查看文章

Many swarm intelligence algorithms study the behavior of animals and make animal-based systems for solving various tasks. Cat swarm optimization (CSO) is a swarm intelligence algorithm which was originally inspired by the resting and tracking behaviors observed naturally in cats. The N-Queens problem is a classical and complex constraint satisfaction problem which has been used as a benchmark for testing AI techniques for years. This paper aims to adapt the original CSO algorithm to the N-Queens problem by replacing the continuous addition/subtraction operations with swapping operations. The results report very fast convergence at smaller numbers of queens and point towards quicker convergence than similar work at larger numbers of queens, with improvements still needed for perfecting the algorithm.