A symbiosis-based artificial fish swarm algorithm

This paper presents a symbiosis-based artificial fish swarm algorithm (AFSA), which employs two artificial fish (AF) populations to collaboratively search the solution space of the problem to be solved. The symbiosis-based framework makes AF individuals adaptively adjust their Visual-limit and Step-limit. Due to the irrationality that AF individual swims toward its partners´ geometric centre when gathering, a weighted centre and the method for calculating it was proposed. Beyond this, this paper also presents a newly designed behavioral strategy for AF. Experiments were conducted on several test functions. The results demonstrate good performance of the symbiosis-based AFSA when compared with several other swarm intelligence-based algorithms.