阈值改进蚂蚁算法在交通诱导中的应用

研究基于阈值改进的蚂蚁算法在交通诱导中的应用，并对动态交通网络条件下算法的优化效果进行仿真.改进算法将已知的行程时间信息作为阈值条件，在路径寻优中预先剔除部分路径，并引入“检查蚁”对已发现的路径进行定期检查与存储.利用R软件对改进前后的两种算法进行仿真分析，对两者的路径寻优结果对比研究.实验发现改进算法下路径检索的平均行程时间较标准蚂蚁算法下降9.57%，证明了基于阈值改进的蚂蚁算法的优化效果，能形成适应城市路网即时高效需求的交通诱导方案.

Threshold Based on Antnet Algorithm for Traffic Guidance

The application of Threshold based Antnet Algorithm in traffic guidancewas presented. Simulation was conducted to evaluate the effect of the algorithm in the dynamic traffic system. Prior information for the traffic network was used as threshold, for instance, good travel time between nodes. The algorithm eliminated certain routes according to those preset values. Besides, "check ants" were introduced to periodically check the assumptions and save the routes. Optimization performance of the presented algorithm was certified with dynamic traffic environment using R software. The simulation shows a 9.57% reduction in the average travel time of ants launched in presented algorithm compared with the classic algorithm. Threshold based traffic guidance scheme can meet real time and high efficiency demand of city traffic.