Method and system for predicting electromagnetic environment effects of laser radar

Abstract

The invention relates to a laser radar electromagnetic environment effect prediction method and system. The method includes: combining the K nearest neighbor algorithm, the decision tree algorithm, the gradient boosting tree algorithm and the extremum gradient boosting algorithm based on the stacking technology to construct an electromagnetic environment effect prediction model; obtaining the point cloud test data when the lidar is not disturbed; obtaining samples Data; the sample data is the electromagnetic environment effect threshold value of lidar under different interference conditions; the electromagnetic environment effect prediction model is trained based on the sample data and the undisturbed point cloud test data, and the trained electromagnetic environment effect prediction model is obtained; Input the predicted interference condition and the point cloud test data without interference into the trained electromagnetic environment effect prediction model, and output the threshold value of the electromagnetic environment effect under the interference condition to be predicted. The invention can comprehensively predict the electromagnetic environment effect and improve the robustness of the prediction model.

Description

technical field [0001] The invention relates to the field of electromagnetic environment effect prediction, in particular to a laser radar electromagnetic environment effect prediction method and system. Background technique [0002] Lidar is an active imaging technology developed on the basis of photoelectric radiation detection and traditional radar. With the development of lidar technology, lidar has a wide range of applications in the fields of terminal guidance, laser active homing, armored vehicle detection under camouflage, automatic landing of planetary probes, and aircraft collision avoidance. The divergence angle of the laser radar detection beam is usually tens of milliradians, and it can only be interfered and monitored if it happens to be on the laser detection optical path. Compared with traditional radar, it has stronger spatial anti-interference ability and confidentiality, showing a good military application prospect. [0003] At the same time, with the rap...

AI Technical Summary

Problems solved by technology

However, in the process of evaluating the electromagnetic environment effect of lidar in a complex electromagnetic environment, because it is not possible to exhaust all the combinations of electromagnetic signal parameters, there is a problem of incomplete evaluation of the electromagnetic environment effect

[0004] Most of the existing predictions of electromagnetic environment effects use traditional simple mathematical models and single algorithm models, but the laws of electromagnetic environment effects are often characterized by nonlinearity, ambiguity, and uncertainty. The mathematical model cannot handle this problem very well, and the single-algorithm model has poor robustness and low generalization ability

Images