Journal of Bionic Engineering ›› 2025, Vol. 22 ›› Issue (5): 2646-2659.doi: 10.1007/s42235-025-00766-w
A Minimalistic and Decentralised Approach to Formation Control for Crowded UUV Swarms Inspired by Fish Schooling
Xuhang Wu1,2,3,4; Xiangyang Deng5,6; Bang Wen1,2,3; Shengzhi Yue1,2,3; Siyang Shao1,2,3; Fumin Zhang7; Fang Wang1,2,3; Yuanshan Lin1,2,3
-
1 School of Information Engineering, Dalian Ocean University,Dalian 116023, China
2 Dalian Key Laboratory of Smart Fisheries, Dalian116023, China
3 Key Laboratory of Environment Controlled Aquaculture(Dalian Ocean University) Ministry of Education,Dalian 116023, China 4 Nanjing NARI Information and Communication TechnologyCo., Ltd, Nanjing 210031, China 5 Naval Aeronautical University, Yantai 264001, China 6 Department of Automation, Tsinghua University,Beijing 100084, China
7 Cheng Kar-Shun Robotics Institute, Hong Kong Universityof Science and Technology, Clear Water Bay, Hong Kong,China
A Minimalistic and Decentralised Approach to Formation Control for Crowded UUV Swarms Inspired by Fish Schooling
Xuhang Wu1,2,3,4; Xiangyang Deng5,6; Bang Wen1,2,3; Shengzhi Yue1,2,3; Siyang Shao1,2,3; Fumin Zhang7; Fang Wang1,2,3; Yuanshan Lin1,2,3
-
1 School of Information Engineering, Dalian Ocean University,Dalian 116023, China
2 Dalian Key Laboratory of Smart Fisheries, Dalian116023, China
3 Key Laboratory of Environment Controlled Aquaculture(Dalian Ocean University) Ministry of Education,Dalian 116023, China 4 Nanjing NARI Information and Communication TechnologyCo., Ltd, Nanjing 210031, China 5 Naval Aeronautical University, Yantai 264001, China 6 Department of Automation, Tsinghua University,Beijing 100084, China
7 Cheng Kar-Shun Robotics Institute, Hong Kong Universityof Science and Technology, Clear Water Bay, Hong Kong,China
摘要： Formation control remains a critical challenge in cooperative multi-agent systems, particularly for Unmanned Underwater Vehicles (UUVs). Conventional approaches often suffer from several limitations, including reliance on global information, limited adaptability, high computational complexity, and poor scalability. To address these issues, we propose a novel bio-inspired formation control method for UUV swarms, drawing inspiration from the self-organizing behavior of fish schools. Our method integrates three key components: (1) a coordinated motion strategy without predefined targets that enables individual UUVs to align their movements via simple left or right rotations based solely on local neighbor interactions; (2) a target-directed movement strategy that guides UUVs toward specified regions; and (3) a dispersion control strategy that prevents overcrowding by regulating local spatial distributions. Simulation results confirm that the method achieves robust formation control and efficient area coverage using only local perception. Validation in a 9-UUV simulation environment demonstrates the approach’s flexibility, decentralization, and computational efficiency, making it particularly suitable for large-scale swarms with limited sensing and processing capabilities.