A mapping method, device, equipment and medium based on pier stone pier

The method and device use pier stones for real-time mapping and optimization to address precision and real-time challenges in vehicle positioning, enhancing navigation accuracy.

CN115183778BActive Publication Date: 2024-12-03BEIJING SENIOR SMART DRIVING TECHNOLOGY CO LTD
View PDF 27 Cites 0 Cited by

Patent Information

Application Number
CN202210774511.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-07-01
Publication Date
2024-12-03
Estimated Expiration
2042-07-01

AI Technical Summary

Technical Problem

Existing vehicle positioning technologies, such as GPS and laser sensors, struggle with precision and real-time performance, especially in obstructed environments, failing to meet high-precision navigation needs.

Method used

A method and device that utilize pier stones for real-time mapping by initializing a pier map and factor graph, integrating laser radar data, and performing nonlinear optimization to establish precise vehicle position and posture.

Benefits of technology

Enables efficient and accurate mapping of pier stones for navigation, overcoming precision and real-time limitations of existing technologies.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115183778B_ABST
    Figure CN115183778B_ABST
Patent Text Reader

Abstract

The present invention is a mapping method based on pier stones, step 1-initialize the pier map M and factor graph G; step 2-calculate the current vehicle's posture; step 3-acquire laser radar observation data and extract pier data; step 4-correspond each pier data to the pier map M and establish data association; step 5-add relevant factors to the factor graph G according to the vehicle posture estimation value and each pier position value; step 6-perform nonlinear optimization solution on the factor graph G to obtain each pier position and update the pier map M; step 7-after there is new laser radar observation data, repeat steps 3-6; if not, end the process; step 8-after the process ends, the final pier map M is the desired result. The present invention uses a factor graph for optimization and solution, and the factors are divided into two categories, one is the posture estimation of the vehicle itself as the prior posture factor of the vehicle, and the other is the observation and extraction of the pier data as the observation position factor of the pier.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a mapping method, device, equipment and medium based on pier stone piers. Background Art

[0002] With the development of driverless technology, automated terminal transportation has gradually become possible. During the transportation process, higher requirements are placed on the vehicle positioning technology. The traditional solution is based on GPS (Global Positioning System) to directly obtain positioning information. In addition, there are also positioning solutions based on existing maps, which usually use vision and laser sensors to pre-map the site, and then use existing maps and real-time observations to estimate the position during operation.

[0003] The GPS-based solution needs to maintain good communication with the satellite. In the case of certain building obstructions, its positioning results will be significantly affected. In addition, the current civilian GPS positioning accuracy is generally at the meter level, which cannot meet the needs of high-precision positioning. Among the positioning solutions based on existing maps, laser sensors are the current mainstream solution because their accuracy and stability are better than those of visual sensors. However, there are also situations where the calculation is complex and the real-time performance is low. Especially for multi-line laser radar, the calculation takes more time because of the large amount of data. Summary of the invention

[0004] The object of the present invention is to provide a method, device, equipment and medium for mapping based on pier stones, which can map pier stones in real time and efficiently to facilitate subsequent navigation control.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A mapping method based on pier stones, characterized in that step 1-initializes a pier map M and a factor graph G; step 2-calculates the current vehicle's position and posture; step 3-acquires laser radar observation data and extracts pier data; step 4-corresponds each pier data to the pier map M and establishes data association; step 5-adds relevant factors to the factor graph G according to the estimated value of the vehicle's position and the position value of each pier; step 6-performs nonlinear optimization solution on the factor graph G to obtain the position of each pier, thereby updating the pier map M; step 7-after there is new laser radar observation data, repeat steps 3-6; if not, end the process; step 8-after the process ends, the final pier map M is the desired result.

[0007] The present invention is further configured that, in an initial state, the stone pier map M is empty and the factor graph G is empty.

[0008] The present invention is further configured to calculate the current vehicle posture according to the GPS and the vehicle's IMU data, and at time t, the vehicle's posture under the global map is calculated as R t .

[0009] The present invention further arranges that the positions of the stone pillars are S={s1, s2, ..., s k}, where the position of the pier is determined by the vehicle coordinate system.

[0010] The present invention further arranges that for each stone pier s i , if you find a stone m within H meters of it in the stone mound map M j , then si and m j Data association occurs. If not found, insert si into the stone pier map M, and set the newly inserted stone pier to m. u , then s i With m u Data association is generated.

[0011] The present invention further provides that when adding related factors to the factor graph G, the factor graph is inserted with a priori factor X=(X t , R t ), and the set of observation factors associated with the stone pier: L = {(X t , L U(i) , S i )|i=1,2,3,...,k},where U(i) represents the number of cells in M ​​that are related to s i The number of the pier where the data association occurs.

[0012] The present invention also provides a mapping device based on a wharf pier, comprising:

[0013] Mapping module, initializing the stone pier map M, factor graph G, updating the stone pier map M, and forming the final stone pier map;

[0014] The calculation module calculates the current vehicle position based on the GPS and vehicle IMU data, and calculates the vehicle's position on the global map as R t ;

[0015] LiDAR observation module, used to obtain observation data;

[0016] The data extraction module obtains the observation data from the laser radar observation module and extracts the stone pier data. The position of each stone pier is recorded as S = {s1, s2, ..., s k};

[0017] The data association module is used to correspond each stone pier data to the stone pier map M and establish data association; the factor addition module adds the estimated value of the vehicle posture and the position value of each stone pier to the factor map G;

[0018] Data optimization module: Perform nonlinear optimization on the factor graph G to obtain the position of each stone pier, and update the stone pier map M accordingly.

[0019] The present invention also provides an electronic device, characterized in that it includes: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement a mapping method.

[0020] The present invention also provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores computer-executable instructions, and when the instructions are executed, the computer performs the mapping method.

[0021] Beneficial effects of the present invention: This algorithm uses a factor graph to perform optimization and solution, in which the factors are divided into two categories, one is the posture estimation of the vehicle itself as the prior posture factor of the vehicle, and the other is the observation and extraction of the stone pier as the observation position factor of the stone pier, and then the factor graph is optimized and solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0023] Figure 1 is a flow chart of an embodiment of the present invention.

[0024] Figure 2 This is an example of the factor graph structure of the present invention. DETAILED DESCRIPTION

[0025] The following will describe the implementation methods of the present application in detail with the help of accompanying drawings and examples, so that the implementation process of how the present application uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.

[0026] Deep learning is used to identify the location of each pier in each frame of laser observation, and then the pier location data of all laser frames are integrated to create a pier map of the entire dock.

[0027] During the whole process, a stone pier map M is maintained, which records the position of each stone pier under the global map. In addition, we also maintain a factor graph G, which is a NonlinearFactorGraph instance of the gtsam library, for solving. The overall steps are as follows, and the flowchart can be found in Figure 1 :

[0028] 1. Initially, the stone pier map M is empty and the factor graph G is empty.

[0029] 2. Based on the GPS and vehicle IMU data, the current vehicle position is calculated. We denote it as: At time t, the vehicle position calculated on the global map is R t .

[0030] 3. Obtain the observation data of the laser radar and extract the data of each stone pillar from the observation data of the radar. We record it as: the position of each stone pillar is s = {s1, s2, ..., s k}, where the position of the pier is relative to the vehicle coordinate system.

[0031] 4. Correspond each pier data to the pier map M and establish data association. Specifically: for each pier s i , if you find a stone m within H meters of it in the stone mound map M j , then s i and m j A data association has occurred. If not found, s i Insert it into the stone pier map M, and let the newly inserted stone pier be m u , then s i With m u Data association is generated.

[0032] 5. According to the estimated value of the vehicle posture in step 2 and the position value of each stone pier in step 3, add relevant factors to the factor graph G. Specifically, we insert the prior factor X = (X t , R t ), and the set of observation factors associated with the stone pier: L = {(X t , L U(i) ,s i )|i=1,2,3,...,k},where U(i) represents the number of cells in M ​​that are related to s i The number of the pier where the data association occurs.

[0033] 6. Perform nonlinear optimization on the factor graph G to obtain the position of each stone pier and update the stone pier map M accordingly.

[0034] 7. When there is new laser observation data, repeat steps 3-6. Otherwise, end the process.

[0035] 8. After the process is completed, the final stone pier map M is the desired result.

[0036] The present invention also provides a mapping device based on a wharf pier, comprising:

[0037] Mapping module, initializing the stone pier map M, factor graph G, updating the stone pier map M, and forming the final stone pier map;

[0038] The calculation module calculates the current vehicle position based on the GPS and vehicle IMU data, and calculates the vehicle's position on the global map as R t ;

[0039] LiDAR observation module, used to obtain observation data;

[0040] The data extraction module obtains the observation data in the laser radar observation module and extracts the stone pier data. The position of each stone pier is recorded as S = {s1, s2, ..., s k};

[0041] The data association module is used to correspond each stone pier data to the stone pier map M and establish data association; the factor addition module adds the estimated value of the vehicle posture and the position value of each stone pier to the factor map G;

[0042] Data optimization module: Perform nonlinear optimization on the factor graph G to obtain the position of each stone pier, and update the stone pier map M accordingly.

[0043] The present invention also provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement a mapping method.

[0044] The present invention also provides a computer-readable storage medium, in which computer-executable instructions are stored. When the instructions are executed, the computer performs the mapping method.

[0045] The present invention uses a factor graph to perform optimization and solution, wherein the factors are divided into two categories, one is the posture estimation of the vehicle itself as the prior posture factor of the vehicle, and the other is the observation and extraction of the stone pier as the observation position factor of the stone pier. Then the factor graph is optimized and solved. The commonly used library of the factor graph is gtsam, and some optimization libraries such as ceres can also be used for independent implementation; when doing data management, the position is used as the judgment standard, and other solutions such as triangular matching can also be used. Identifying stone piers is not limited to a certain deep learning algorithm.

[0046] For example, certain words are used in the specification and claims to refer to specific components. Those skilled in the art should understand that hardware manufacturers may use different terms to refer to the same component. This specification and claims do not use differences in names as a way to distinguish components, but use differences in the functions of components as the criteria for distinction. For example, "including" mentioned throughout the specification and claims is an open term, so it should be interpreted as "including but not limited to". "Approximately" means that within an acceptable error range, those skilled in the art can solve the technical problem within a certain error range and basically achieve the technical effect.

[0047] It should be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a product or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such product or system. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the product or system including the elements.

[0048] The above description shows and describes several preferred embodiments of the present invention, but as mentioned above, it should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the invention concept described herein through the above teachings or the technology or knowledge of the relevant field. Changes and variations made by those skilled in the art do not depart from the spirit and scope of the present invention, and should be within the scope of protection of the claims attached to the present invention.

Claims

1. A mapping method based on pier stone piers, characterized in that: Step 1- Initialize the stone pile map M and factor graph G; Step 2-Calculate the current vehicle's position; Step 3-acquire lidar observation data and extract stone pier data; Step 4: Correspond each pier data to the pier map M and establish data association, specifically: for each pier s i , if you find a stone m within H meters of it in the stone mound map M j , then s i and m j A data association occurs. If not found, s i Insert it into the stone pier map M, and let the newly inserted stone pier be m u , then s i With m u Data association is generated; Step 5-Add relevant factors to the factor graph G according to the estimated values ​​of the vehicle posture and the position values ​​of each stone pier; Step 6: Perform nonlinear optimization on the factor graph G to obtain the position of each stone pier, and update the stone pier map M accordingly; Step 7: Repeat steps 3-6 when new lidar observation data is available. If not, end the process; Step 8-After the process is completed, the final stone pier map M is the desired result.

2. A method for mapping based on pier stones as claimed in claim 1, characterized in that: In the initial state, the stone pier map M is empty and the factor graph G is empty.

3. A method for mapping based on pier stone piers as claimed in claim 2, characterized in that: When calculating the current vehicle's position, the vehicle's position on the global map is calculated based on the GPS and vehicle's IMU data, and at time t, the vehicle's position is calculated as R t .

4. A method for mapping based on pier stone piers as claimed in claim 3, characterized in that: The position of each stone pillar is S = {s1, s2, ..., sk}, where the position of the stone pillar is determined by the vehicle coordinate system.

5. The method for mapping based on pier stone piers as claimed in claim 1, characterized in that: When adding related factors to the factor graph G, the factor graph inserts the prior factor X = (X t ,R t ), and the set of observation factors associated with the stone pier: L = {(X t ,L U(i) ,s i )|i=1,2,3,...,k}, where X t represents the prior factor at time t, R t represents the position of the vehicle in the global map at time t, and U(i) represents the position of the vehicle in M ​​with s i The number of the pier where the data association occurs, L U(i) Represents the observation factor of the pier number U(i).

6. A mapping device based on pier stone piers, characterized in that: include: Mapping module, initializing the stone pier map M, factor graph G, updating the stone pier map M, and forming the final stone pier map; The calculation module calculates the current vehicle position based on the GPS and vehicle IMU data, and calculates the vehicle's position on the global map as R t ; LiDAR observation module, used to obtain observation data; The data extraction module obtains the observation data in the laser radar observation module, extracts the stone pier data, and records the position of each stone pier as S = {s1, s2, ..., sk}; The data association module is used to correspond each pier data to the pier map M and establish data association, specifically: for each pier s i , if you find a stone m within H meters of it in the stone mound map M j , then s i and m j A data association occurs. If not found, s i Insert it into the stone pier map M, and let the newly inserted stone pier be m u , then s i With m u Data association is generated; Factor adding module, the estimated value of vehicle posture and the position value of each stone pier, add relevant factors in the factor graph G; Data optimization module: Perform nonlinear optimization on the factor graph G to obtain the position of each stone pier, and update the stone pier map M accordingly.

7. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; the processor executes the computer-executable instructions stored in the memory to implement the mapping method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the instructions are executed, the computer executes the mapping method according to any one of claims 1 to 5.

Citation Information

Patent Citations

  • Parking lot map construction method and system, mobile terminal and storage medium

    CN108959321A

  • Simultaneous positioning mapping method and device

    CN110751722A

  • Mobile robot synchronous positioning and map building method

    CN110895146A

  • Map construction method and device based on laser radar

    CN111856499A

  • Laser mapping positioning method and device, storage medium and electronic equipment

    CN112068154A