运行状态密切相关，交叉口及路段中节点均有成为最优节点的可能；RSU布设方案将受线网/设备布设成本和信息生命周期的直接影响；对比现有模型，在Sioux Falls路网和所给定参数的条件下，所提方法在信息覆盖率和布设后的综合收益方面相较现有方法均提升了30%以上。

Road-side Unit Optimal Layout Scheme by Considering Coexisting of V2I and V2V Communication

As the key junction between traffic sensing-control network and vehicular ad-hoc network, the road side unit (RSU) is one of essential the parts of cooperative vehicle infrastructure system (CVIS). The layout of RSUs critically affects the vehicle-infrastructure interactive efficiency, CVIS construction cost, and service level.Currently, most studies mainly focus on finding the optimal RSU layout by maximizing the information coverage range by vehicle-to-infrastructure (V2I) communications. Notwithstanding, it ignores the forwarding and relaying of vehicle-to-vehicle (V2V) communications in information propagation. The equipment and network layout costs are also not considered in previous RSU layout schemes. Motivated by the above research gaps, this study provides a RSU optimal layout model by maximizing the comprehensive profit, which considers the coexisting of V2I and V2V communication. This study firstly investigates the relationship function between the profit of the information coverage and the RSU layout plan, which is obtained by information-traffic flow coupling propagation model for existing research and a weight constraint reachability algorithm designed by this study. According to a temporal RSU layout plan, a minimum network layout plan among the selected RSUs is provided by a modified minimum spanning tree model. The proposed RSU optimal layout model is established by integrating the profit-information coverage relationship function and the modified minimum spanning tree model. This study is verified by a well shape toy network and the Sioux Falls network. The reachability of the information flow, the optimal layout of given RSU number or candidates' position sets, and the sensitivity of key parameters are analyzed in numerical experiments by the toy network. It also compares with an existing study without the information-traffic flow coupling considering by the Sioux Falls network. The results indicate that the proposed relationship function between information coverage and the RSU layout plan is close to reality with the error less than 5% in the terms of coverage rate. The optimal RSU layout position can be an intersection or a node in a segment. The RSU layout design critically affects by the network traffic condition, equipment and network layout cost, and the information life cycle. According to the Sioux Falls network and given parameters, the proposed study can provide more than 30% enhancement in information coverage rate and comprehensive profit by comparing with the existing study.