Enhancing target search efficiency of centralized and distributed intelligence in UAV swarm operations

• Main Author: Xu, Xiaotian
• Other Authors: Chau Yuen
• Format: Thesis-Master by Coursework
• Language: English
• Published: Nanyang Technological University 2024
• Subjects: Engineering; UAV; Swarm intelligence; Tracking and seeking; Centralized intelligence; Distributed intelligence
• Online Access: https://hdl.handle.net/10356/175437
• Institution: Nanyang Technological University

This dissertation tackles the challenge of improving the efficiency of search and track maneuvers for static targets in multi-Unmanned Aerial Vehicle (UAV) systems. Key objectives include developing advanced control algorithms, implementing autonomous dynamics using the steering force method, integrating collision prevention mechanisms via a potential field model, and adapting formation control in dynamic environments. Focused enhancements of the Rowscan algorithm, including Distributed Search (DS), Centralized Search (CS), Distributed Map-known Search (DMS), Centralized Map-known Search (CMS), and known-Drones’ Positions (DP) algorithms, are explored along with evaluating shared exploration areas, considering known map sizes, and establishing mutual orientation information among drones. The methodology combines distributed and centralized decision-making frameworks to empower UAVs with enhanced autonomy during search missions. Through rigorous algorithm design, simulation validation, and comparative analysis, this research demonstrates significant improvements in search efficiency and target localization. The findings contribute to robotics research, addressing contemporary challenges in multirobot and swarm systems with implications for search and rescue, environmental monitoring, and surveillance applications.