Target situation fusion sensing method and system based on multiple sensors

Abstract

The invention provides a target situation fusion sensing method based on multiple sensors. The target situation fusion sensing method comprises the steps: monitoring a detection region, and obtainingmulti-target situation information data in a recognition environment; reconstructing a point cloud target and a three-dimensional scene model based on the visual features; detecting a target track, and carrying out registration fusion on a local track and an overall track; obtaining a target identity through a DS evidence theory to realize target friend or foe grouping; and analyzing the behaviorsituation of the enemy target, obtaining the threat degree of the enemy target, and taking suppression measures according to the threat degree level. The invention further provides a device which comprises a measuring and positioning system, an information fusion system, a situation awareness and threat evaluation system, a decision-making system and a display system. The target situation fusion sensing method realizes target situation awareness and threat assessment based on DS evidence theory and target data multi-fusion, and realizes target and scene reconstruction by using point cloud so that accurate information can be provided for detection and identification of the shielded target, scene reproduction, track detection and prediction, battlefield target situation awareness and the like.

Description

technical field [0001] The invention relates to the fields of artificial intelligence, target fusion, situation awareness, etc., and in particular relates to a multi-sensor based target situation fusion perception method and system. Background technique [0002] In recent years, the rapid development of robot technology with remote control and autonomous control as the core has led to the emergence of combat systems such as drones, unmanned ships, and unmanned underwater pre-submersible vehicles, which have been put into the battlefield one after another. A new combat style is emerging. In order to have a complete and comprehensive perception of the battlefield environment, unmanned platforms are equipped with a large number of sensors to collect environmental information in real time. [0003] In order to overcome the problem of insufficient capability of a single platform, unmanned platforms often form multi-platform unmanned teams to perform tasks. There are technical dif...

AI Technical Summary

Problems solved by technology

[0003] In order to overcome the problem of insufficient capability of a single platform, unmanned platforms often form multi-platform unmanned teams to perform tasks. There are technical difficulties in such manned and unmanned coordinated control, which is specifically reflected in the fact that commanders need to monitor the current situation based on a large amount of sensor data. For real-time analysis, it is necessary to introduce an auxiliary situation awareness system to help integrate and screen information, provide reference situation information, and push important information related to the situation and decision-making, so that the commander's main energy can be devoted to decision-making

[0004] In terms of sensor information integration methods, the method of introducing situational awareness modules is mostly adopted at present, that is, according to actual needs, the overall situation is divided into easy-to-understand situations, and the situation understanding results of the auxiliary system are reported to the control center, thereby reducing combat personnel As far as the mathematical process is concerned, the situational awareness module in the auxiliary control system is a reasoning process. The current mainstream technology includes: using Bayesian reasoning to build a situational awareness network. The network architecture has a clear hierarchy, but in actual use Facing the problem that the prior probability is difficult to determine

An algorithm based on fuzzy cognitive maps, which weakens the influence of sensor errors, but lacks the inclusion of sample randomness, and there is a reasonable way to deal with the differences in expert experience in expert systems

[0005] Although the methods obtained in the above studies have a very good effect on completing the task, in the actual complex battlefield environment, it is not enough for the auxiliary command and control system to simply perform situational awareness on sensor information directly. How to ensure the accuracy of situational awareness, How to conduct reasonable information screening and sorting to ensure that command intervention and decision-making can be carried out quickly and effectively is an urgent problem that needs to be solved now

Images