建立基于路段的时间依赖网络模型,将转向延误时间引入到FIFO(先进先出)条件的定义中,并给出满足FIFO条件的路段到达时间和转向延误时间计算式。通过将时间因子引入到启发式评价函数中,发展了基于路段标号的时间依赖A*最短路径算法。试验表明,所提出的算法能预测并回避即将发生的交通拥堵,有效节省用户的出行时间。而其平均计算时间仅比传统算法增加10%左右。由于不再需要进行频繁的路径重优化,该算法能提高路径规划的整体效率。
A link-based time-dependent network model was built by introducing the turn delay time into the definition of "first in first out (FIFO)" condition. A link-labelling time-dependent A* shortest path algorithm is developed by adapting temporally the heuristic evaluation function and using Euclidian distance divided by maximum possible driving speed as the heuristic evaluator. An experiment on the real road network showed that the proposed algorithm is capable of forecasting and bypassing those forthcoming traffic congestions and then shortening travel time, only with a cost of about 10% more computational time than the traditional algorithms. Moreover, it is able to improve overall efficiency of route planning heavily because the frequent path re-optimization processes are no longer needed.