Cooperative Detection-Oriented Formation Design and Optimization of USV Swarms via an Improved Genetic Algorithm

• Published in: Sensors (Basel, Switzerland) Vol. 25; no. 10; p. 3179
• Main Authors: Liang, Rui, Li, Dingzhao, Sun, Haixin, Hong, Liangpo
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 18.05.2025; MDPI
• Subjects: Algorithms; Analysis; Autonomous underwater vehicles; Collaboration; Communication; Efficiency; Energy consumption; Evacuations & rescues; formation optimization; genetic algorithm; Genetic algorithms; marine environment; Marine pollution; Mathematical models; multi-objective optimization; Optimization; Probability; Sensors; Signal processing; unmanned surface vehicle (USV); Vehicles; genetic algorithm; formation optimization; unmanned surface vehicle (USV); multi-objective optimization; marine environment
• ISSN: 1424-8220; 1424-8220
• DOI: 10.3390/s25103179

Efficient and adaptive formation planning is critical for unmanned surface vehicle (USV) swarms equipped with sensor networks and smart sensors to perform cooperative detection tasks in complex marine environments. Existing formation optimization methods often overlook the nonlinear coupling between sensor-based detection performance, communication constraints, and obstacle avoidance. We propose a multi-objective formation optimization framework based on an improved genetic algorithm that simultaneously considers the detection coverage area, forward detection width, inter-agent communication, and static obstacle avoidance. We formulate a probabilistic cooperative detection model, introduce normalized detection efficiency indicators, and embed multiple geometric and environmental constraints into the optimization process. Simulation results show that the proposed method significantly improves the spatial efficiency of cooperative sensing, yielding a 32.76% increase in effective coverage area and 20.97% improvement in forward detection width compared to unoptimized formations. This strategy, supported by multi-sensor positioning and navigation, offers a robust and generalizable approach for intelligent maritime USV deployment in dynamic, multi-constraint scenarios.