Research On Cooperative Path Planning For Multiple Aircrafts Based On Improved Ant Colony Algorithm

The research on path planning of cooperative penetration for multiple aircrafts is of great importance to enhance the penetration ability of combat aircrafts. With the development of anti-aircraft weapons, the battlefield becomes more dangerous and uncertain, and the mission constraints become diversified. Under such circumstances, the mission environment makes path planning of cooperative penetration more difficult. The main achievements and progress are summarized as follows:1. The cooperative path planning problem of multiple aircrafts is analyzed and modeled. A model is constructed to formulate the threat space by using terrain masking. An area-light-of-sight criteria is designed and developed to evaluate uncertain threats distribution. A flying height estimating algorithm is presented to improve traditional terrain smoothing methods. This presented algorithm can simplify the three-dimensional planning problem into a two-dimensional form. Both terminal and time mission constraints are considered, along with aircraft maneuvering characteristics and cooperative constraints. Furthermore, a theoretical model of cooperative path planning is established.2. A novel path planning method is proposed based on the improved ant colony algorithm (ACA). A diversity choice operator is designed for the characteristics of premature convergence in ACA. This operator considers both path cost value and space distribution of a solution in the process of pheromone updating, which guarantees population diversity. A bidirectional searching mechanism and a Hopfield neural network based flight time estimating method is respectively developed and integrated into the ACA framework to complement the shortage of ACA in solving problems of path planning with terminal and time mission constraints.3. A coevolution based problem solving method is developed for cooperative path planning of multiple aircrafts. First of all, a coevolution based solving framework is constructed for cooperative path planning. This framework can divide the high dimensional and large-scale cooperative path planning problem into several low dimensional and small-scale single aircraft path planning sub-problems. Cooperation and iterative optimization are conducted among all the sub-problems to maintain the optimal or sub-optimal solutions. The cooperation between each sub-problem includes cooperative time constraint, cooperative space constraint and cooperative path cost evaluation. Eventually, a multiple ant colony coevolution algorithm is designed to enable each ant colony to plan an individual path through the evolution and coevolution between different colonies.The work in this thesis provides a method to solve the cooperative path planning problem. Moreover, the proposed and developed methods and algorithms have partially been applied into a related high-tech project as an important sub-module.