ABSTRACT
The advancement and research in artificial intelligence (AI) and evolutionary computing have led to the invention of swarm intelligence (SI) techniques. SI works on the analogy of collective and collaborative behavior of the social species in nature. Self-organization, global ordering, and coordination, which result from interactive communication in the swarm, are the key features of the SI system, which is exploited for optimization in computational problems. Particle swarm optimization (PSO) is one of the most popular SI-based optimization algorithms inspired by social behavior observed in animals and insects. Each member in the swarm is treated as a particle. The collective intelligence of these decentralized particles is attained by cooperation and learning. For decades, PSO has been considered as one of the most efficient optimization techniques because of its simplicity in implementation, the number of parameters used for manipulation, quick convergence, and scalability. The main objective of this chapter is to provide a comprehensive investigation of PSO and its variants like unified PSO, memetic PSO, vector evaluated PSO, composite PSO to name a few. The chapter also focuses on the relevance of PSO in real-world applications.
