Multi-unmanned aerial vehicle cooperation sequential coupling task distribution method of mixing gravitation search algorithm

Abstract

The present invention provides a multi-unmanned aerial vehicle cooperation sequential coupling task distribution method of a mixing gravitation search algorithm, and relates to the unmanned aerial vehicle cooperation task distribution field. The method comprises: a multi-unmanned aerial vehicle cooperation task distribution model is constructed in the time coupling constraint, a fitness function and a task constraint are obtained, in the gravitation search algorithm based on genetic operators, the individual discretization coding and the population are initialized, the individual is decoded, and the fitness function is employed to calculate the fitness and perform individual update. Because the genetic operators are added in the gravitation search algorithm, the multi-unmanned aerial vehicle cooperation sequential coupling task distribution method of the mixing gravitation search algorithm has good general applicability, the number of times of long-term simulation tests and data statistics constructs a more improved database to allow the model to be more improved; and compared to the discrete particle swarm algorithm, the mixing gravitation search algorithm can be rapidly converged, the searching optimization result is optimal, the iteration process is brief, and the convergence speed is fast.

Description

technical field [0001] The invention relates to the field of unmanned aerial vehicle (UAV) collaborative task assignment, in particular to a multi-unmanned aerial vehicle task assignment method under time coupling constraints. Background technique [0002] Multi-UAV cooperative timing coupling task assignment plays a very important role in scientific research and engineering applications. The traditional gravitational search algorithm has outstanding global optimization ability in solving such problems, but also has many defects, such as easy to fall into local optimum, and the quality of the global optimal particle is low. The essence of the Gravitation Search Algorithm (GSA) is to simulate the gravitational phenomenon in nature and evolve it into a process of randomly searching for the optimal solution. [0003] Due to the calculation and search of global optimal particles (selection of guided particles) in multi-UAV cooperative timing coupling task assignment, it has an ...

AI Technical Summary

Problems solved by technology

[0005] However, when the gravitational search algorithm is applied to multi-UAV cooperative timing coupling task allocation, some of its own shortcomings lead to low quality of the global optimal particle in the algorithm, and the effect of timing coupling task allocation needs to be improved

First of all, in the gravitational search algorithm, only the current position information plays a role in the iterative update process, that is, the algorithm is an algorithm that lacks memory, which leads to no information exchange between the upper and lower generations of the population, and it is easy to fall into premature convergence

On the other hand, due to the high particle speed in the gravitational search algorithm, all of them move towards the particles with large mass, and the convergence is very fast, so the diversity of the population decreases rapidly, that is, the diversity is lost rapidly, and the diversity of non-dominated solutions cannot be guaranteed. and distribution

Images