Please use this identifier to cite or link to this item:
Title: An MINLP Formulation for Multi-UAV Cooperative Search Path Planning Problem
Keywords: multi-UAV
cooperative search
path planning
obstacle avoidance
mixed-integer nonlinear optimization
Issue Date: 2008
Citation: ZHANG YIFENG (2008). An MINLP Formulation for Multi-UAV Cooperative Search Path Planning Problem. ScholarBank@NUS Repository.
Abstract: There has been an increasing effort of research in the broad area of Unmanned Aerial Vehicles (UAV) for military as well as civilian applications in recent years. While a single UAV is capable of missions such as reconnaissance and surveillance operations, target searching, and target tracking, cooperative control of a fleet of UAVs is believed to be able to accomplish the missions more effectively. This thesis addresses a multi-UAV cooperative search problem whereby a fleet of UAVs are dispatched to search for targets of interest in an uncertain environment. We propose a Mixed-Integer Non-Linear Programming (MINLP) formulation for the problem at hand. Various features related to the multi-UAV cooperative search problems are incorporated. These include collision avoidance among the UAVs as well as with the surrounding obstacles, adaptive search, whereby earlier search result can be used to influence future decisions on path selection, cooperative control which aims at achieving good team performance rather than individual UAV performance. We also model UAV dynamics using simple logic. We present a few numerical examples to show that this MINLP formulation is flexible and capable of top-level path planning for the cooperative search problem. We also discuss the limitation of such an approach in terms of memory and computational requirements, and make suggestions for future work.
Appears in Collections:Master's Theses (Restricted)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
Zhang Yifeng_SMA_Thesis_ZhangYifeng.pdf453.04 kBAdobe PDF


NoneLog In

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.