车辆动态合乘匹配算法研究

发布时间：2019-01-28 12:18

【摘要】：随着物质生活水平的提高,越来越多的家庭开始拥有私家车。我国车辆数量的快速增长,伴随而来的是交通拥堵、环境污染等一系列问题。同时,虽然道路上车辆众多,但各大城市却仍然存在着不同程度的打车难问题。一味地拓宽道路和增加出租车数量都是无法很好解决这个问题的。为此,车辆合乘开始走入人们生活,将车上空余的座位作为乘坐资源利用起来。目前,欧美国家车辆合乘已经常态化了,而我国尚处于起步阶段。早期,一些提供车辆合乘信息的网站只提供基本的出行时间和起终点信息,采用文字匹配的技术使得乘客与车主之间难以得到有效配对,没有被大众广泛接受。因而,一个有效的车辆合乘匹配算法是车辆合乘问题研究的关键。当前,不少学者对车辆合乘匹配问题的研究大多只是研究确定性静态车辆匹配问题,研究中缺乏考虑外界因素对问题的影响,如道路通行情况、车主与乘客自身因素等。尤其是确定性静态车辆匹配问题在车辆出发前就已经规定了乘客上下车时间和位置,实际生活中车主与乘客很难能够严格按照时间窗规定到达乘车点,只要有一个环节出错,整个合乘很可能就是一个失败的合乘匹配。本文着眼于研究车辆动态合乘匹配问题,只有实时位置在乘客附近的车辆才有可能与之匹配,以此解决车辆到达时间不确定性的问题。借助于Dijkstra算法和遗传算法,使得随着乘客的加入,车辆总能得到当前新的最短路径。并在此基础上,检验时间窗,计算乘客与车辆之间的匹配程度,确定匹配关系。论文最后在仿真系统中构建了包含昆明市主要干道的路网,在此之上试验乘客与车辆的匹配过程,证明本文的算法能够有效解决车辆动态合乘问题。
[Abstract]:With the improvement of material standard of living, more and more families begin to own private cars. The rapid growth of the number of vehicles in China is accompanied by a series of problems, such as traffic congestion, environmental pollution and so on. At the same time, although there are a lot of traffic on the road, the major cities still have varying degrees of difficulty in taxi problems. Simply widening roads and increasing the number of taxis will not solve this problem very well. To this end, car-sharing began to walk into people's lives, using the spare seats as a ride resource. At present, the European and American countries have regular vehicle ride, and China is still in its infancy. In the early days, some websites that provided information on vehicle rides provided only basic travel time and terminal information, and the use of text matching technology made it difficult for passengers and car owners to match effectively and were not widely accepted by the public. Therefore, an effective vehicle matching algorithm is the key to the research of vehicle ridership problem. At present, many scholars only study the static vehicle matching problem, and lack of considering the influence of external factors, such as road traffic situation, vehicle owner and passenger factors, etc. In particular, the deterministic static vehicle matching problem has already stipulated the time and location of passengers getting on and off before the departure of the vehicle. In actual life, it is very difficult for the owner and passenger to reach the point of the ride strictly according to the time window, as long as there is a link error. The whole combination is likely to be a failed multiplicative match. In this paper, we focus on the dynamic matching problem of vehicles, which can only be matched by vehicles located near passengers in real time, so as to solve the problem of vehicle arrival time uncertainty. With the help of Dijkstra algorithm and genetic algorithm, with the addition of passengers, the vehicle can always get the new shortest path. On this basis, the time window is checked, the matching degree between passenger and vehicle is calculated, and the matching relationship is determined. Finally, the paper constructs a road network including the main trunk roads in Kunming, and tests the matching process of passenger and vehicle on the simulation system, which proves that the algorithm in this paper can effectively solve the problem of vehicle dynamic matching.
【学位授予单位】：昆明理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：U491

【参考文献】