Multi-vehicle Air Conflict Resolution Based On Genetic Algorithm And Particle Swarm Optimization

With the rapid development of economy and technique, the civil aviation industryhas also entered a period of rapid development, and the growing needs caused ourairspace crowded and the safety problems increase. Currently the air traffic managementhas dealt with the flight conflict by changing the flight plan, which resulted in a backlogof large number of flights and the phenomenon of flight delay more serious. Theimplementation of “Free Flight” can change this problem.Free flight allows pilots to select the most suitable flight route and speed accordingto the actual situation. In free flight, route congestion will be improved, but the abolitionof the route structure restrictions and the growth of flight plan could increase theprobability of flight collision, and the solution to conflict between planes will becomemore complex. Flight conflict detection and resolution in free flight become animportant research topic of the air traffic management.In this article, the application of background is Air Traffic Control SystemSimulation Platform. To enhance the search speed and resolution path close to theoriginal route and suitable for actual flight, flight conflict resolution model wasimproved, and then particle swarm optimization and a algorithm based on geneticalgorithm and particle swarm algorithm was proposed to multi-vehicle conflictionresolution. Specific work can be briefly summarized as follows:①Considering the current conflict resolution model and the actual flight situation,flight conflict resolution model was improved and the sum of the distances between allresolution track points to the original path is designed to planning objective.②Particle swarm algorithm is proposed to resolve multi-vehicle conflictionresolution and this algorithm is also improved by taking the deviation distance of theentire track as a planning goal. Then some simulation results show the feasibility of thealgorithm;③To enhance the search speed and avoid falling into the local optimization andpremature convergence by particle swarm algorithm，a algorithm based on geneticalgorithm and particle swarm algorithm was proposed to multi-vehicle conflictionresolution．This algorithm was integrated the global search capability of geneticalgorithm and the memory function and fast convergence properties of particle swarmalgorithm, which is able to draw closer resolution path to the original flight plan and better in the search accuracy. The algorithm is applied to air traffic control simulationplatform and the validity is proved by using the flight data.