Automatic driving control method

Abstract

The embodiment of the invention relates to an automatic driving control method. The automatic driving control method comprises the steps of: obtaining initial position information and target position information of a vehicle, and generating an automatic driving path in a navigation map; determining whether the vehicle is in a preset normal state or not; acquiring distance data of obstacles on two sides and binocular vision data of the obstacles in front; determining whether collision trends exist on two sides of the vehicle or not; performing 3D modeling on the front obstacle; identifying the front obstacle according to the front obstacle 3D model data; determining the highest value of the collision risk according to the distance between the vehicle and the front obstacle and the recognition result of the front obstacle; determining whether the highest value of the collision risk is lower than a preset threshold value or not; and when the vehicle is in a normal state, having no a collision trend on two sides of the vehicle, and when the highest collision risk value is lower than the preset threshold value, continuing to drive according to the automatic driving path. The method has high precision, and the data can be transmitted to the processor without reducing the frame rate when the vehicle is driven at high speed.

Description

technical field [0001] The invention relates to the field of automatic driving, in particular to an automatic driving control method. Background technique [0002] With the continuous improvement of the level of science and technology in recent years, the rapid development of artificial intelligence has been widely used in various fields. Among them, since autonomous vehicles can efficiently utilize traffic resources, alleviate traffic congestion, and reduce carbon emissions, autonomous driving technology has increasingly become the focus of attention. [0003] Self-driving vehicles rely on the cooperation of artificial intelligence, visual computing, radar, monitoring devices and global positioning systems, allowing computers to automatically and safely operate motor vehicles without any active human operation. For example, monitoring devices including visual sensors, radar sensors, and laser range finders are used to obtain surrounding traffic conditions, and to navigate ...

AI Technical Summary

Problems solved by technology

[0004] Due to the low accuracy of the current visual sensor for image acquisition of obstacles around the vehicle, the error is generally within a few centimeters, so the description accuracy of the distance between the obstacle and the vehicle, the trajectory of the obstacle, etc. is low, which cannot meet the requirements of automatic driving. The precision required by the vehicle

In addition, since the frame rate of the current visual sensor to transmit images to the on-board computer can only reach 60 frames, when the vehicle is driving at high speed, the transmission speed cannot meet the high-speed driving requirements of the vehicle

If the frame rate is reduced in order to be able to return images in time, it will further reduce the description accuracy of obstacles, which will easily lead to safety accidents

Images