Research On UAV Track Planning Based On Swarm Intelligence Algorithm

With the development of science and technology,UAVs are playing an increasingly important role in modern battlefields because of their flexibility,lightness,and strong concealment.A perfect mission planning system is the prerequisite guarantee for the UAV to complete the mission,and track planning is the core part of its mission planning system.With the advantages of good robustness and flexibility,swarm intelligence algorithms have become the most widely studied algorithms for track planning problems.The use of swarm intelligence algorithm for track planning has the problems of slow convergence speed and easy to fall into local optimum.In view of the above problems,it is very meaningful to use the swarm intelligence algorithm for track planning.Firstly,a digital map model for navigation space of UAV is established and used to transform the real-world environment information into an expressible mathematical model.The constraints of UAV itself and the objective function of completing the track planning are analyzed.Aiming at the problems of UAV 3D trajectory planning based on Artificial Bee Colony(ABC),it is easy to fall into local optimal,global search performance is not strong,and the convergence is poor.Based on the analysis of artificial bee colony algorithm,an algorithm based on Metropolis(Metropolis Artificial Bee Colony,M-ABC)is proposed.More relevant population initialization methods and search strategies based on step size in the bee colony field were introduced.The parallel selection mechanism of roulette and anti-roulette for honey source selection is utilized.and the Metropolis guidelines is used to exploit the potential of honey sources.Enhancing the correlation between the track points during the track planning of the algorithm,ensuring the continuity of the algorithm's vitality,and avoiding the poor development performance caused by ignoring potential honey sources and focusing too much on excellent honey sources.Through simulation experiment analysis,compared with the ABC algorithm,it can achieve faster convergence and less comprehensive cost of the track route.Secondly,aiming at the improvement of sparse A * Search(SAS)of the traditional artificial intelligence algorithm in the field of UAV track planning,a bi-directional Sparse A * Search(Bi-directional Sparse A* Search,Bi-SAS)algorithm was proposed.Analyzing the dimension explosion problem and search space redundancy problem of SAS algorithm during flight path planning.Based on SAS algorithm,a new cost function with weights is proposed,a two-way search strategy from the starting point to the target point in both directions is adopted.Compared with the SAS algorithm,the OPEN set is simplified,the planning time is shortened,and the algorithm execution efficiency is greatly improved.Finally,in order to solve the problem of coordinated track planning for multiple UAVs,an algorithm based on the fusion of M-ABC and Bi-SAS algorithms is proposed.A two-layer planning mechanism is established.First,the proposed fusion algorithm is used for single-UAV track planning to obtain multiple candidate tracks,and then it is extended to multi-machine coordinated track planning based on the established space-time constraints.Simulations show that the algorithm not only solves the problem of Poor obstacle avoidance performance of the SAS algorithm,but also improves the premature convergence problem of the ABC algorithm,proving the effectiveness and feasibility of the combination of the two algorithms.Under the established time constraints,the tasks of simultaneous arrival and interval arrival of multiple machines are completed.