Vehicle longitudinal positioning method and system based on least square method

Abstract

The invention relates to a vehicle longitudinal positioning method and system based on a least square method. The method comprises the following steps: acquiring coarse positioning information of a vehicle; obtaining side marker information in the vehicle driving process, and obtaining longitudinal relative position information of the vehicle and a marker; loading map information near the coarse positioning information by using a high-precision map, and determining absolute position information of the marker based on the map information; based on the absolute position of each marker and the longitudinal relative position information of the vehicle, performing linear fitting by using a least square method to obtain longitudinal position information of the vehicle, and correcting longitudinal positioning information in the coarse positioning information by using the longitudinal position information of the vehicle; the longitudinal positioning precision of shielded and semi-shielded scenes is improved, the overall positioning precision requirement is helped, and the method has practical significance in the field of automatic driving.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving technology, and in particular to a vehicle longitudinal positioning method and system based on the least squares method. Background Art

[0002] In the field of intelligent driving technology, accurate vehicle positioning is a very important part. The positioning of the vehicle determines whether it can make correct and timely decisions and complete the driving action of autonomous driving in assisted driving or autonomous driving. Providing the wrong vehicle position may lead to traffic accidents, so the accuracy and correctness of positioning is the prerequisite for ensuring the development of intelligent driving.

[0003] The vehicle positioning technology currently used is the fusion of RTK (real-time motion positioning) and IMU (inertial measurement unit) as the positioning result. Through mechanical arrangement, the original output of IMU is converted into attitude, speed, and position information as the state prediction, and the position information is corrected with the real-time motion positioning result of RTK as the state observation.

[0004] Generally, the existing vehicle navigation uses the mean value of IMU to calculate the position, and the RTK data is used for position correction. This combination inputs RTK and IMU into the combined framework as two independent sets of position and attitude parameters, and generates a single combined navigation solution by feedback, and finally outputs the result smoothly through the Kalman filter. The characteristic of this method is that if the RTK signal is interrupted, the position can still be calculated by IMU, but the IMU device contains static errors and dynamic errors. As time goes by, the position error will accumulate, and the position can only be corrected when the next RTK signal is restored. Therefore, in order to solve the problem of occlusion and semi-occlusion scenes, new observations need to be introduced. Camera is a good choice as one of the powerful perception sensors.

[0005] Currently, in the field of autonomous driving, most of the lateral corrections are made by observing the distance to the lane line with a camera, and the longitudinal corrections are only based on the results of combined navigation. In this way, in a no-signal scenario, the lateral error can be corrected by the camera to maintain the lane. However, the longitudinal error cannot be guaranteed, and it is important to keep a safe distance from the vehicle in front during autonomous driving. Summary of the invention

[0006] In response to the technical problems existing in the prior art, the present invention provides a vehicle longitudinal positioning method and system based on the least squares method, which improves the longitudinal positioning accuracy in occluded and semi-occluded scenes, helps meet the overall positioning accuracy requirements, and has practical significance in the field of autonomous driving.

[0007] According to a first aspect of the present invention, a vehicle longitudinal positioning method based on least squares method is provided, comprising:

[0008] Step 1, obtaining the rough positioning information of the vehicle; obtaining the information of the markers on the side of the vehicle during the driving process, and obtaining the longitudinal relative position information of the vehicle and the markers;

[0009] Step 2, using a high-precision map to load map information near the rough positioning information, and determining the absolute position information of the marker based on the map information;

[0010] Step 3, based on the absolute position of each marker and its longitudinal relative position information with the vehicle, use the least squares method to perform linear fitting to obtain the longitudinal position information of the vehicle, and use the longitudinal position information of the vehicle to correct the longitudinal positioning information in the coarse positioning information.

[0011] Based on the above technical solution, the present invention can also make the following improvements.

[0012] Optionally, the process of obtaining the rough positioning information of the vehicle in step 1 includes:

[0013] The positioning result is obtained by fusing RTK and IMU, and then the positioning result is passed through a Kalman filter to obtain the coarse positioning information.

[0014] Optionally, the type of the marker acquired in step 1 includes: ground marker and road marker; the marker information includes: the content and location of the marker;

[0015] When the marker is a ground marker, the longitudinal distance between the sensing system and the marker is used as the longitudinal relative position information between the vehicle and the marker;

[0016] When the marker is a road marker, the longitudinal relative position information of the vehicle and the marker is calculated based on the installation parameters of the perception system and the observed longitudinal distance to the marker. The installation parameters include: the height and viewing angle of the installation position of the perception system relative to the ground.

[0017] Optionally, the step 1 further includes: calculating a compensation distance according to a timestamp difference between the marker and the current position and the wheel speed, and compensating the calculated longitudinal relative position information based on the compensation distance.

[0018] Optionally, the process of calculating the longitudinal position information of the vehicle in step 3 and using the longitudinal position information of the vehicle to correct the longitudinal positioning information in the coarse positioning information includes:

[0019] Step 301, traverse and classify the landmarks sensed by the perception system and store them;

[0020] Step 302, converting the coarse positioning information from the vehicle body coordinate system to the perception system coordinate system, and then to the Odometry coordinate system;

[0021] Step 303, searching for a landmark map identifier closest to the rough positioning information in the high-precision map in the Odometry coordinate system and saving the landmark ID;

[0022] Step 304, transferring the found landmark map identification to the perception system coordinate system, recording the identification distance of each landmark map identification from the vehicle and the corresponding landmark ID, and saving them in a container with time as a variable;

[0023] Step 305, when the nearest identification distance is less than the threshold and the data meets the set requirements, outliers are screened out through MAD-based Z-score anomaly detection;

[0024] Step 306: Perform a least squares linear fit on the data obtained in step 305, and use the linear fit to calculate the mean square error between the current position and the true position to compensate for the longitudinal position.

[0025] Optionally, in step 302, the nearest point search is performed using KD and / or divide-and-conquer method.

[0026] Optionally, the type of the marker acquired in step 1 includes: a lane line; the marker information includes: the type and position of the lane line;

[0027] The step 3 also includes: correcting the lateral positioning information of the coarse positioning information based on the lateral position information of the lane line.

[0028] According to a second aspect of the present invention, there is provided a vehicle longitudinal positioning system based on the least squares method, comprising: an integrated navigation system, a perception system, a high-precision map and a longitudinal positioning algorithm module;

[0029] The integrated navigation system is used to obtain rough positioning information of the vehicle;

[0030] The perception system allows the user to obtain information about the markers on the side of the vehicle during driving, and obtain longitudinal relative position information between the vehicle and the markers;

[0031] The longitudinal positioning algorithm module uses the high-precision map to load map information near the coarse positioning information, and determines the absolute position information of the marker based on the map information; based on the absolute position of each marker and its longitudinal relative position information with the vehicle, the least squares method is used for linear fitting to obtain the longitudinal position information of the vehicle, and the longitudinal positioning information in the coarse positioning information is corrected using the vehicle longitudinal position information.

[0032] According to a third aspect of the present invention, there is provided an electronic device comprising a memory and a processor, wherein the processor is configured to implement the steps of a vehicle longitudinal positioning method based on the least squares method when executing a computer management program stored in the memory.

[0033] According to a fourth aspect of the present invention, there is provided a computer-readable storage medium on which a computer management program is stored. When the computer management program is executed by a processor, the steps of a vehicle longitudinal positioning method based on the least squares method are implemented.

[0034] The least squares method-based vehicle longitudinal positioning method, system, electronic device and storage medium provided by the present invention have the following advantages:

[0035] (1) The present invention can correct and eliminate positioning errors in the longitudinal position on complex urban roads (including occluded and semi-occluded scenes) based on the coarse positioning information provided by the integrated navigation system through lane lines, ground signs and road marker information.

[0036] (2) The longitudinal positioning method of the present invention can quickly find the nearest landmark information and match it. The search method used is of low complexity, high search efficiency and reliability. The longitudinal distance is linearly fitted by the least squares method, which can improve the accuracy of the longitudinal error. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A flow chart of a vehicle longitudinal positioning method based on the least squares method provided by the present invention;

[0038] Figure 2 A flowchart of a longitudinal positioning algorithm provided by an embodiment of the present invention;

[0039] Figure 3 A flowchart of a KD tree search provided by an embodiment of the present invention;

[0040] Figure 4 A schematic diagram of a least squares method provided by an embodiment of the present invention;

[0041] Figure 5 A structural block diagram of a vehicle longitudinal positioning system based on the least squares method provided by the present invention;

[0042] Figure 6 A schematic diagram of the hardware structure of a possible electronic device provided by the present invention;

[0043] Figure 7 A schematic diagram of the hardware structure of a possible computer-readable storage medium provided by the present invention;

[0044] In the accompanying drawings, the components represented by the reference numerals are listed as follows:

[0045] 10-Integrated navigation system; 20-Camera perception system; 30-High-precision map; 40-Vertical positioning algorithm. DETAILED DESCRIPTION

[0046] The principles and features of the present invention are described below in conjunction with the accompanying drawings. The examples given are only used to explain the present invention and are not used to limit the scope of the present invention.

[0047] Figure 1 A flow chart of a vehicle longitudinal positioning method based on the least squares method provided by the present invention is as follows: Figure 1 As shown, the positioning method includes:

[0048] Step 1, obtain the rough positioning information of the vehicle; obtain the marker information on the side of the vehicle during the driving process, and obtain the longitudinal relative position information between the vehicle and the marker.

[0049] Step 2: Use a high-precision map to load map information near the coarse positioning information, and determine the absolute position information of the landmark based on the map information.

[0050] Step 3, based on the absolute position of each marker and its longitudinal relative position information with the vehicle, use the least squares method to perform linear fitting to obtain the longitudinal position information of the vehicle, and use the longitudinal position information of the vehicle to correct the longitudinal positioning information in the coarse positioning information.

[0051] The present invention provides a vehicle longitudinal positioning method based on the least squares method, which utilizes lane lines, ground signs and road markers observed by a camera sensor system to correct the longitudinal positioning with errors provided by a combined navigation system, so as to correct the longitudinal position of the vehicle during driving. The longitudinal positioning accuracy in occluded and semi-occluded scenes is improved, which helps meet the overall positioning accuracy requirements and has practical significance in the field of autonomous driving.

[0052] Example 1

[0053] Embodiment 1 provided by the present invention is an embodiment of a vehicle longitudinal positioning method based on the least squares method provided by the present invention, combined with Figure 1 It can be seen that the embodiment of the positioning method includes:

[0054] Step 1, obtain the rough positioning information of the vehicle; obtain the marker information on the side of the vehicle during the driving process, and obtain the longitudinal relative position information between the vehicle and the marker.

[0055] In specific implementations, lane lines, road signs and other information can be sensed by camera sensors or lidar perception systems, and transmitted to the image algorithm in the camera perception system or lidar perception system to obtain the perception information of the markers. Lidar, like cameras, is a popular perception sensor at present. Its accuracy in identifying lane lines and traffic signs is better than that of cameras, and it has stronger anti-interference ability in bad weather.

[0056] In a possible implementation manner, the process of obtaining the rough positioning information of the vehicle in step 1 includes:

[0057] The positioning result is obtained by fusing RTK and IMU, and then the positioning result is passed through the Kalman filter to obtain coarse positioning information.

[0058] In a possible implementation manner, the types of markers acquired in step 1 include: ground markers and road markers; the marker information includes: the content and location of the marker.

[0059] When the marker is a ground marker, the longitudinal distance between the perception system and the marker is used as the longitudinal relative position information between the vehicle and the marker.

[0060] When the marker is a road marker, the longitudinal relative position information between the vehicle and the marker is calculated based on the installation parameters of the perception system and the observed longitudinal distance to the marker. The installation parameters include: the height and viewing angle of the installation position of the perception system relative to the ground.

[0061] In a specific implementation, step 1 also includes: recording the installation parameters of the camera sensor, including the height and viewing angle of the camera from the ground when installed; recording the installation parameters of the IMU sensor, including the IMU installation angle and static and dynamic zero bias parameters.

[0062] In a possible implementation manner, step 1 further includes: calculating a compensation distance according to a timestamp difference between the marker and the current position and a wheel speed, and compensating the calculated longitudinal relative position information based on the compensation distance.

[0063] In the specific implementation, the coarse positioning results are received by the high-precision map and longitudinal positioning algorithm module. There is a certain gap between the time when the landmark is observed and the time when the positioning is sent from upstream to this module. The distance during this period needs to be made up based on the vehicle speed.

[0064] Step 2: Use a high-precision map to load map information near the coarse positioning information, and determine the absolute position information of the landmark based on the map information.

[0065] Step 3, based on the absolute position of each marker and its longitudinal relative position information with the vehicle, use the least squares method to perform linear fitting to obtain the longitudinal position information of the vehicle, and use the longitudinal position information of the vehicle to correct the longitudinal positioning information in the coarse positioning information.

[0066] In a possible embodiment, Figure 2 The flowchart of a longitudinal positioning algorithm provided by an embodiment of the present invention is shown in FIG. Figure 1 and Figure 2 It can be seen that the process of calculating the longitudinal position information of the vehicle in step 3 and correcting the longitudinal positioning information in the coarse positioning information using the longitudinal position information of the vehicle includes:

[0067] Step 301, traverse the landmarks perceived by the perception system and store them by category.

[0068] In the specific implementation, the landmarks perceived by the perception system are traversed, and the perception system will classify and store the landmarks, so different landmarks need to be classified and processed in this step. The perception system will perceive three types of landmarks in total: 1. Lane lines, which are used to calculate the lateral error of the vehicle. The camera observes the distance to the left and right lane lines for lateral correction; 2. Ground landmarks, which are used to calculate the longitudinal error of the vehicle. They are on the same level as the ground, and the distance observed by the camera to this landmark can be directly used as the longitudinal distance (before the longitudinal positioning algorithm); 3. Road landmarks, which are used to calculate the longitudinal error of the vehicle. They are not on the same level as the ground, so it is necessary to record the installation parameters of the camera sensor, including the height and viewing angle of the camera from the ground when it is installed. Affine transformation is performed through these two parameters to calculate the horizontal distance from the camera to the landmark.

[0069] The compensation distance is then compensated to the distance of the marker perceived by the perception system. Next, the marker needs to be filtered. Longitudinal positioning will only begin when the distance between the marker and the vehicle is less than a certain threshold.

[0070] Step 302 , convert the coarse positioning information from the vehicle body coordinate system to the perception system coordinate system, and then to the Odometry coordinate system.

[0071] In a possible implementation manner, in step 302, the nearest point search is performed by using a KD (K-dimensional tree) and / or a divide-and-conquer method.

[0072] KD tree is a data structure that divides data points in K-dimensional space. It is essentially a balanced binary tree. KD tree is a space partition tree that divides the space to be searched (here is a two-dimensional plane) into several parts and searches in specific parts of the space. Figure 3The figure shows a flow chart of a KD tree search provided by an embodiment of the present invention. First, the marker coordinates within a certain range are placed in the same coordinate system, the variance of these coordinates in the horizontal and vertical coordinates are calculated, the point with the larger variance is selected as the segmentation domain, the median of these data is selected as the segmentation threshold, and these points are divided into two left and right subtrees. The cycle is repeated. After the KD tree is built, the distances of the points contained in the area of ​​this point are compared to find the point with the smallest distance.

[0073] Step 303: Search for the map identifier of the landmark closest to the coarse positioning information in the high-precision map in the Odometry coordinate system and save the landmark ID.

[0074] Step 304, transfer the found landmark map identification to the perception system coordinate system, record the identification distance of each landmark map identification from the vehicle and the corresponding landmark ID, and save them in a container with time as a variable.

[0075] Step 305: When the nearest identification distance is less than the threshold and the data meets the set requirements, outliers are screened out through MAD-based Z-score anomaly detection.

[0076] MAD is the median absolute deviation, and the calculation formula is:

[0077]

[0078] Z-score determines whether the data deviates from the normal range by calculating the standard deviation and mean of the data.

[0079] Step 306, performing a least squares linear fit on the data obtained in step 305, and using the linear fit to calculate the mean square error between the current position and the true position to compensate for the longitudinal position.

[0080] like Figure 4 The figure shows a schematic diagram of a least square method provided by an embodiment of the present invention. In the specific implementation process, the linear fitting process includes: establishing a linear regression model between the camera perception data and the distance data on the map, so that the selected regression model should minimize the sum of all residual squares. The residual is the difference between the predicted value (upstream positioning result) and the observed value (camera sensor data), and the minimum residual sum of squares is to find the minimum error. The minimum residual sum of squares can be used to obtain the optimal parameter value of the fitting curve. The formula is:

[0081]

[0082] Finally, the minimum residual sum of squares, that is, MSE (mean square error), is used as the optimal longitudinal correction distance. Through the above steps, the predicted value and the actual situation of the observed value can be combined to the greatest extent to obtain the longitudinal correction with the minimum error. Not only can the longitudinal position of the vehicle be corrected by the observation of the camera data, but also the error in the camera data can be considered and compensated by the predicted value.

[0083] In a possible implementation manner, the type of the marker acquired in step 1 includes: a lane line; and the marker information includes: the type and position of the lane line.

[0084] Step 3 also includes: correcting the lateral positioning information of the coarse positioning information based on the lateral position information of the lane line.

[0085] The present invention provides a vehicle longitudinal positioning method based on the least squares method, which utilizes lane lines, ground signs and road markers observed by a camera sensor system to correct the longitudinal positioning with errors provided by a combined navigation system, so as to correct the longitudinal position of the vehicle during driving. The longitudinal positioning accuracy in occluded and semi-occluded scenes is improved, which helps meet the overall positioning accuracy requirements and has practical significance in the field of autonomous driving.

[0086] Example 2

[0087] Embodiment 2 provided by the present invention is an embodiment of a vehicle longitudinal positioning system based on the least squares method provided by the present invention. Figure 5 A structural diagram of a vehicle longitudinal positioning system based on the least squares method provided by an embodiment of the present invention, combined with Figure 5 It can be seen that the embodiment of the positioning system includes:

[0088] Combined navigation system, perception system, high-precision map and longitudinal positioning algorithm modules.

[0089] The integrated navigation system is used to obtain the rough positioning information of the vehicle.

[0090] In a specific implementation, the integrated navigation system includes position information obtained by fusion of RTK (real-time kinematic positioning) and IMU (inertial measurement unit) sensors.

[0091] The perception system allows users to obtain information about side markers while the vehicle is driving, and obtain longitudinal relative position information between the vehicle and the markers.

[0092] The perception system can be a camera sensor or a LiDAR perception system, which senses lane lines, road signs and other information, and transmits it to the image algorithm in the camera perception system or LiDAR perception system to obtain the perception information of the landmarks. LiDAR, like cameras, is a popular perception sensor at present. Its accuracy in identifying lane lines and traffic signs is better than that of cameras, and it has stronger anti-interference ability in bad weather.

[0093] In specific implementation, the high-precision map includes the real location information of map logos and landmarks, as well as the topological structure of the road and other relevant geographic information, which are searched and matched with the camera perception system.

[0094] The longitudinal positioning algorithm module uses a high-precision map to load map information near the coarse positioning information, and determines the absolute position information of the marker based on the map information; based on the absolute position of each marker and its longitudinal relative position information with the vehicle, the least squares method is used for linear fitting to obtain the longitudinal position information of the vehicle, and the longitudinal positioning information in the coarse positioning information is corrected using the vehicle longitudinal position information.

[0095] The present invention combines a navigation system, a camera perception system and a high-precision map. The combined navigation system provides an initial position to a positioning matching system based on a coarse-precision position obtained by fusing RTK and IMU sensors. The camera perception system is used to transmit lane lines, road markers and ground marker information during vehicle driving to the positioning system. The high-precision map provides map identification and real position information of markers. The algorithm module searches and matches the marker closest to the vehicle on the map, records a series of distances from the marker (two types: the distance observed by the camera and the distance of the vehicle on the map), selects relatively abnormal values, and relies on the least squares method to estimate the minimum error of the longitudinal deviation. The minimum mean square error obtained is compensated to the longitudinal result of positioning, thereby realizing high-precision positioning of the longitudinal position of the vehicle. The system improves the driving safety and stability of the vehicle, provides more reliable navigation and position information for the driver, and can be widely used in fields such as autonomous driving and traffic management.

[0096] It can be understood that the vehicle longitudinal positioning system based on the least squares method provided by the present invention corresponds to the vehicle longitudinal positioning method based on the least squares method provided by the aforementioned embodiments. The relevant technical features of the vehicle longitudinal positioning system based on the least squares method can refer to the relevant technical features of the vehicle longitudinal positioning method based on the least squares method, which will not be repeated here.

[0097] See also Figure 6 , Figure 6 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 6As shown, an embodiment of the present invention provides an electronic device, including a memory 1310, a processor 1320, and a computer program 1311 stored in the memory 1310 and executable on the processor 1320. When the processor 1320 executes the computer program 1311, the following steps are implemented: obtaining coarse positioning information of the vehicle; obtaining marker information on the side of the vehicle during driving, and obtaining longitudinal relative position information between the vehicle and the marker; using a high-precision map to load map information near the coarse positioning information, and determining the absolute position information of the marker based on the map information; based on the absolute position of each marker and its longitudinal relative position information with the vehicle, using the least squares method to perform linear fitting to obtain the longitudinal position information of the vehicle, and using the vehicle longitudinal position information to correct the longitudinal positioning information in the coarse positioning information.

[0098] See also Figure 7 , Figure 7 Schematic diagram of an embodiment of a computer-readable storage medium provided by the present invention. Figure 7 As shown, this embodiment provides a computer-readable storage medium 1400, on which a computer program 1411 is stored. When the computer program 1411 is executed by a processor, the following steps are implemented: obtaining coarse positioning information of the vehicle; obtaining marker information on the side of the vehicle during driving, and obtaining longitudinal relative position information between the vehicle and the marker; using a high-precision map to load map information near the coarse positioning information, and determining the absolute position information of the marker based on the map information; based on the absolute position of each marker and its longitudinal relative position information with the vehicle, using the least squares method to perform linear fitting to obtain the longitudinal position information of the vehicle, and using the vehicle longitudinal position information to correct the longitudinal positioning information in the coarse positioning information.

[0099] The least squares method-based vehicle longitudinal positioning method, system, electronic device, and storage medium provided by the embodiments of the present invention have the following advantages:

[0100] (1) The present invention can correct and eliminate positioning errors in the longitudinal position on complex urban roads (including occluded and semi-occluded scenes) based on the coarse positioning information provided by the integrated navigation system through lane lines, ground signs and road marker information.

[0101] (2) The longitudinal positioning method of the present invention can quickly find the nearest landmark information and match it. The search method used is of low complexity, high search efficiency and reliability. The longitudinal distance is linearly fitted by the least squares method, which can improve the accuracy of the longitudinal error.

[0102] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0103] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0104] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0105] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0106] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0107] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0108] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A vehicle longitudinal positioning method based on least squares method, characterized in that: The positioning method comprises: Step 1, obtaining the rough positioning information of the vehicle; obtaining the information of the markers on the side of the vehicle during the driving process, and obtaining the longitudinal relative position information of the vehicle and the markers; Step 2, using a high-precision map to load map information near the rough positioning information, and determining the absolute position information of the marker based on the map information; Step 3, based on the absolute position of each marker and its longitudinal relative position information with the vehicle, use the least squares method to perform linear fitting to obtain the longitudinal position information of the vehicle, and use the longitudinal position information of the vehicle to correct the longitudinal positioning information in the coarse positioning information.

2. The positioning method according to claim 1, characterized in that: The process of obtaining the rough positioning information of the vehicle in step 1 includes: The positioning result is obtained by fusing RTK and IMU, and then the positioning result is passed through a Kalman filter to obtain the coarse positioning information.

3. The positioning method according to claim 1, characterized in that: The types of the markers obtained in step 1 include: ground markers and road markers; the marker information includes: the content and location of the marker; When the marker is a ground marker, the longitudinal distance between the sensing system and the marker is used as the longitudinal relative position information between the vehicle and the marker; When the marker is a road marker, the longitudinal relative position information of the vehicle and the marker is calculated based on the installation parameters of the perception system and the observed longitudinal distance to the marker. The installation parameters include: the height and viewing angle of the installation position of the perception system relative to the ground.

4. The positioning method according to claim 3, characterized in that: The step 1 also includes: calculating a compensation distance according to a timestamp difference between the marker and the current position and the wheel speed, and compensating the calculated longitudinal relative position information based on the compensation distance.

5. The positioning method according to claim 1, characterized in that: The process of calculating the longitudinal position information of the vehicle in step 3 and using the longitudinal position information of the vehicle to correct the longitudinal positioning information in the rough positioning information includes: Step 301, traverse and classify the landmarks sensed by the perception system and store them; Step 302, converting the coarse positioning information from the vehicle body coordinate system to the perception system coordinate system, and then to the Odometry coordinate system; Step 303, searching for a landmark map identifier closest to the rough positioning information in the high-precision map in the Odometry coordinate system and saving the landmark ID; Step 304, transferring the found landmark map identification to the perception system coordinate system, recording the identification distance of each landmark map identification from the vehicle and the corresponding landmark ID, and saving them in a container with time as a variable; Step 305, when the nearest identification distance is less than the threshold and the data meets the set requirements, outliers are screened out through MAD-based Z-score anomaly detection; Step 306: Perform a least squares linear fit on the data obtained in step 305, and use the linear fit to calculate the mean square error between the current position and the true position to compensate for the longitudinal position.

6. The positioning method according to claim 1, characterized in that: In step 302, the nearest point search is performed using KD and / or divide-and-conquer methods.

7. The positioning method according to claim 1, characterized in that: The type of the marker acquired in step 1 includes: a lane line; the marker information includes: the type and position of the lane line; The step 3 also includes: correcting the lateral positioning information of the coarse positioning information based on the lateral position information of the lane line.

8. A vehicle longitudinal positioning system based on the least squares method, characterized in that: The positioning system includes: an integrated navigation system, a perception system, a high-precision map and a longitudinal positioning algorithm module; The integrated navigation system is used to obtain rough positioning information of the vehicle; The perception system allows the user to obtain information about the markers on the side of the vehicle during driving, and obtain longitudinal relative position information between the vehicle and the markers; The longitudinal positioning algorithm module uses the high-precision map to load map information near the coarse positioning information, and determines the absolute position information of the marker based on the map information; based on the absolute position of each marker and its longitudinal relative position information with the vehicle, the least squares method is used for linear fitting to obtain the longitudinal position information of the vehicle, and the longitudinal positioning information in the coarse positioning information is corrected using the vehicle longitudinal position information.

9. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the processor is used to implement the steps of the vehicle longitudinal positioning method based on the least squares method as described in any one of claims 1 to 7 when executing a computer management program stored in the memory.

10. A computer-readable storage medium, characterized in that: A computer management program is stored thereon, and when the computer management program is executed by a processor, the steps of the vehicle longitudinal positioning method based on the least squares method as described in any one of claims 1 to 7 are implemented.

Images