基于Petri网与遗传算法的航空器滑行初始路径规划

为支持先进机场场面活动引导与控制系统(A-SMGCS,advanced surface movement guidance and control system)实施航空器滑行的精确引导,将场面分为滑行道交叉口和直线段等典型运行单元,利用改进的扩展赋时库所Petri网,建立了场面运行模块化模型;采用该模型进行染色体编码,并考虑场面运行管制规则,提出了染色体合法性检测与修复算法,以及染色体交叉和变异算法.基于首都国际机场01号跑道实际运行数据,用本文模型和算法进行了多个航班滑行初始路径规划,研究结果表明:与节点-路段类模型相比,本文模型能更充分地描述场面管制规则约束,可避免生成违反管制规则的路径;本文算法的每个航班初始路径规划耗时小于10 s,符合A-SMGCS的要求;由于考虑了航空器滑行速度调整特征,更符合场面运行的实际情况. 

Aircraft Initial Taxiing Route Planning Based on Petri Net and Genetic Algorithm

For the precision guidance of aircraft movement under the control of the advanced surface movement guidance and control system (A-SMGCS), airport surface was divided into typical operation units, such as taxiway intersection and line segment. A modular surface operation model was built based on the extended timed place Petri net (ETPPN), and used to code chromosomes. Then, algorithms for chromosome validation and amendment, crossover and mutation were proposed respectively, taking the surface operation rules into account. Using the daily operation data collected from runway 01 at Beijing Capital International Airport, and the initial taxiing routes were planned for each flight using the model and algorithms. The result demonstrates that, compared with the node-link model, the proposed model could depict surface operation rules more thoroughly and avoid generating illegal routes. The computation time of initial taxi routes for each flight is less than 10 s, which meets the requirement of A-SMGCS. In addition, as the aircraft taxi speed adjustment process is taken into account by the model, the traffic situation could be described more closely to the real surface operation.