【摘要】 为加快交通信号控制中迭代学习控制的收敛速度，解决迭代学习控制历史信息利用率低的问题，提出一种基于历史信息迭代学习控制的交通信号控制方法。利用欧氏距离从历史信息数据库中找到相似的历史数据，对数据进行加权处理，给相似度高的数据赋予更大的权重；根据数据对应的信号配时及权重，计算得到适合的初次迭代控制信号，应用到PD型迭代学习控制中。与自主设置初次迭代控制信号的PD型迭代学习控制相比，以更快速度使各个交叉口的排队长度趋于均衡状态，充分利用一个周期的绿灯时长，提升路网的通行效率。利用仿真实验验证了算法的有效性。更多还原

【Abstract】 To accelerate the convergence speed of iterative learning control in traffic signal control and solve the problem of low utilization of historical information in iterative learning control, a traffic signal control method based on iterative learning control of historical information was proposed. Similar historical data were found from the historical information database using Euclidean distance, and the data were weighted to give greater weights to the data with high similarity. The suitable initial iterative control signal was subsequently calculated based on the signal timing and weights corresponding to the data and applied to PD-type iterative learning control. Compared with the PD-type iterative learning control that sets the initial iterative control signal autonomously, it achieves higher speed to equalize the queue length of each intersection, fully utilizes the green light duration of one cycle, and improves the traffic efficiency of the road network. The effectiveness of the algorithm is verified using simulation experiments.更多还原