Energy Efficient Data Evacuation Path for Multi-UAV System Based on Multi-Objective Optimization Method
Linbo Zhai, Xiumin Zhu, Cheng Chen
Abstract
To avoid the loss of data collected by unmanned aerial vehicles (UAVs) in multi-UAV system, the collected data should be transmitted to ground station for processing in time. Due to insufficient energy and computation resources of UAVs, an interactive multiple data evacuation paths problem of multi-UAV system is a challenge. In this paper, we study energy-saving and time-saving evacuation path for collected data. We formulate a Time and Energy Saving Multi-objective Path Optimization Problem (TESMPOP) to minimize the total transmission time of data and energy consumption of relay UAVs. Because of the NP-hardness of formulated TESMPOP, we propose a Data Temporary Staging Point Selection and Multiple Data Evacuation Paths Ant Colony System (DTSPS-MDEPACS) algorithm. Based on load balance of relay UAV and time constraints of data UAV, DTSPS algorithm is designed to select the suitable relay UAVs to temporarily store data to avoid the loss of data. Then, the MDEPACS algorithm is proposed to find the multiple data evacuation paths for data from the selected relay UAVs to base station. In MDEPACS algorithm, we design hybrid data transmission mode for relay UAV. Each relay UAV's transmission mode can be different. A novel solution construction which includes the mode and trajectory of relay UAVs is proposed. A new pheromone update rule based on nondominated solutions is designed to ensure that ant colony can quickly find the best evacuation paths. Simulation results illustrate that the proposed DTSPS-MDEPACS algorithm has better performance than some other benchmark methods.