ABSTRACT
Glowworm swarm optimization (GSO) algorithm is a swarm intelligence algorithm, which follows the behavior of lightning worms in order to solve complex mathematical problems. The algorithm was introduced by Krishnanand and Ghose [1]. GSO algorithm is very efcient in solving multimodal equations and has been used in many applications such as clustering, routing, swarm robotics, image processing, localization, and so on. GSO works on the mechanism of the physical behavior of insects called glowworms. The agents for the solution are seen to as glowworms, which contain luciferin, a light emitting compound found in many rey species. The glowworms calculate the tness of their current position in the solution space; get to use the objective function into a luciferin value, which they then broadcast to all their neighbors. The glowworm nds its neighbors and then calculates its position changes by manipulating its adaptive neighborhood, which is bordered by its sensor range. Every glowworm gets attracted by the brighter glow of their other neighboring glowworms and they then select, using a probabilistic method, a neighbor that has a luciferin value more than its own and moves in the direction of it. The higher the intensity of luciferin, the better is the location of glowworm in the search space. These movements-derived only from the local information and selective neighbor interaction enables the glowworm swarm to divide into disjoint subgroups that converge on multiple peaks of a given multimodal function. This algorithm is considered similar to that of other swarm intelligent solutions such as particle swarm optimization (PSO) and ant colony optimization (ACO) as discussed in the previous chapters but there are several notable differences in its working from the PSO:
• The velocity update equation of PSO includes a memory element, whereas in GSO no information is taken from the memory.
