提出一种基于改进遗传算法,并将其应用于城市交通诱导策略.该算法采用邻近交叉策略增加种群的多样性,采用贪婪前向变异策略增强算法的局部搜索能力,并且采用稳态繁殖策略更好地保护适应度高的遗传个体.在适应度函数的设计中考虑实际路网中的交叉口延误,使所求最优路径更加贴近实时的交通状态.数值实验结果表明,所提算法的全局搜索能力和收敛速度均有显著提高. A genetic algorithm based on improved genetic algorithm is proposed and applied to urban traffic guidance strategy. This algorithm uses adjacent crossover strategy to increase the diversity of the population, and uses greedy forward mutation strategy to enhance the local search ability of the algorithm. So as to well protect genetic individuals with high adaptability.In the design of fitness function, the delay of intersection in the actual road network is considered, so that the optimal path is closer to the real-time traffic state.Numerical experimental results show that the global Search capabilities and convergence rates have significantly improved.