Multi-unmanned aerial vehicle task decision-making method based on MADDPG

Abstract

The invention discloses a multi-unmanned aerial vehicle task decision method based on MADDPG, which introduces an MADDPG algorithm into multi-unmanned aerial vehicle task allocation, and comprises thesteps of firstly establishing a two-dimensional combat environment model required by deep reinforcement learning according to the actual combat environment of multiple unmanned aerial vehicles, secondly establishing mathematical description of various threats such as air defense missiles in the combat environment of the multiple unmanned aerial vehicles, and finally, taking the tracks and distances of the multiple unmanned aerial vehicles and the defense threats of the battlefield as constraint conditions, and performing learning training to obtain a multi-unmanned aerial vehicle task decision model. According to the method, an experience pool and a double-network structure are adopted, so that the operation and convergence speed of the whole network is greatly improved, a result can be obtained more quickly in the high-speed flight process of the unmanned aerial vehicles, the purpose of autonomous decision making of the multiple unmanned aerial vehicles can be achieved, and the highefficiency of task decision making of the unmanned aerial vehicles can also be guaranteed in an unknown combat environment.

Description

technical field [0001] The invention belongs to the field of flight control, and in particular relates to a method for multi-unmanned aerial vehicle task decision-making. Background technique [0002] For the military of various countries, drones will become one of the indispensable weapons in the future battlefield. UAVs are likely to become the target of attacks and counterattacks by multiple combat platforms, and become the most common and deadly air combat "sharp sword". Although the existing multi-UAV system can complete some complex combat tasks, as a huge fleet coordination system, the traditional multi-UAV mission decision-making is generally carried out under the conditions of the known battlefield environment. If the set combat plan cannot meet the actual combat environment, it is difficult to make a timely response plan. Therefore, it is particularly important for future UAV operations to find a method that can quickly and efficiently assign tasks to multiple UA...

AI Technical Summary

Problems solved by technology

However, in a multi-UAV environment, it is difficult for traditional reinforcement learning methods to play a role, because in a multi-UAV environment, each UAV is constantly changing, and the environment is no longer stable, and for traditional reinforcement learning algorithms For the strategy gradient method in , as the number of drones increases, the complexity of the environment also increases, which leads to the optimization method of estimating the gradient by sampling, the variance increases sharply, and it is difficult to calculate the final result

Images