Multilayer parking lot positioning method based on high-precision map data

Abstract

The invention relates to a multilayer parking lot positioning method and device based on high-precision map data. The method comprises the steps of firstly receiving a GPS coordinate signal, acquiringmap data around a vehicle from the high-precision map data according to the GPS coordinate signal, and generating a LocationMap; then acquiring external sensor data, fusing the external sensor data with the LocationMap, positioning hierarchy information of the vehicle, and denoising the LocationMap by using the hierarchy information; and finally, correcting the longitude and latitude informationand the altitude information of the GPS by using the denoised LocationMap to obtain the current position and the hierarchy of the vehicle. According to the multilayer parking lot positioning method and device, the external sensor data, the GPS data and the high-precision map data are subjected to fusion processing, so that the problem of insufficient positioning precision caused by single-mode sensing is avoided, and the positioning precision of the vehicle in automatic driving is improved.

Description

technical field [0001] The invention relates to the technical field of automatic driving, in particular to a multi-storey parking lot positioning method based on high-precision map data. Background technique [0002] With the rise of autonomous driving technology, lane-level safe driving maps have entered everyone's field of vision and received more and more attention. Therefore, the safety of autonomous driving is put on the agenda, such as how to make decisions when encountering obstacles and still reach the destination, anti-collision and so on. However, the hardware perception in the automatic driving system, such as ultrasonic radar, binocular camera, etc., has limited perception conditions and a limited range of perception, and the map can well make up for these hardware perception deficiencies, so the map It is reasonable and economical as a sensor in an automatic driving system. [0003] At present, high-precision maps can make up for the lack of hardware perceptio...

AI Technical Summary

Problems solved by technology

[0003] At present, high-precision maps can make up for the lack of hardware perception, but whether the information of high-precision map sensors can be accurately output depends on whether the vehicle position can be correctly matched to the map. In this environment, the incoming location information (especially the altitude) is often wrong. Once an error occurs, the location information of the vehicle cannot be guaranteed, and the output map information around the vehicle is also incorrect.

Images