为解决目前城市轨道交通平面线路设计受设计人员水平和经验的约束，耗时长且难以获得较优方案的问题，构建基于改进粒子群优化（PSO）算法的城市轨道交通线路平面避障优化设计系统，实现城市轨道交通平面线路的自动避障和优化设计。考虑到在城市轨道交通平面线路设计过程中，平面线路长度在很大程度上决定了项目的规模和投资额，采用区间线路双线总长度作为优化目标；除了考虑城市轨道交通设计规范对线路的要求，还考虑线路需要避开障碍物的约束。在给定站点和障碍物信息的前提下，所构建的平面线路优化设计模型具有高度非线性和复杂度。通过启发式逐段搜索算法提供可行的初始线路方案，并采用结合PSO算法和Rosenbrock搜索的改进PSO算法对初始线路方案进行优化，从而得到站点区间内优化的平面线路。以某条城市轨道交通线路为例进行实例分析，与实际设计方案相比，所得到的设计方案在成功避开障碍物的同时，可以有效减少平面线路长度，节约工程造价。

The current urban rail transit plane line design is limited by the level and experience of designers, which takes a long time and is difficult to obtain a better scheme. Therefore, an urban rail transit plane obstacle avoidance optimization design system based on improved particle swarm optimization (PSO) algorithm is constructed to realize automatic obstacle avoidance and optimal design of urban rail transit plane lines. In the process of urban rail transit plane line design, the length of plane line largely determines the scale and investment of the project, and the total length of double lines is adopted as the optimization objective. In addition to considering the requirements of the urban rail transit design code for the line, it also considers the constraints of the line to avoid obstacles. Under the premise of given station and obstacle information, the plane route optimization design model is highly nonlinear and complex. The heuristic segmental search algorithm is used to provide feasible initial route scheme, and the improved PSO algorithm combining PSO algorithm and Rosenbrock search is used to optimize the initial route scheme, so as to obtain the optimized plane route in the station area. A case study is conducted on an urban rail line, the obtained design scheme can effectively reduce the length of the plane line and save the project cost while successfully avoiding obstacles compared with the actual design scheme.