Unmanned aerial vehicle task allocation method based on group entropy and Q learning

Abstract

The invention provides an unmanned aerial vehicle task allocation method based on group entropy and Q learning. The unmanned aerial vehicle task allocation method comprises the steps that parameters are initialized; obtaining a neighbor unmanned aerial vehicle set of each unmanned aerial vehicle according to the position information of the unmanned aerial vehicles and the tasks; determining a sensitive unmanned aerial vehicle and a non-sensitive unmanned aerial vehicle; initializing self-evolution algorithm parameters of the cellular automaton; obtaining an evolution result, and measuring a result of each round of evolution by adopting group entropy; optimal group entropy information is selected from the group entropy information, and a Q table of each unmanned aerial vehicle in Q learningis initialized according to the information; and then initializing Q learning parameters to perform Q learning on the unmanned aerial vehicle, and finally obtaining an unmanned aerial vehicle task allocation result. According to the method, Q learning convergence is accelerated by adopting the method of taking the group entropy as the prior value, global information and local information are fully considered in the Q learning process, the learning efficiency of the algorithm is improved, and the convergence speed of unmanned aerial vehicle task allocation is increased on the basis of ensuringthe reliability of the unmanned aerial vehicle task allocation result.

Description

technical field [0001] The invention belongs to the technical field of unmanned aerial vehicle task allocation, and relates to an unmanned aerial vehicle task allocation method, in particular to an unmanned aerial vehicle task allocation method based on group entropy and Q learning, which can be used for unmanned aerial vehicle detection task allocation. Background technique [0002] Due to the low price, small size, low fuel consumption, and strong mobility of UAVs, UAV task allocation is often used in scenarios such as environmental detection and geographic surveying and mapping. The UAV task allocation problem can be described as multiple UAVs in the scene to perform multiple tasks. Each task requires the UAV to provide a corresponding amount of execution and the execution amount decays with the distance from the UAV to the task. The purpose is Find an effective and reasonable allocation scheme to assign tasks to different UAVs, so that all tasks can get the required amou...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that it is easy to fall into local optimal convergence by excessively seeking the global optimal result, resulting in a slow convergence of task allocation.

Images