Robotic systems have become essential tools in disaster response, enabling rapid, safe, and persistent operations in environments that are often inaccessible or hazardous to humans. Among these systems, unmanned aerial vehicles (UAVs) play a particularly critical role in search and rescue (SAR) missions, providing wide-area surveillance, rapid deployment, and real-time situational awareness. Several recent studies have addressed the challenges of multi-UAV coordination and multi-modal search strategies in SAR operations. However, these approaches primarily focus on optimizing UAV trajectories across discretized altitude levels, without explicitly modeling biologically inspired behaviors or integrating real-time altitude-role specialization into sensing and decision-making. To overcome the above issues, this work proposes the development and testing of a coordinated multi-drone approach to achieve efficiency and robustness in such missions. The framework utilizes particle swarm-based path planning to synchronize high-altitude surveying with low-altitude refinement, coordinating thermal and visual sensors to enhance classification performance. Four scenarios are presented, varying in complexity levels, ranging from the simplest, such as detecting humans, to scenes with animal-like thermal signatures for the targets. The method has proven effective in achieving fast coverage, effective trajectory distribution, and stable victim detection. In particular, the low-altitude refinement enhanced the localization precision, while the hybrid sensing was capable of alleviating false positives. The mission time was increased in cases where the environment is more complex, but the overall accuracy and robustness are significantly enhanced. The conclusion is that the coordinated multi-drone approach has great potential for developing SAR missions. Furthermore, the work demonstrates that bio-inspired multi-UAV coordination, combined with multimodal sensor fusion, can significantly enhance the scalability, reliability, and spatial accuracy of SAR operations. The presented framework provides an excellent foundation for future implementation in the context of convention-based disaster response missions.
Kareem et al. (Tue,) studied this question.