Unmanned aerial vehicle control method and system based on multi-agent deep reinforcement learning

Abstract

The invention provides an unmanned aerial vehicle control method and system based on multi-agent deep reinforcement learning. The unmanned aerial vehicle control method comprises the steps of: establishing an information acquisition task model according to parameters of an unmanned aerial vehicle group information acquisition system, wherein an information acquisition task is divided into an acquisition subtask and a calculation subtask; constructing a deep neural network model according to the task model, and training the deep neural network model by using a multi-agent deep reinforcement learning algorithm combined with an attention mechanism; and controlling an unmanned aerial vehicle group in an actual environment to complete an information acquisition task by using the trained deep neural network model. According to the unmanned aerial vehicle control method and the system, each unmanned aerial vehicle is used as an intelligent agent, a critic network with an attention unit is used for evaluating the performance of an actor network, and the training speed of the actor network can be accelerated with a more accurate evaluation value; and when an information acquisition task isexecuted, each unmanned aerial vehicle does not need to communicate with other intelligent agents, so that the communication time delay is reduced.

Description

technical field [0001] The present invention relates to the technical field of wireless communication, in particular to a control method and system for unmanned aerial vehicles based on multi-agent deep reinforcement learning. Background technique [0002] Unmanned Aerial Vehicles (UAV) is an unmanned aircraft that is remotely controlled by an operator through a radio remote control device or automatically controlled by a computer program. Most of the applications of UAVs are information collection tasks. In the prior art, the control instructions for multi-UAV system data collection tasks are mainly solved by two methods, namely the heuristic method and the method based on machine learning. [0003] Among them, the heuristic algorithm needs to go through multiple rounds of calculations after receiving the task to get the best information collection and calculation migration plan, resulting in a large time delay, which is not conducive to some urgent tasks; the depth enhance...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a UAV control method and system based on multi-agent deep reinforcement learning to solve the problem that UAV systems are difficult to perform in a short time delay when facing various complex tasks and environments. technical issues

Images