PSO based on chaotic map and its application to PID controller self-tuning

As a kind of iterative learning algorithm, PSO algorithm is analogous to the stochastic behaviors of creatures in nature for foraging such as birds and fish, through self-learning strategies and synergy of swarm to determine their searching directions. In order to strengthen diversity and searching ergodicity of particles, this paper proposed an initial method of adaptive inertia weight based on chaotic map and proved the swarm's convergence is prior to stochastic initialization by embedding in three common improved PSOs with test of three benchmark functions. The proposed algorithm is applied to self-turn a PID controller which is widely used in precise positioning realms such as electronic packing technology subsequently. The outperformed performance of MSPO based on chaotic map is calculated and verified by simulated results.