Vehicle motion state information extraction method based on visual and radar collaborative awareness

Abstract

The invention provides a vehicle motion state information extraction method based on visual and radar collaborative awareness. A control module collects motion state information such as speed, position, orientation, and the like of other mobile platforms obtained by a visual measurement module and a radar measurement module, and filters out two groups of motion state information of the two modules on the same mobile platform; two groups of information is used as an object, an error parameter of visual measurement information is extracted by taking radar measurement information as the standard by an error parameter extraction module; a calibration parameter calculation module uses the error parameter to measure a calibration parameter of the visual measurement; and a calibration module calibrates the visual measurement information by using the calibration parameter and sends the calibrated information to the control module. According to the vehicle motion state information extraction method based on the visual and the radar collaborative awareness provided by the invention, the visual and radar awareness modes can work collaboratively, so that the advantages are complementary, and the cost, the precision, the operating distance and visualization and other characteristics can be taken into account, which is beneficial to improve the cost performance of a vehicle motion state information acquisition system and promote the wide application.

Description

technical field [0001] The invention relates to the fields of unmanned driving, automotive electronics, etc., and specifically relates to a method for extracting vehicle motion state information through vision and radar cooperative perception. Background technique [0002] Accurately extracting the motion state information of surrounding vehicles is one of the key technologies for realizing advanced assisted driving and unmanned driving. Among them, the vehicle motion state information acquisition technology based on vision and radar perception are two effective methods. However, both of these two vehicle electronic devices have some disadvantages, which restrict their wide application in the field of advanced assisted driving and unmanned driving. The visual perception method has the advantages of low cost, long operating distance, wide field of view, intuitive visualization, and good interaction with people, but the measurement accuracy is low, especially at low speeds, i...

AI Technical Summary

Problems solved by technology

However, both of these two types of on-board electronic devices have some disadvantages, which restrict their wide application in the field of advanced assisted driving and unmanned driving

The visual perception method has low cost, long working distance, wide field of view, intuitive visualization, and good interaction with people, but the measurement accuracy is low, especially at low speeds, it cannot work effectively

On the other hand, the radar method can obtain high precision and can still work effectively at low speeds. However, under the requirements of wide field of view and long-distance work, the cost is generally high, and it is difficult to meet the small size requirements of the vehicle.

Images