Cooperative Multi-Agent UAS Task Assignment for Disaster Response Scenario
Nicholas S. DeGroote, Kelly De Oliveira Cohen · AIAA SCITECH 2022 Forum · 2022
View Video Presentation: https://doi.org/10.2514/6.2022-0792.vid Task assignment problems have evolved as computation methods and resources have allowed, utilizing Unmanned Aerial Systems (UAS) as a resource for accomplishing a variety of complex tasks. The ability of UAS to operate in complex environments, cover large areas, and be deployed quickly opens the door for a variety of applications, with one such being disaster response. This research is inspired by a scenario where UAS are used to help dispatch aid to a region after a natural disaster such as a hurricane. A heterogeneous fleet of both fixed-wing and multirotor UAS is available to be allocated to the region of interest, with the goal of accomplishing all known tasks in the minimum amount of time. The multiple-depot multiple travelling salesman problem (MDMTSP) is used as a benchmark to show the effectiveness of the developed algorithm. Several additional constraints are then added to the MDMTSP to better represent the challenges of disaster response task assignment including search tasks, heterogeneous sensors, fuel constraints, and Dubins vehicles. Ongoing research focuses on applying the solution in a hardware integration and performing a real-world flight test demonstration.