Unscented Particle Filter Algorithm Based on Divide-and-Conquer Sampling for Target Tracking

An Improved Particle Swarm Optimization Algorithm and Its Application to the Extreme Value Optimization Problem of Multivariable Function

It is proposed to improve the study of particle optimization and its application in order to solve the problem of inefficiency and lack of local optimization skills in the use of particle herd optimization. Firstly, the basic principle, mathematical description, algorithm parameters, and flow of the original (Particle Swarm Optimization, PSO) algorithm are introduced, and then the standard PSO algorithm is introduced; thirdly, over the last 10 years, four types of improvements have been proposed through the study of improved particle algorithms. The improved algorithm is applied to the extreme value optimization problem of multivariable function. The simulation results show that the basic (Cloud Particle Swarm Optimization, CPSO) algorithm within 500 generations has not reached convergence for 8 times, 6 times, 4 times, and 5 times, respectively. In the case of convergence, the average number of steps is much higher than ICPSO, and the improved algorithm converges completely. In terms of time performance, the convergence time of ICPSO is much better than that of CPSO algorithm. Therefore, the improved particle optimization algorithm ensures the effectiveness of the improvement measures, such as small optimization algebra, fast merging speed, high efficiency, and good population diversity.