Fireworks algorithm based on search space partition

Abstract

Fireworks algorithm is a novel swarm intelligence optimization framework which focuses on the potential of collaboration among multiple subpopulations with independent search ability. Although it has been proved to perform excellently in many tasks, the collaborative mechanism of fireworks is still quite undeveloped. In this paper, a theoretical model of fireworks algorithm based on search space partition is proposed, analyzed and implemented. The local search of each firework is replaced by covariance matrix adaptation evolution strategy for efficient exploitation. A coordination strategy inspired from Voronoi diagram is proposed to approximate the theoretical model for stable global exploration. Experimental results show that the proposed algorithm not only outperforms previous variants of fireworks algorithm significantly, but also achieves competitive results compared with state‐of‐the‐art evolutionary algorithms, which are intensively fine‐tuned on the objective functions.

Research on a local path planning algorithm based on multivehicle collaborative mapping and a potential field method

“…The collaborative mechanism of the Firework Algorithm is relatively underdeveloped. The theoretical model of the Firework Algorithm based on search space partition replaces the local search of each firework with the covariance matrix-adapted evolutionary strategy, aiming to achieve efficient development [31]. Shoufei han et al [32] make improvements by incorporating an adaptive strategy and special archive guidance into it, where special archives are established for each firework, and two strategies (i.e.…”

“…The collaborative mechanism of the Firework Algorithm is relatively underdeveloped. The theoretical model of the Firework Algorithm based on search space partition replaces the local search of each firework with the covariance matrix‐adapted evolutionary strategy, aiming to achieve efficient development [31]. Shoufei han et al.…”