Course prediction method based on dynamic data real-time adaptation

Abstract

The invention relates to a course prediction method based on dynamic data real-time adaptation. The method comprises the following steps of: processing sound signals collected in a vehicle advancing process, and obtaining a plurality of pieces of orientation angle data; placing the obtained multiple pieces of orientation angle data in a continuous interval in a coordinate rotation mode; adopting the multiple pieces of orientation angle data as a training set to train a linear regression model, and determining the change trend of an orientation angle; and according to the change trend of the orientation angle, judging the movement direction of a field moving vehicle relative to the arrangement of a sensor, and predicting the advancing course of the field moving vehicle on the basis of the orientation angle of the sound sensor when the sensor is arranged. According to the method, course judgment can be carried out on the data generated by the sensor in real time, the model is simple, the robustness is high, rigid requirements for data sets do not exist, the rough course of a moving target can be obtained under a few data sets, and the more accurate course is obtained along with increase of the data sets.

Description

technical field [0001] The invention relates to the technical field of speech signal orientation angle judgment, in particular to a heading prediction method based on dynamic data real-time adaptation. Background technique [0002] The infrared sensor function is generally used to predict the course of the field vehicle, but in some extreme cases, the infrared sensor cannot work: (1) the external environment temperature is too high; (2) when there is an obstacle between the sensor and the field moving target, the infrared sensor (3) Infrared has a low detection rate for long-distance vehicles. Therefore, it is very important to study the heading according to the orientation angle obtained by the sound signal processing: it not only allows the composite sensor to predict the direction of the moving target well when the vehicle is traveling in the field, but also allows each sensor to complement each other, so that the composite The accuracy of the sensor in monitoring moving...

AI Technical Summary

Problems solved by technology

[0002] The infrared sensor function is generally used to predict the course of the field vehicle, but in some extreme cases, the infrared sensor cannot work: (1) the external environment temperature is too high; (2) when there is an obstacle between the sensor and the field moving target, the infrared sensor (3) Infrared has a low detection rate for long-distance vehicles

Images