Dragonfly Algorithm for Multi-target Search Problem in Swarm Robotic with Dynamic Environment Size

• Published in: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
• Main Author: Hamami M.G.M.; Ismail Z.H.
• Format: Conference paper
• Language: English
• Published: Springer Science and Business Media Deutschland GmbH 2022
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85137986571&doi=10.1007%2f978-3-031-08530-7_21&partnerID=40&md5=51e0b70c2c4bb14e77fe3200fb51cbc3

Target search elements are very important in real-world applications such as post-disaster search and rescue missions, and pollution detection. In such situations, there will be time limitations, especially under a dynamic environment size which makes multi-target search problems are more demanding and need a special approach and intention. To answer this need, a proposed multi-target search strategy, based on Dragonfly Algorithm (DA) has been presented in this paper for a Swarm Robotic application. The proposed strategy utilized the DA static swarm (food hunting process) and dynamic swarm (migration process) to achieve the optimized balance between the exploration and exploitation phases during the multi-target search process. For performance evaluation, numerical simulations have been done and the initial results of the proposed strategy show more stability and efficiency than the previous works. © 2022, Springer Nature Switzerland AG.