Ultrasonic radar in-loop automatic driving test method based on Gaussian clustering

Abstract

The invention provides an ultrasonic radar in-loop automatic driving test method based on Gaussian clustering, and the method comprises the steps: determining a dangerous scene when an ultrasonic radar works through a Gaussian mixture model, building the dangerous scene through a virtual simulation platform such as Prescan, introducing a real ultrasonic radar into a test loop, and improving the test reliability. Echo signals received by the ultrasonic radar are generated by an echo simulator, Carsim software and Simulink are fused into a control algorithm of the vehicle to jointly carry out simulation testing, and finally a testing result is obtained to judge the reliability and the safety of the control algorithm of the automatic driving vehicle. According to the method, the defects of inconvenience and uneconomy caused by actual field testing are overcome, the cost of manpower and material resources is effectively reduced, and the testing efficiency is greatly improved.

Description

technical field [0001] The invention relates to the field of in-the-loop testing methods for automatic driving vehicles, in particular to an ultrasonic radar-in-the-loop automatic driving testing method based on Gaussian clustering. Background technique [0002] Automatic parking is an integral part of automatic driving technology. Most of the automatic parking systems on the market today involve ultrasonic radar sensors. How to reliably test the reliability of autonomous vehicle control algorithms has become a more important issue. The traditional field test has relatively large limitations, and it has become a problem to be solved to use a suitable simulation method to test the vehicle. [0003] Ultrasonic radar uses the principle of ultrasonic ranging. The ultrasonic radar fixed on the car body sends ultrasonic signals to the surroundings when the car is automatically parked, and receives the reflected echo signals when encountering obstacles to obtain the specific posit...

AI Technical Summary

Problems solved by technology

[0006] The invention provides an ultrasonic radar-in-the-loop automatic driving test method based on Gaussian clustering to realize the test of the automatic parking control algorithm of the automatic driving vehicle, which not only overcomes the low efficiency and safety problems of manual testing, but also solves the problem of virtual testing For the neglect of the sensor itself, more accurate test results can be obtained

Images