Abstract:

With the development of urban economy and the accelerated pace of people’s life, research on the travel time of people in the field of smart transportation has become a hot issue.The passing time in the estimated itinerary before travel makes it easier for people to plan the travel route more reasonably. Path planning based on time state characteristics is also one of the important means to solve the traffic problem.The existing models pay more attention to the vehicle arrival time or most of the predictions combined with the real and complete historical transit time data. There are few studies on whether the floating vehicle’s running state, vehicle speed, etc. have an impact on time.Based on this situation, this paper proposes a road time prediction model based on state features.In a fixed period of time, the trajectory data is deeply analyzed in combination with the situation of renting the vehicle passenger or not, the double parameter convolution theory model based on density partition is constructed based on vehicle speed, and the travel time is calculated by the final speed value.The experimental results show that the proposed model algorithm has higher accuracy and practicability than the traditional time prediction algorithm, which improves people’s rationalization and hierarchy of travel arrangements, and has important significance for formulating urban road travel strategies.

Key words: urban transportation, time series prediction, Pearson correlation coefficient, double parameter convolution, travel time, machine learning

摘要：

随着城市经济的发展和人们生活节奏的加快，智慧交通领域针对出行时间的研究已经成为热点问题。出行前预估行程中的通行时间便于人们更合理地规划出行路径，基于时间状态特征的路径规划就是解决交通问题的重要手段之一。现有模型多关注于车辆到达时间或多结合于真实历史时间数据进行预测，对浮动车的运行状态、车速等是否对时间存在影响的问题研究较少。基于此现状，提出了一种基于状态特征的道路时间预测模型，在固定时段内，利用出租车载客与否情况对轨迹数据进行深度相关性分析，结合车辆行驶速度构建一个基于密度划分的双参卷积理论模型，用得到的最终速度值对通行时间进行计算。实验结果表明该模型算法与传统时间预测算法相比有更高的精确度和实用性，提高了人们对出行安排的合理化和层次化，对制定城市道路出行策略具有重要的意义。

关键词: 城市交通, 时间预测, 皮尔逊相关系数, 双参卷积理论模型, 出行时间, 机器学习