Automatic-drive vehicle positioning method

Abstract

The invention discloses an automatic-drive vehicle positioning method. The automatic-drive vehicle positioning method comprises the following steps that S1, a guideboard is mounted; S2, automatic recognizing is conducted; S3, database comparison is conducted; and S4, position output is conducted. According to the automatic-drive vehicle positioning method, through the guideboard arranged on a roadand a unique corresponding coding arranged on the guideboard, the coding corresponds to the position, and is stored in a database, and the coding can be recognized by a vision or other sensors, and therefore the position is determined through comparison with the database. According to recognition to the guideboard by the sensor, the relative position between a vehicle and the guideboard can be determined by the position of the guideboard in the field of vision, and the included angle between the vehicle and the current road can be determined by further analysis. In the current automatic driving field, the problem about positioning accuracy and real-time positioning limiting the volume production prospect of automatic driving is solved, and the automatic-drive vehicle positioning method can promote the volume production of automatic driving.

Description

technical field [0001] The invention relates to the field of automatic driving, in particular to a positioning method for an automatic driving vehicle. Background technique [0002] The goal of autonomous driving technology is becoming more and more perfect, and the technical solutions are becoming more and more mature. In order to realize the automatic driving of the car, the precise positioning of the vehicle itself must be solved. The current technical solution first requires the centimeter-level GPS / Beidou satellite system to first determine the approximate position, which is the basis for further precise positioning. However, the current satellite positioning such as GPS / Beidou mainly has the following disadvantages: [0003] 1) Errors caused by clock errors, signal reflections, atmospheric effects, etc. are difficult to eliminate and have strong uncertainties. Although the use of carrier phase difference technology can reduce the error to a certain extent, which can ...

AI Technical Summary

Problems solved by technology

[0003] 1) Errors caused by clock errors, signal reflections, atmospheric effects, etc. are difficult to eliminate and have strong uncertainties

Although the use of carrier phase difference technology can reduce the error to a certain extent, which can reduce the error in most cases, but in other cases, the error between the satellite to the reference station and the satellite to the GPS system of the vehicle They are not equal, and the distance from the base station to the vehicle is always changing. There is also an error in the radio transmission here.

Therefore, the above-mentioned errors cannot be completely avoided under the current technology.

[0004] 2) The signal is blocked: GPS receivers cannot receive signals in real time, such as between tunnels and tall buildings

[0005] 3) The signal update frequency is low: about 10 Hz, calculated at a speed of 30m / s, the vehicle has traveled a distance of 3 meters in 100ms

[0006] 4) In addition to knowing the position of the vehicle, the angle of the vehicle is also very important for self-driving vehicles. For the measurement of the vehicle angle, satellite positioning is currently impossible

[0008] For the disadvantage 2), there is no feasible solution at present, and it can only be compensated by multiple satellite positioning systems. However, due to the longer time span between multiple satellite positioning position information, for the IMU, the greater the cumulative error, there is an unresolved risk

Therefore, the main technical disadvantages are: 1) The basic method used for position positioning, the error of satellite positioning is difficult to eliminate, which is mainly a problem that always exists in long-distance positioning; 2) For the accuracy requirements of vehicle angle and microcosm, currently There are no fixed and reliable road signs on the road that can be recognized by lidar or cameras to determine the above requirements

[0011] At present, there is no automatic driving method that can determine the precise position and angle of the vehicle by applying a high-definition vehicle camera to detect preset characteristic objects in the driving environment (road environment) without relying on the satellite positioning system.