A method and device for establishing map information and vehicle positioning

Abstract

The purpose of this application is to provide a method or device for establishing map information and vehicle positioning; obtain the route information of the target road; determine the reference point of the target road based on the route information; determine the coordinate position information of the reference point ; Establish a mapping relationship between the coordinate position information of the target object and the map positioning information based on the coordinate position information of the reference point. Compared with the prior art, this application can effectively save the cost of map information creation and reduce data processing time, thereby promoting the development and promotion of vehicle driving technology based on the application of map information, especially automatic driving technology.

Description

technical field [0001] This application relates to the field of automatic driving, in particular to a technology for establishing map information and vehicle positioning. Background technique [0002] During the driving process of existing vehicles, especially the driving process of self-driving vehicles, high-precision maps are usually used to achieve lane-level positioning of vehicles. At this time, the collection and establishment of high-precision maps need to rely on high-precision differential GPS, lidar or camera establishment, so that the location of various attribute targets on the road (traffic signs, ramps, roadsides, traffic lights, etc.) High precision; at the same time, self-driving vehicles also need to have a GPS receiving method with centimeter-level high-precision positioning function (the commonly used method is to build by yourself or use the GPS base station of the service provider). Such a method of map creation and use usually requires high equipment ...

AI Technical Summary

Problems solved by technology

Such a method of map creation and use usually requires high equipment costs and long data processing time, which to a certain extent makes it difficult for the rapid development and popularization of vehicle autonomous driving technology

In addition, the current creation and application of high-precision maps are mainly aimed at highways with relatively simple and clear road environments and target attributes, because in more complex urban road environments (dense high-rise buildings and trees, narrow lanes, The promotion of this technology is still limited by two reasons: high-precision mapping will be more time-consuming and laborious; high-precision GPS reception cannot be effectively guaranteed

The collection and establishment of high-precision maps will take more time and labor costs to filter non-road target attribute information. At the same time, frequent updates of urban roads also require frequent updates to the high-precision map database; and once the high-precision GPS on the vehicle cannot effectively receive Data, not only the aforementioned map building results have not been used, but also the behavior control and decision-making of the vehicle will also produce errors

At the same time, even on highways with relatively simple and clear target attributes, it is feasible to implement the above-mentioned high-precision map establishment and use methods, but it is not reasonable and effective

Due to the clear road structure and high degree of regularization of highways, in most cases, vehicles do not need to obtain too much meaningless target information (for example: utility poles, lampposts, etc.); in addition, many road target attributes can also be Recognize by visual means or write directly into map information (for example: ramp position, speed limit signs, etc.); thirdly, the control technology of many self-driving vehicles is not very fine at present, and cannot achieve centimeter-level motion control

Implementing the centimeter-level positioning described above does not allow the vehicle's motion planning to obtain benefits commensurate with its high cost

Therefore, there is a lack of an effective, low-cost, easier to implement and popularize map creation method and vehicle positioning method that can be applied to vehicle driving in the prior art

Images