The IoT Data Collection by UAV, Based on Multi-Objective Evolutionary Algorithm Using Gaussian Process Based on Reverse Modeling

Today UAVs are used as data collection platforms for a group of IoT devices around the world. Determining the optimal number and locations of UAVs can minimize the energy consumption of the IoT data collection system. The use of modern multi-objective optimization algorithms can achieve this goal. In this research, a model-based multi-objective evolutionary optimization algorithm (multi-objective evolutionary algorithm using Gaussian process based on reverse modeling) is used on the data collection platform for a group of IoT devices around the earth. The results of this method have been compared with other evolutionary multi-objective optimization algorithms based on the inverse generation distance evaluation criterion. The analysis of the results of the evaluation criterion used shows that the optimization method used has a significant impact on the process of optimal energy consumption in this system. The results of this research can be used for identification of the optimization parameters in this system.