Abstract:

To improve the traffic efficiency of urban regional road network, this paper proposes a real-time timing method for regional traffic lights based on improved immune clonal selection algorithm. This timing method takes minimizing the total number of detained vehicles in the regional road network as the optimization objective, and converts the problem of traffic light state setting into the problem of clonal selection algorithm to search for the optimal solution, and dynamically searches the traffic light timing scheme with the highest capacity of regional road network according to the real-time traffic flow per unit time. In order to improve the convergence performance of clonal selection algorithm, a double-layer dynamic mutation operator is proposed, and the clonal suppression operator and population refresh operator are improved. Taking a regional road network in Xi’an as an example, the simulation results show that the total number of detained vehicles in the regional road network is 38.93% less than the fixed timing, and 20.33% less than the timing method based on standard genetic algorithm.

Key words: regional traffic, traffic signal timing, intelligent traffic light, clonal selection algorithm

摘要：

为提高城市区域路网通行效率，提出一种基于改进的克隆选择算法的区域交通灯实时配时方法。该配时方法以最小化区域路网总滞留车辆数为优化目标，将交通灯状态设置问题转换成克隆选择算法搜索最优解问题，在每个单位时间根据实时车流量动态搜索出使区域路网通行能力达到最高的交通灯配时方案。为提高克隆选择算法寻优性能，提出双层动态变异算子，并对克隆抑制算子与种群刷新算子进行改进。以西安市某区域路网为仿真实验参考对象，仿真结果表明：提出的配时方法的区域路网总滞留车辆数比固定配时减少了38.93%，比基于标准遗传算法的配时方法减少了20.33%。

关键词: 区域交通, 交通信号配时, 智能交通灯, 克隆选择算法