自适应变异粒子群算法在交通控制中的应用

提出了自适应粒子群算法结合实数遗传算法中变异算子的混合算法，它能提高算法的收敛性和稳定性。同时，通过对交通路口的通行情况的研究提出了一种新颖的离散交通信号控制模型。此模型以交叉路口各方向车流支路为基本单元，以各支路车流信息为输入，得出交通信号控制的各项性能指标。在此模型的基础上，应用自适应变异粒子群算法实现交通信号优化控制及验证算法。仿真结果表明自适应变异粒子群算法能够有效实现交通信号优化控制。

Application of Adaptive Mutation- particle Swarm Optimization Algorithm in Traffic Control

A hybrid algorithm, which combines mutation operator in real-code genetic algorithm with adaptive particle swarm optimization algorithm (AMPSOA), was proposed. The new method increases its convergence rate and stability. A novel discrete model of traffic signal control was proposed based on the research of the situation of an intersection. It is composed of branches of each direction of an intersection. According to the discrete input information of traffic flow in each branch of each direction, the performance indexes of traffic signal control can be achieved. Based on the model, the method can be applied to the optimal control of traffic signal and examined. Simulation demonstrates the effectiveness of the AMPSOA which can realize the optimal control of in traffic signal.