A Decision Support System for Dynamic Heterogeneous Unmanned Aerial System Fleets


Ercan C., Gencer C.

GAZI UNIVERSITY JOURNAL OF SCIENCE, cilt.31, ss.863-877, 2018 (ESCI) identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 31
  • Basım Tarihi: 2018
  • Dergi Adı: GAZI UNIVERSITY JOURNAL OF SCIENCE
  • Derginin Tarandığı İndeksler: Emerging Sources Citation Index (ESCI), Scopus, TR DİZİN (ULAKBİM)
  • Sayfa Sayıları: ss.863-877
  • Gazi Üniversitesi Adresli: Evet

Özet

The Dynamic Unmanned Aerial System Routing Problem (DUASRP) is a variant of the classic Vehicle Routing Problem (VRP) in which both planned and unplanned targets are observed by a fleet of Unmanned Aerial Systems (UASs). In the dynamic environment of UAS, the rapid response for the new important targets is a very critical process, especially for the military operations in battle space conditions. This study describes a heuristic method for the solution of the dynamic heterogeneous UAS routing problems without causing the initial tour to be completely changed. For the dynamic routing of Unmanned Aerial Vehicles (UAV), it is necessary to determine a combination of the least additional costs of vehicle routes through a set of geographically scattered targets, and quick responses for immediate targets during the reconnaissance missions. The most frequent cases assumed in the existing literature of classical DUASs consider all UASs as identical (homogenous), all targets as having two geographical coordinates, and the thread of the targets are ignored. In this paper, a dynamic routing decision support system based on both fuzzy clustering and leveraged cheapest insertion neighborhood method is studied for pop-up threat in the case where the UAV fleet is heterogeneous, and targets have both three-dimensional information and threads. Instead of selecting an a priori code, the proposed control methodology dynamically starts with the route based on observed behavior of the new target and the routes. It describes an efficient heuristic method capable of producing quick dynamic solutions on a series of empirical test problems.