Drone's trajectory optimization for searching a disaster collapsed area using data mining techniques

Drones have unique characteristics such as mobility, flexibility, and role adaptability. Hence, they are useful tools during urban search and rescue (USAR) operations in areas where many buildings have suffered structural collapse leading to difficulties in reaching locations where people may be trapped. However, drones are limited by the life of the batteries allowing their flight. Clearly, reducing flight time to useful destinations can help to optimize the use of drones with limited battery life. Therefore, this study considers trajectory optimization as a critical research problem for USAR applications. This paper explores optimal trajectory paths for USAR operations for a single drone by selecting nodes within a disaster area from the start point to the destination point for the drone's flight. The following analysis applies density-based spatial clustering of applications with noise (DBSCAN) data clustering algorithm to the dataset obtained from the scanning from the top of the disaster collapse area. The goal is to find the optimized shortest paths for USAR drone operations in order to conserve drone flight time. The computational results are compared to physical experimental results.