Target tracking and hunting method for unmanned aerial vehicle group adaptive environment

Abstract

The invention discloses a target tracking and hunting method for an unmanned aerial vehicle group adaptive environment, and the method comprises the following steps: (1), building a multi-agent collaborative planning model through employing an MADDPG algorithm, and achieving the tracking and hunting of a target through the unmanned aerial vehicle group; and (2) when the unmanned aerial vehicle group approaches the threat area, performing autonomous adjustment and re-planning on the positions of the unmanned aerial vehicles by using a GA algorithm to avoid entering the threat area, wherein the survival rate of the unmanned aerial vehicles is improved, and meanwhile, a hunting task is completed. A layered hunting model is established and divided into two layers: a hunting layer and a multi-agent training layer. The unmanned aerial vehicle group interacts with the environment in real time, so that the current environment state can be obtained at any time. The surrounding layer determines whether to adjust the formation from the current state, and calculates a surrounding position distribution scheme. Aiming at the dynamic change of the environment and the task, the task execution success rate is improved in the relatively complex environment with the threat, meanwhile, the surrounding position is autonomously changed to avoid the threat area, and the risk of the unmanned aerial vehicle group is reduced.

Description

technical field [0001] The invention relates to the technical field of unmanned aerial vehicle swarm mission planning, and more specifically, it relates to a target tracking and rounding-up method for an unmanned aerial vehicle swarm to adapt to the environment. Background technique [0002] UAVs have unique advantages such as zero casualties, continuous operation, low cost, and excellent mobility, making UAV swarm combat a research hotspot in recent years. In terms of UAV cluster command and control collaborative task decision-making, most of them adopt swarm intelligence algorithms based on ant colony algorithm and wolf pack algorithm. [0003] However, the swarm intelligence algorithm is still unable to meet the characteristics of autonomy and autonomy of UAV clusters. Reinforcement learning has been widely concerned and applied in recent years. However, the existing reinforcement learning-based UAV round-up target problem scene setting is relatively simple, less threat ...

AI Technical Summary

Problems solved by technology

[0003] However, the swarm intelligence algorithm is still unable to meet the characteristics of autonomy and autonomy of UAV clusters. Reinforcement learning has been widely concerned and applied in recent years.

However, the existing reinforcement learning-based UAV round-up target problem scene setting is relatively simple, less threat factors are considered, and the round-up method is not flexible enough.

Images