A route planning method for multi-UAV cooperative area surveillance based on DQN

Abstract

The invention belongs to the field of multi-UAV regional monitoring route planning, and discloses a DQN-based route planning method for multi-UAV regional monitoring, including: setting the target monitoring region of the UAV group, including N UAVs, Set an airborne radar on each drone, and each drone flies at a constant speed; model the route planning of the drone group; determine the state and action of the drone; establish a deep neural network; get the target value of the neural network ; Plan the route of N drones, get a limited number of training samples during the flight of the drone group, and train the deep neural network through the samples. The trained neural network can predict the samples that have not been seen. When the current state of the UAV swarm is input, the neural network can be used to know which action to take to fly, and the monitoring coverage of the next step will be the largest. The route planned in this way can enable the UAV group to effectively cover the surveillance area in real time.

Description

technical field [0001] The invention belongs to the field of UAV regional monitoring route planning, and in particular relates to a DQN (Deep Q-Network)-based route planning method for multi-UAV collaborative regional monitoring, which is suitable for UAV groups to perform real-time maximum coverage of target areas Continuously monitor for issues. Background technique [0002] Unmanned Aerial Vehicle (UAV) refers to an aircraft that does not require a pilot to drive, and is controlled by radio remote control equipment and its own program control device. With its low cost, strong mobility, and zero casualty rate, it plays an irreplaceable role in both the military and civilian fields. Among the various tasks that UAVs can complete, area surveillance is a very important task for UAV systems at present, especially when undertaking border air defense and alert tasks, the area that needs to be monitored is usually relatively wide, resulting in a single unmanned aerial vehicle. ...

AI Technical Summary

Problems solved by technology

If there is a lack of effective coordinated flight of UAV groups, not only will it not be able to fully reflect the advantages of multiple UAVs in cooperative monitoring coverage and detection, but there will even be a risk of conflict and collision between UAVs.

Images