基于瞬时阻抗的路径规划策略仿真性能评估

摘要

根据对实时路况的数据处理方式,常见的动态路径规划策略可分为瞬时阻抗T-SPF和平均阻抗A-SPF两种方式.如何验证这些策略的个体效益及其对城市路网的中长期影响一直是相关系统部署的核心问题.针对这一问题,通过Vissim的Car2X模块与自定义程序实现了仿真中的动态转向,并以现实路网为蓝本,对T-SPF和A-SPF在不同程度的市场占有率情况下的运行性能进行了测试.测试结果表明2种现有策略都存在各自的问题;当市场占有率达到15%以上时,T-SPF 策略可能令路网的通行容量处于振荡状态,而实施了 A-SPF 的路网的平均延误和通行量表现稳定,但在个体效益方面表现还不如静态策略.建议为城市路网实施路径规划时设计新的策略以及与之匹配的信号优化机制.%Traditional strategies of dynamic route planning can be divided into Transient-value-based Shortest-Path-Finding (T-SPF) and Average-value-based Shortest-Path-Finding (A-SPF) in term of processing for real-time traffic conditions.It is significant for traffic management departments to know how to balance individual benefits and long-term impacts of these strategies.To this end, a series of situations based on actual road networks are simulated.Vissim Car2X module and customized programs are applied to achieve dynamic steering during the simulations.The results show that the performances of T-SPF and A-SPF under different levels of market share have their own problems.When the market share is higher than 15%, the T-SPF strategy may cause an oscillatory state in traffic capacity of urban road networks.The A-SPF strategy can achieve stable performance in average delay and traffic throughput of road networks under the same conditions.However, it shows poor performance in individual benefit.Suggestions for solving these problems are coordination of a new route planning strategy, and enhancement of an adaptive signal control policy.