Multi-UAV cooperative system for search and rescue based on YOLOv5

Wilderness search and rescue (WiSAR) usually need to search a large and complex area and has high speed requirements. When an unmanned aerial vehicle (UAV) is used for auxiliary operation, it usually needs to carry a variety of sensors, and it is difficult to ensure real-time performance. In recent years, through the continuous development of target detection algorithms based on deep learning, the accuracy and real-time performance of target detection for visible images are getting better. This study aims to shorten the time of search and rescue (SAR) and improve the success rate by using UAV and target detection technology. In this study, a multi-UAV cooperative system for SAR based on YOLOv5 is proposed. The system has these functions of free-grafting multiple UAVs, independent control of each UAV, real-time target detection, monocular positioning. The performance of this system is tested through several simulated search and rescue missions, and the results prove that the system can meet the requirements of search and rescue operations.