基于Storm的城市智能交通规划方法

Urban Intelligent Transportation Planning Method Based on Storm

为了缓解城市交通拥堵,提高城市道路利用率,增强路径规划的实时性,提出基于Storm的城市智能交通规划方法。该方法在Storm框架的基础上,设计了城市实时动态路径规划模型(UR-MODE),将用户选择的静态目标和实时路况的动态目标相结合,实现城市交通路径的智能规划;针对城市交通数据实时性较强的特点,利用用户偏好值动态选择最优粒子,并结合自适应惯性权重策略和小规模扰动策略,提出改进的自适应粒子群优化(adaptive partner-particle swarm optimization,APPSO)算法,保证模型的高效求解;结合开源实时处理系统Storm,实现了海量实时交通数据处理。仿真实验结果表明,相较于现有路径规划算法,本文方法收敛速度更快、稳定性更强,能减少17%的车辆平均行驶时间,道路资源利用率平均提高58%,大大缓解了城市交通拥堵问题。

To alleviate the traffic congestion,raise the road utilization ratio and enhance the real-time performance of path planning,urban intelligent transportation planning method based on Storm is proposed.First,we design an urban real-time dynamic path planning model(UR-MODE)based on the Storm framework,realizing the intelligent planning of urban traffic route by combining static object of user selection and dynamic object of real-time road condition.Second,by considering the real-time characteristics of urban traffic data,we adopt the adaptive inertia weight and small-scale perturbation strategies to improve the performance of particle swarm optimization(PSO)algorithm and propose the adaptive partner-particle swarm optimization(AP-PSO)algorithm,which can ensure the efficiency of our proposed model.Third,we implement the improved algorithm on the opensource Storm real-time processing system and realize the mass real-time traffic data processing.Compared with the existing path planning algorithms,the simulation results show that the proposed method can achieve the faster convergence rate and stronger stability.The results also reveal that the proposed method can reduce the travel time by 17%on average and traffic resource utilization have risen by an average of 58%,greatly alleviating the urban traffic congestion problem.