Multi-objective resource optimization of unmanned aerial vehicles and internet of things devices in time critical applications using evolutionary approaches

In this thesis, investigations are done on the use of UAVs to support IoT devices on the ground in time critical applications. Due to advancements in aerial and sensor technology, a team of UAVs can be used to bring about a collaborative effort in supporting IoT devices due to their flexibility and versatility. However, both UAVs and ground IoT devices have limited energy resources since they are powered by battery and as a result, there exists a need to effectively manage their operation. The first part of the investigation involves the use of UAVs to provide aerial services such as video or image data to get more information about the situation on the ground. A multi-parameter encoded model based on a multi-objective optimization algorithm is developed to generate a set of 3D waypoints for each UAV to visit within an expected time frame using variable speeds to maximize service provision without violating any mission constraints and also minimizes the total energy consumed and response time by the UAV team. The proposed model is able to give solutions that improve the energy consumption by 8%. The second part of the investigation involves the use of UAVs as aerial base stations to collect data from IoT devices in the absence of critical communication infrastructure. Another model employing an improved hybrid multi-objective optimization algorithm, is developed to jointly determine the UAV-IoT device association, UAV locations, UAV speeds and hovering durations and the transmit power used by the IoT devices when sending data. The aim is to minimize the energy consumption of both the UAVs and IoT devices as well as maximize the data collected by UAVs. An improvement of 12% is seen in the amount of data collected when compared to similar approaches.