A Novel Classified Self-Organising Map Applied to Task Assignment. (27th September 2020)

• Record Type: Journal Article
• Authors: Qu, Yun; Zhu, Daqi
• Is Part Of: Journal of navigation. Volume 73:Number 5(2020)
• Page Start: 1129
• Page End: 1145
• Publication Date: 2020-09-27
• Subjects: Autonomous Underwater Vehicles (AUVs), -- Classified Self-Organising Map, -- Matched, -- Order; Navigation -- Periodicals; 623.8905
• DOI: 10.1017/S037346332000020X
• Languages: English
• ISSNs: 0373-4633

Abstract

Abstract : With the development of sensor technology, sensor nodes are increasingly being used in underwater environments. The strategy presented in this paper is designed to solve the problem of using a limited number of autonomous underwater vehicles (AUVs) to complete tasks such as data collection from sensor nodes when the number of AUVs is less than the number of target sensors. A novel classified self-organising map algorithm is proposed to solve the problem. First, according to the K-means algorithm, targets are classified into groups that are determined by the number of AUVs. Second, according to the self-organising map algorithm, AUVs are matched with groups. Third, each AUV is provided with the accessible order of the targets in the group. The novel classified self-organising map algorithm can be used not only to reduce the total energy consumption in a multi-AUV system, but also to give the most efficient accessible order of targets for AUVs. Results of simulations conducted to prove the applicability of the algorithm are given.