Map construction method and device and electronic equipment

Abstract

The embodiment of the invention discloses a map construction method, which comprises the following steps of: determining a loading range according to the current position of an AGV (Automatic Guided Vehicle), and loading preliminary constraint data corresponding to the loading range; performing back-end graph optimization processing on the preliminary constraint data; and constructing and / or updating a global map of the AGV in real time according to a back-end graph optimization result. The map construction method provided by the embodiment of the invention aims to alleviate the problem of low memory consumption of real-time storage of memory data, reduce the overall mapping back-end calculation amount, realize high-precision mapping under a large-scene map, and maintain low memory and low computing power consumption at the same time. The embodiment of the invention further provides a map construction device, electronic equipment and a storage medium.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of intelligent warehousing. Specifically, the embodiments relate to a method, device, and electronic device for map construction. Background Art

[0002] In the field of warehousing logistics, 2D laser SLAM (Simultaneous Localization and Mapping) technology has become a key means for devices such as AGVs (Automated Guided Vehicles) to construct maps and perform positioning and navigation. However, with the continuous expansion of the warehousing scale and the significant increase in the cargo storage density, the mapping environment gradually exhibits the characteristics of high-density constraints.

[0003] Specifically, obstacles such as densely distributed shelves and mountains of goods in the warehouse cause an explosive growth in the number of constraint data (such as feature points, edge constraints, etc.) collected by the lidar during the scanning process.

[0004] Traditional mapping methods often adopt the strategy of calling all constraint data and performing unified calculations on all data in the backend graph optimization stage when dealing with high-density constraint data. However, this approach has many drawbacks: First, the massive constraint data will over-occupy memory resources, and in the construction of large-scale maps, it is extremely easy to cause memory shortage problems, which in turn affect the stable operation of the entire system. Second, the amount of data that needs to be processed in the backend graph optimization is too large, which not only significantly increases the computing power cost but also leads to low mapping efficiency and is difficult to meet the strict requirements for real-time performance in practical applications. Summary of the Invention

[0005] In view of the problems existing in the above-mentioned prior art, the embodiments of the present application provide a method, device, electronic device, and storage medium for map construction, aiming to alleviate the problem of low memory consumption for real-time storage of in-memory data, reduce the overall backend calculation amount of mapping, achieve high-precision mapping in large-scale maps, and at the same time maintain low memory and low computing power consumption.

[0006] In a first aspect, the embodiments of the present application provide a method for map construction, including the following steps: Determine a loading range according to the current position of the AGV, and load preliminary constraint data corresponding to the loading range; Perform backend graph optimization processing on the preliminary constraint data; and According to the result of the backend graph optimization, construct and / or update the global map of the AGV in real time.

[0007] Further, the performing backend graph optimization processing on the preliminary constraint data includes: Based on the pose nodes and constraint edges of the AGV, the pose estimation and map model of the AGV are optimized by minimizing the error function.

[0008] Further, the error function to be minimized is based on the reprojection error and pose error of the AGV.

[0009] Further, determining the loading range according to the current position of the AGV and loading the preliminary constraint data corresponding to the loading range includes: Setting the area within a preset distance centered on the current position of the AGV as the loading range; and According to the current position and loading range of the AGV, loading the preliminary constraint data corresponding to the loading range, where the preliminary constraint data includes odometer constraint data and loop detection constraint data.

[0010] Further, before determining the loading range according to the current position of the AGV and loading the preliminary constraint data corresponding to the loading range, it further includes: Determining the mapping area and planning the movement path of the AGV; and According to the collected laser point cloud data, performing preprocessing and generating the preliminary constraint data, and adding a position identifier to the preliminary constraint data.

[0011] Further, the position identifier is the position information of the AGV in the global coordinates.

[0012] Further, constructing and / or updating the global map of the AGV in real time according to the result of the back-end map optimization includes: Combining the laser point cloud data and the result of the back-end map optimization to construct the global map; and According to changes in the external environment, updating the global map in real time.

[0013] In a second aspect, an embodiment of the present application provides a map construction device, including: A constraint data loading module, configured to determine a loading range according to the current position of the AGV and load the preliminary constraint data corresponding to the loading range; A back-end map optimization module, configured to perform back-end map optimization processing on the preliminary constraint data; and A map construction module, configured to construct and / or update the global map in real time according to the result of the back-end map optimization.

[0014] In a third aspect, an embodiment of the present application further provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor is configured to implement the map construction method according to the first aspect described above when executing the program.

[0015] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and the computer program is used to implement the map construction method according to the first aspect described above.

[0016] The embodiments of the present application bring the following beneficial effects: In the map construction method provided by the embodiments of the present application, first, a local loading range is delimited according to the current position of the AGV, and only the preliminary constraint data within this range is loaded to avoid global data redundancy. Then, the loaded preliminary constraint data is converted into a graph model, and backend graph optimization processing is performed. The optimized local map is incrementally fused with the global map to achieve dynamic expansion and real-time construction and / or update of the global map of the AGV. In the map construction method provided by the embodiments of the present application, only the preliminary constraint data near the current position is loaded into the memory when needed, avoiding a large amount of constraint data occupying the memory simultaneously, effectively alleviating the memory pressure, and is particularly suitable for mapping tasks in large-scale and high-density constraint environments. In addition, the backend graph optimization only processes the constraint data near the current position, which can reduce the amount of calculation, lower the requirement for computing power, improve the mapping efficiency, and meet the real-time mapping requirements. Moreover, dynamically loading constraint data can ensure that effective constraints related to the current position are used during the graph optimization process, improve the accuracy of pose estimation, and thus achieve high-precision map construction, providing a reliable map basis for the precise positioning and navigation of warehousing and logistics equipment. Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on the structures shown in these drawings.

[0018] Figure 1 It is a flowchart of a map construction method provided by an embodiment of the present application; Figure 2 It is a structural block diagram of a map construction device provided by an embodiment of the present application; Figure 3 It is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

[0019] The realization of the purpose of the present application, functional features, and advantages will be further described in conjunction with the embodiments and with reference to the drawings. Detailed Embodiments

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the protection scope of the present application.

[0021] In the description of the present application, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present application, unless otherwise specified, the meaning of "a plurality" is two or more. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances.

[0022] The following describes a map construction method, apparatus, electronic device, and storage medium according to an embodiment of the present application with reference to the accompanying drawings.

[0023] Figure 1 is a flowchart of a map construction method according to an embodiment of the present application. As Figure 1 shown, the map construction method of the embodiment of the present application includes the following steps: S101: Determine a loading range according to the current position of the AGV, and load preliminary constraint data corresponding to the loading range; Specifically, during the map construction process, the current position of the AGV is obtained in real time. The current position of the AGV can be obtained through odometer, IMU (Inertial Measurement Unit) fusion, or other positioning methods. According to the current position of the AGV, a loading range is determined. For example, a circular area with the current position as the center and a radius of R. All constraint data within this loading range is read from the hard disk and loaded into the memory. As the AGV moves, the current position is updated in real time, and the loading range is continuously adjusted to ensure that only the preliminary constraint data near the current position is stored in the memory all the time.

[0024] In an embodiment of the present application, the AGV obtains the current position (x_current, y_current) through odometer and IMU fusion. The loading range radius R is set to 10 meters. All preliminary constraint data files within a range of 10 meters centered on (x_current, y_current) are read from the hard disk and loaded into the memory. During loading, the preliminary constraint data is parsed and verified to ensure the validity of the data.

[0025] As the AGV moves, its current position is updated every 0.5 seconds, and the loading range is recalculated. Preliminary constraint data outside the range is unloaded, and new preliminary constraint data that enters the range is loaded to maintain the timeliness and relevance of the preliminary constraint data in memory.

[0026] S102: Perform backend graph optimization processing on the preliminary constraint data; and That is, using the preliminary constraint data near the current position loaded into memory, perform backend graph optimization calculations. After each graph optimization is completed, save the optimized pose nodes and map information, and at the same time update the relevant information of the preliminary constraint data in the hard disk as needed, such as the validity of the constraint data, associated pose nodes, etc.

[0027] S103: According to the backend graph optimization result, construct and / or update the global map of the AGV in real time.

[0028] That is, according to the result of the backend graph optimization, gradually construct and / or update the global map. When constructing the map, combine the laser point cloud data and the optimized pose information to generate an accurate 2D grid map or feature map.

[0029] Therefore, in the map construction method provided in the embodiments of the present application, first, a local loading range is delimited according to the current position of the AGV, and only the preliminary constraint data within this range is loaded to avoid global data redundancy. Then, the loaded preliminary constraint data is converted into a graph model, and backend graph optimization processing is performed. The optimized local map is incrementally fused with the global map to achieve dynamic expansion, and the global map of the AGV is constructed and / or updated in real time. The map construction method provided in the embodiments of the present application only loads the preliminary constraint data near the current position into memory when needed, avoiding a large amount of constraint data from occupying memory at the same time, effectively alleviating the memory pressure, and is especially suitable for mapping tasks in large-scene and high-density constraint environments. In addition, the backend graph optimization only processes the constraint data near the current position, which can reduce the amount of calculation, lower the requirement for computing power, improve the mapping efficiency, and meet the real-time mapping requirements. Moreover, the dynamic loading of constraint data can ensure the use of effective constraints related to the current position during the graph optimization process, improve the accuracy of pose estimation, and thus achieve high-precision map construction, providing a reliable map basis for the precise positioning and navigation of warehousing and logistics equipment.

[0030] Further, in some embodiments of the present application, the performing backend graph optimization processing on the preliminary constraint data includes: Based on the pose nodes and constraint edges of the AGV, optimize the pose estimation and map model of the AGV by minimizing the error function.

[0031] Specifically, for example, the g2o general graph optimization framework is used as the backend graph optimization tool to construct a graph optimization model from the preliminary constraint data loaded into the memory. The pose nodes are the poses (x, y, θ) of the AGV at different times, and the constraint edges are the odometry constraints and loop closure detection constraints between adjacent pose nodes. The error function is defined as the difference between the observed value and the predicted value of each constraint, and the Levenberg-Marquardt algorithm is used for iterative optimization to minimize the overall error, obtaining the optimized pose nodes and map parameters. After each graph optimization is completed, the optimized pose nodes and map information are saved to the global map file, and at the same time, the associated pose node information of the constraint data in the hard disk is updated for subsequent loading and use. Therefore, in the map construction method provided by the embodiments of the present application, through the backend graph optimization framework and the dynamic loading constraint data strategy, the efficient joint optimization of AGV pose estimation and map construction is realized. While ensuring the accuracy, it can reduce the memory and computing power requirements, providing reliable technical support for real-time positioning and map construction in complex scenarios.

[0032] Further, in some embodiments of the present application, the minimizing the error function is based on the reprojection error and pose error of the AGV.

[0033] Specifically, in tasks such as visual SLAM (Simultaneous Localization and Mapping) or 3D reconstruction, minimizing the error function is the core method for optimizing the camera pose (position and orientation) and the position of map points. The error function usually includes various error terms such as reprojection error and pose error. Among them, the reprojection error is the difference between the observed point (pixel coordinates in the image) and the point reprojected through the camera model and the current pose estimation. The pose error is the difference between the pose transformation between adjacent frames and the pose transformation obtained through feature matching or optical flow estimation.

[0034] The reprojection error and pose error are the core error terms in visual SLAM and 3D reconstruction, respectively reflecting the observation accuracy and pose consistency. By minimizing these error functions, the camera pose and the position of map points can be optimized, improving the accuracy and stability of the system. In practical applications, it is necessary to select appropriate error terms for optimization according to the specific scenario and construct a comprehensive optimization objective function in combination with other error terms (such as feature point error) to achieve more accurate positioning and reconstruction.

[0035] Further, in some embodiments of the present application, the determining the loading range according to the current position of the AGV and loading the preliminary constraint data corresponding to the loading range includes: Setting the area within a preset distance range centered on the current position of the AGV as the loading range; and Load preliminary constraint data corresponding to the loading range according to the current position and loading range of the AGV, where the preliminary constraint data includes odometer constraint data and loop detection constraint data.

[0036] Specifically, in the navigation and positioning system of the AGV, the loading range and preliminary constraint data are key elements to ensure positioning accuracy and system efficiency. Centered on the current position of the AGV, a reasonable loading range is dynamically determined to reduce unnecessary data loading and improve system real-time performance. For example, a circular or square area with a radius of R centered on the current position of the AGV is set as the loading range. The odometer constraint data is based on the odometer sensor of the AGV, providing pose constraints at adjacent moments for pose prediction and local optimization in a short period of time. The loop detection constraint data is based on the loop detection results of vision or lidar, providing globally consistent pose constraints for eliminating cumulative errors. Therefore, the map construction method provided by the embodiments of the present application delimits a local loading range according to the current position of the AGV and only loads the preliminary constraint data within this range, which can avoid global data redundancy.

[0037] It should be noted that the constraint data generated per frame (including constraint type, relevant pose nodes, constraint parameters, etc.) will be saved in a specified directory on the hard disk in JSON format, and each constraint data file is named with the corresponding timestamp or position identifier for easy subsequent search and loading. For example, for the constraint data located at the global coordinates (x, y), the file name can be named "constraint_xy.json".

[0038] Furthermore, in some embodiments of the present application, before determining the loading range according to the current position of the AGV and loading the preliminary constraint data corresponding to the loading range, it further includes: Determine the mapping area and plan the movement path of the AGV; and According to the collected laser point cloud data, perform preprocessing and generate the preliminary constraint data, and add a position identifier to the preliminary constraint data.

[0039] Specifically, first, calibrate the 2D lidar and IMU carried by the AGV to ensure the accuracy of sensor data. Configure the parameters of the SLAM system, such as the acquisition frequency of the laser point cloud, the threshold of feature extraction, etc. Divide the working area based on the application scenario, plan the optimal path in combination with the environmental map, clarify the AGV movement trajectory, and provide a spatial framework for data acquisition. For example, according to the layout of the warehousing and logistics center, plan the movement path of the AGV so that it can cover all areas that need to be mapped, including the shelf area, the passage area, the entrance and exit, etc. The AGV moves along the planned path, and the lidar collects the laser point cloud data of the environment at a frequency of, for example, 10 Hz. After collecting the environmental point cloud through the lidar, perform denoising, registration, and feature extraction to eliminate noise and errors, and extract key features (such as edges, planes) to provide a high-quality data basis for constraint generation. Based on the preprocessed point cloud, generate odometry constraints (pose changes between adjacent frames) and loop detection constraints (global feature matching), and associate position tags (coordinates, timestamps) to achieve the spatial positioning of the constraint data. Through the above step-by-step processing, ensure that the AGV system can dynamically load efficient and accurate constraint data during navigation, reduce computational losses, improve the positioning accuracy and system robustness, and lay a foundation for real-time navigation in complex environments.

[0040] Further, in some embodiments of the present application, the position identifier is the position information of the AGV in the global coordinate system.

[0041] Specifically, in the AGV navigation system, the position identifier is the core for associating constraint data with the spatial position. The embodiments of the present application use the position information in the global coordinate system as the identifier. Through the position identifier in the global coordinate system, the AGV system can achieve efficient and accurate loading of constraint data and provide reliable support for navigation in complex scenarios.

[0042] Further, in some embodiments of the present application, the real-time construction and / or update of the global map of the AGV according to the back-end map optimization result includes: Construct the global map by combining the laser point cloud data and the back-end map optimization result; and Update the global map in real time according to changes in the external environment.

[0043] Specifically, in the map construction method provided by the embodiments of the present application, first, a globally consistent map framework is generated by combining the lidar point cloud data with the results of backend map optimization (such as pose graph, loop closure detection constraints). The point cloud data is aligned with the optimized poses to construct a global map that includes three-dimensional structures (such as walls, obstacles) and semantic information (such as passable areas). By continuously collecting point cloud data, external environmental changes (such as newly added obstacles) are detected in real time, and only the affected areas are updated. Combining backend optimization, the poses and point cloud registration are recalculated to ensure global consistency and reduce computational overhead. Therefore, the map construction method provided by the embodiments of the present application improves the accuracy and real-time performance of the map through multi-source data fusion and incremental update mechanisms, enhances the navigation robustness of the AGV in a dynamic environment, and provides reliable support for autonomous operations in complex scenarios.

[0044] In one embodiment of the present application, first, according to the optimized pose nodes and lidar point cloud data, a 2D grid map is generated using a grid map construction algorithm. The resolution of the map is set to 0.05 meters, which can clearly represent structures such as shelves and aisles in the warehousing environment. When the AGV detects new obstacles (such as temporarily stacked goods, handling of goods, and adjustment of shelves) or finds a deviation between the map and the actual environment, the map update mechanism is triggered. The lidar point cloud data of this area is re-collected, new constraint data is generated, and local map optimization is performed to update the global map. Therefore, the map construction method provided by the embodiments of the present application can effectively achieve high-precision construction of large-scale maps in a high-density constrained environment in the warehousing and logistics industry, while maintaining low memory consumption and low computing power consumption, and has great practical value and promotional significance.

[0045] Figure 2 is a structural block diagram of the map construction device 200 according to the embodiments of the present application. As Figure 2 shown, the map construction device 200 according to the embodiments of the present application includes: a constraint data loading module 210, a backend map optimization module 220, and a map construction module 230, where: The constraint data loading module 210 is configured to determine a loading range according to the current position of the AGV and load preliminary constraint data corresponding to the loading range; The backend map optimization module 220 is configured to perform backend map optimization processing on the preliminary constraint data; and The map construction module 230 is configured to construct and / or update a global map in real time according to the results of backend map optimization.

[0046] In the map construction device provided in the embodiments of the present application, first, a local loading range is delimited according to the current position of the AGV, and only the preliminary constraint data within this range is loaded to avoid global data redundancy. Then, the loaded preliminary constraint data is converted into a graph model, and backend graph optimization processing is performed. The optimized local map is incrementally fused with the global map to achieve dynamic expansion and real-time construction and / or update of the global map of the AGV. The map construction method provided in the embodiments of the present application only loads the preliminary constraint data near the current position into the memory when needed, avoiding a large amount of constraint data occupying the memory simultaneously, effectively alleviating the memory pressure, and is particularly suitable for mapping tasks in large-scale and high-density constraint environments. In addition, the backend graph optimization only processes the constraint data near the current position, which can reduce the amount of calculation, lower the requirement for computing power, improve the mapping efficiency, and meet the real-time mapping requirements. Moreover, the dynamic loading of constraint data can ensure that effective constraints related to the current position are used during the graph optimization process, improve the accuracy of pose estimation, and thus achieve high-precision map construction, providing a reliable map basis for the precise positioning and navigation of warehousing and logistics equipment.

[0047] It should be noted that the specific implementation manner of the map construction device in the embodiments of the present application is similar to that of the map construction method in the embodiments of the present application. For details, please refer to the description in the method section and will not be elaborated here.

[0048] Figure 3 It is a schematic structural diagram of the electronic device 300 according to the embodiments of the present application.

[0049] As Figure 3 shown, the electronic device 300 includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage section 302 into the random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the electronic device 300 are also stored. The CPU 301, ROM 302, and RAM 303 are connected to each other through a bus 304. The input / output (I / O) interface 305 is also connected to the bus 304.

[0050] The following components are connected to the I / O interface 305: an input section 306 including a keyboard, a mouse, etc.; an output section 307 including, for example, a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as required. A removable medium 311 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc. is installed on the drive 310 as required so that a computer program read out therefrom is installed into the storage section 308 as required.

[0051] Specifically, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product that includes a computer program carried on a machine-readable medium, and the computer program includes program codes for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through the communication section 309, and / or installed from the removable medium 311. When the computer program is executed by a central processing unit (CPU) 301, the above functions defined in the electronic device of the present application are executed.

[0052] It should be noted that the computer-readable medium shown in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. A computer-readable storage medium may be, for example, but not limited to, an electronic device, apparatus, or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples of the computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0053] In this application, a computer-readable storage medium can be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction-executing electronic device, apparatus, or device. In this application, a computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction-executing electronic device, apparatus, or device. The program code contained on a computer-readable medium can be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination of the above.

[0054] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of a processing receiving device, method, and computer program product according to various embodiments of this application. In this regard, each block in a flowchart or block diagram can represent a module, program segment, or part of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in a block may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented by a dedicated hardware-based electronic device that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0055] The units or modules involved in the embodiments described in this application can be implemented in software or in hardware. The described units or modules can also be provided in a processor, and when the processor executes the program, it implements the map construction method: Determine a loading range according to the current position of the AGV, and load preliminary constraint data corresponding to the loading range; Perform backend graph optimization processing on the preliminary constraint data; and According to the backend graph optimization result, construct and / or update the global map of the AGV in real time.

[0056] As another aspect, the present application also provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiments; or may exist separately without being assembled into the electronic device. The above computer-readable storage medium stores one or more programs, and when the foregoing programs are executed by one or more processors to perform the map construction method described in the present application: Determine a loading range according to the current position of the AGV, and load preliminary constraint data corresponding to the loading range; Perform back-end graph optimization processing on the preliminary constraint data; and According to the back-end graph optimization result, construct and / or update the global map of the AGV in real time.

[0057] As another aspect, the present application also provides a computer program product, which may be included in the electronic device described in the above embodiments; or may exist separately without being assembled into the electronic device. The above computer program product stores one or more programs, and when the foregoing programs are executed by one or more processors to perform the map construction method described in the present application: Determine a loading range according to the current position of the AGV, and load preliminary constraint data corresponding to the loading range; Perform back-end graph optimization processing on the preliminary constraint data; and According to the back-end graph optimization result, construct and / or update the global map of the AGV in real time.

[0058] The above are only the preferred embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structural transformation made by using the specification and drawings of the present application under the application concept of the present application, or direct / indirect application in other related technical fields, is included in the patent protection scope of the present application.

Claims

1. A method for map construction, characterized in that, It includes the following steps: Determine the loading range according to the current position of the AGV, and load the preliminary constraint data corresponding to the loading range; Perform back-end graph optimization processing on the preliminary constraint data; and According to the back-end graph optimization result, construct and / or update the global map of the AGV in real time.

2. The map construction method according to claim 1, characterized in that, The performing back-end graph optimization processing on the preliminary constraint data includes: Based on the pose nodes and constraint edges of the AGV, optimize the pose estimation and map model of the AGV by minimizing the error function.

3. The map construction method according to claim 2, wherein The error function to be minimized is based on the reprojection error and pose error of the AGV.

4. The map construction method according to claim 1, characterized in that, The determining the loading range according to the current position of the AGV and loading the preliminary constraint data corresponding to the loading range includes: Set the area within a preset distance centered on the current position of the AGV as the loading range; and According to the current position and loading range of the AGV, load the preliminary constraint data corresponding to the loading range, and the preliminary constraint data includes odometry constraint data and loop detection constraint data.

5. The map construction method according to claim 1, characterized in that Before the determining the loading range according to the current position of the AGV and loading the preliminary constraint data corresponding to the loading range, it further includes: Determine the mapping area and plan the movement path of the AGV; and According to the collected laser point cloud data, perform preprocessing and generate the preliminary constraint data, and add a position identifier to the preliminary constraint data.

6. The map construction method according to claim 5, wherein The position identifier is the position information in the global coordinates of the AGV.

7. The map construction method according to claim 6, characterized in that, The constructing and / or updating the global map of the AGV in real time according to the back-end graph optimization result includes: Combine the laser point cloud data and the back-end graph optimization result to construct the global map; and According to the changes in the external environment, update the global map in real time.

8. A map construction device, characterized in that, It includes: A constraint data loading module, configured to determine the loading range according to the current position of the AGV and load the preliminary constraint data corresponding to the loading range; A back-end graph optimization module, configured to perform back-end graph optimization processing on the preliminary constraint data; and A map construction module, configured to construct and / or update the global map in real time according to the back-end graph optimization result.

9. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the map construction method according to any one of claims 1-7.

10. A computer-readable storage medium, on which a computer program is stored, and the computer program is used to implement the map construction method according to any one of claims 1-7.

Images