A multi-sensor adaptive management method for multi-UAV cooperative detection

Abstract

The invention relates to a multi-sensor adaptive management method oriented to multi-unmanned aerial vehicle cooperative detection. The method includes a target cooperative recognition model using Bayesian network reasoning technology; using Bayesian network to obtain the multi-sensor feature level cooperative perception probability through offline learning and online learning; using perceptual information entropy to represent the "uncertainty" of the target category Based on perceptual information entropy, a closed-loop detection loop of "perception-learning-decision-action" from target cooperative recognition to sensor dynamic management is established; using perceptual information entropy increment as the dynamic scheduling of sensor resources in multi-UAV cooperative detection network Judgment basis; use the intelligent optimization algorithm combining genetic algorithm and particle swarm algorithm to improve the allocation efficiency of "multi-UAV-multi-target"; from the perspective of comprehensive identification of multi-sensor multi-target features, use multiple sensors to detect multiple features to identify unknown targets.

Description

technical field [0001] The invention belongs to the technical field of multi-UAV cooperative detection, in particular to a multi-sensor self-adaptive management method for multi-UAV cooperative detection. Background technique [0002] With the rapid development of science and technology, UAV has become a research hotspot at home and abroad. As a kind of unmanned aerial vehicle that can be controlled and can perform multiple tasks, UAV has a wide range of application prospects in both military and civilian applications. In the future, swarms will be the development trend of drones. [0003] In the process of cluster information sharing, how to transmit the perceived target and platform status information to other individuals, so that the whole system can not only meet the available bandwidth limit to reduce the probability of being detected, but also meet the requirements of collaborative control and decision-making need is a very important question. In the multi-UAV coope...

AI Technical Summary

Problems solved by technology

Most of these technologies perform sensor optimization management on the basis of comprehensive target recognition at the decision-making level, and cannot fully utilize the maximum effectiveness of different detection characteristics of various types of sensors; in addition, due to the complexity of the battlefield environment, two sensors may be used against the same target. Making an opposite judgment may cause the failure of the sensor adaptive scheduling algorithm based on D-S evidence theory

Images