A multi-UAV charging and task scheduling method based on deep reinforcement learning

Abstract

A multi-UAV charging and task scheduling method based on deep reinforcement learning relates to the technical field of UAV scheduling, and solves the problem that the load of the charging station and the influence of the UAV's charging strategy on the task scheduling are not considered in the prior art, and the method It is: according to the tasks to be performed and the load of the charging station, the schedulable UAVs are scheduled through the deep reinforcement learning model; after the UAVs perform the tasks, according to the number of unexecuted tasks, the number of schedulable UAVs and unmanned aerial vehicles The remaining power of the drone is used to charge the drone to be charged that is staying on the charging station. The invention effectively solves the problem that the charging of multiple drones and task scheduling need to be optimized and improved, and can achieve the goal of minimizing the overall time for executing tasks on the premise of ensuring that the drones will not run out of energy, and finally obtain each drone. Therefore, the corresponding multi-UAVs can traverse these task points in order from the starting point, and perform adaptive charging at the corresponding charging station.

Description

technical field [0001] The invention relates to the technical field of UAV scheduling, in particular to a multi-UAV charging and task scheduling method based on deep reinforcement learning. Background technique [0002] The current methods for multi-UAV charging and task scheduling mainly include heuristic algorithms and algorithms based on reinforcement learning. Heuristic algorithms generally optimize the charging and task scheduling of UAVs through human-designed rules. The disadvantage of this method is that it requires artificial design rules. When the problem is complex and multiple factors need to be considered, this rule often cannot be effectively designed, and the obtained dissociated optimal solution is far away. Based on this, the heuristic algorithm that introduces the exchange operator continuously updates the solution after obtaining the initial solution to obtain a better solution, but this will lead to an increase in time complexity and is not suitable for ...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the problem that the existing multi-UAV charging and task scheduling method still needs to be improved, the present invention provides a multi-UAV charging and task scheduling method based on deep reinforcement learning

Images