Dynamic flight path planning method based on adaptive mutation genetic algorithm

Abstract

The invention discloses a dynamic flight path planning method based on adaptive mutation genetic algorithm, comprising the steps of: generating an initial species group at random and calculating the degree of adaptability thereof; counting feasible solutions E(t) in the species group and performing elite preservation operation; performing intersection, mutation, insertion and deleting operations on non-feasible solutions S(t); calculating the degree of adaptability of a transitional species group composed of the operated E(t) and S(t); obtaining individuals and the feasible solutions generated at random in the individuals by E(t) single point random mutation in order to replace the non-feasible solutions in the transitional species group; judging whether the evolution reaches the extent that an algebra selected according to the complexity of problem or the species group is converged; and if so, taking the optimal invividual in the species group as the planned flight path. The method of the invention can improve the variety and convergence capacity of the species groups; a feasible solution path can be provided for the problem of time-varying dynamic optimization for the dynamic flight path planning.

Description

technical field [0001] The invention relates to the technical field of dynamic optimization and the technical field of aircraft path planning. Background technique [0002] Dynamic flight path planning refers to real-time flight path planning when there are motion obstacles in the environment, or there are obstacles unknown in advance. Since the external environment changes with time, dynamic flight path planning is actually a dynamic optimization problem. Solving dynamic optimization problems with genetic algorithms is a challenging task. [0003] Among the existing methods for solving dynamic optimization problems based on genetic algorithms, many of them use the stochastic transition method and the elite mutation method. The random change method is to replace some of the original individuals in the population by randomly generated individuals to increase the diversity of the population. The elite mutation method is to use the elite individuals in the previous generatio...

AI Technical Summary

Problems solved by technology

[0006] In order to overcome the deficiency that the existing technology is not suitable for solving the problem of dynamic flight path planning, the present invention provides a dynamic flight path planning method based on adaptive mutation genetic algorithm, which can adaptively transform random transitions according to the number of feasible solutions in the current population Combining with elite mutation dynamics, it improves the diversity and convergence ability of the population, and provides a feasible solution to the time-varying dynamic optimization problem such as dynamic flight path planning

Images