Abstract:

Aiming at the problems of the population diversity reduction in the late evolution of particle swarm optimization (PSO),and the information exchange error of PSO-DE,this paper presents a PSO-DE algorithm based on the optimal selection strategy.First,a weighted network (WN) is constructed to calculate the systematic biases.Then the optimal selection strategy mechanism is introduced,and the fitness function is constructed as an evaluation criterion.Finally,the systematic deviation estimate is used to register the target sensor measurement.In the test of population diversity and fitness,the algorithm proposed in the paper has a richer population diversity,and the optimal fitness value of the individual is 2.0194×10^-5.In the experiments on non-maneuvering and maneuvering targets,the deviation value rapidly converges to the true deviation value after about 2s,with the shortest convergence time being 201.8s,and the RMS error value is reduced by more than 10 times.Simulation results show that the algorithm not only increases the population diversity,but also improves the convergence speed and the accuracy.

Key words: systematic biases registration, weight networks, optimal selection strategy, differential evolution particle swarm optimization algorithm