Map update method and apparatus, and electronic device

Abstract

The present application discloses a map update method and apparatus, and an electronic device. The method comprises: acquiring a reference ground texture map; if it is determined that a target area in the reference ground texture map needs to be updated, controlling a first vehicle in an automated guided vehicle group to move to a first position in the target area, and controlling a second vehicle in the automated guided vehicle group to move to a second position in the target area, the first position and the second position being located at two ends of the target area; acquiring environmental image data collected by the first vehicle when the second vehicle serves as a temporary reference object; and updating map data of the target area in the reference ground texture map on the basis of the environmental image data, so as to obtain a target ground texture map. By adopting the method, when a local portion of a ground texture map becomes invalid, the ground texture map can be automatically updated, thereby avoiding manual intervention and reducing maintenance costs.

Description

Map update methods, devices, and electronic equipment

[0001] Cross-references

[0002] This application claims priority to Chinese Patent Application No. 2025101160913, filed with the Chinese Patent Office on January 24, 2025, entitled “Map Updating Method, Apparatus and Electronic Device”, the entire contents of which are incorporated herein by reference. Technical Field

[0003] This application relates to the field of computer technology, and more specifically, to a map updating method, apparatus, and electronic device. Background Technology

[0004] AGVs (Automated Guided Vehicles) are widely used in logistics and production line handling. A typical AGV handling system includes a scheduling system and a swarm of robots. The scheduling system distributes tasks to each AGV, which then uses specific positioning and navigation methods to move goods. The AGVs use their positioning and navigation modules to provide pose information during task execution. Specific positioning and navigation methods include LiDAR SLAM (simultaneous localization and mapping based on laser radar) or ground texture positioning.

[0005] In related technologies, if the automated guided vehicles (AGVs) in an AGV handling system operate based on texture-based positioning and navigation, there is a risk of failure due to ground contamination. Because the ground texture map and the ground trajectory are dependent on each other, the ground texture map is more prone to failure when localized ground texture is altered by contamination. However, related technologies require manual intervention to address this risk, resulting in high maintenance costs. Summary of the Invention

[0006] In view of the above problems, this application proposes a map updating method, apparatus and electronic device that can automatically update ground texture maps and solve the problem of high manual maintenance costs in related technologies.

[0007] In a first aspect, embodiments of this application provide a map updating method, the method comprising: acquiring a reference ground texture map; if it is determined that a target area in the reference ground texture map needs to be updated, controlling a first vehicle in an automated guided vehicle (AGV) group to move to a first position in the target area and controlling a second vehicle in the AGV group to move to a second position in the target area, wherein the first position and the second position are located at opposite ends of the target area, the first vehicle is an AGV with ground texture map acquisition function and environmental image data acquisition function, and the second vehicle serves as a temporary reference when the first vehicle acquires image data; acquiring environmental image data acquired by the first vehicle when the second vehicle serves as a temporary reference; updating the map data of the target area in the reference ground texture map based on the environmental image data to obtain a target ground texture map.

[0008] Secondly, embodiments of this application provide a map updating device, the device comprising: a reference map acquisition module for acquiring a reference ground texture map; a movement control module for controlling a first vehicle in an automated guided vehicle (AGV) group to move to a first position in the target area and controlling a second vehicle in the AGV group to move to a second position in the target area when it is determined that a target area in the reference ground texture map needs to be updated, wherein the first position and the second position are located at opposite ends of the target area, the first vehicle is an AGV with ground texture map acquisition function and environmental image data acquisition function, and the second vehicle serves as a temporary reference when the first vehicle acquires image data; an image data acquisition module for acquiring environmental image data acquired by the first vehicle when the second vehicle serves as a temporary reference; and a map updating module for updating the map data of the target area in the reference ground texture map based on the environmental image data to obtain a target ground texture map.

[0009] Thirdly, embodiments of this application provide an electronic device including one or more processors; a memory; and one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, and the one or more application programs are configured to perform the methods described above.

[0010] Fourthly, embodiments of this application provide a computer-readable storage medium storing program code, which can be invoked by a processor to execute the above-described method.

[0011] The map updating method, apparatus, and electronic device provided in this application acquire a reference ground texture map; and when a target area in the reference ground texture map needs updating, control a first vehicle in an automated guided vehicle (AGV) convoy to move to a first position in the target area and control a second vehicle in the AGV convoy to move to a second position in the target area. The first and second positions are located at opposite ends of the target area. The first vehicle is an AGV with ground texture map acquisition and environmental image data acquisition capabilities, and the second vehicle serves as a temporary reference point for the first vehicle when acquiring image data. Environmental image data acquired by the first vehicle while the second vehicle serves as a temporary reference point is acquired. Based on the environmental image data, the map data of the target area in the reference ground texture map is updated to obtain the target ground texture map. By employing this method, automatic updating of the ground texture map can be achieved when there is partial failure, avoiding manual intervention and reducing maintenance costs. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 shows a schematic flowchart of a map update method provided in an embodiment of this application;

[0014] Figure 2 shows a flowchart of step S110 in Figure 1;

[0015] Figure 3 shows a schematic diagram of a map provided in one embodiment of this application;

[0016] Figure 4 shows a schematic diagram illustrating the positional relationship between the first vehicle and the second vehicle and the target area according to an embodiment of this application;

[0017] Figure 5 shows another schematic flowchart of a map updating method provided in an embodiment of this application;

[0018] Figure 6 shows a schematic diagram of a map update method provided in an embodiment of this application;

[0019] Figure 7 shows a schematic diagram of an automated guided vehicle group provided in one embodiment of this application;

[0020] Figure 8 shows a block diagram of a map updating device provided in an embodiment of this application;

[0021] Figure 9 shows a block diagram of an electronic device for performing a map update method according to an embodiment of this application;

[0022] Figure 10 illustrates a storage unit according to an embodiment of the present application for storing or carrying program code implementing the map update method according to an embodiment of the present application. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0024] Please refer to Figure 1, which is a schematic flowchart of a map update method provided in one embodiment of this application. This method can be applied to an electronic device, which may be an automated guided vehicle (AGV) in an AGV group (e.g., the first vehicle), a terminal device associated with each AGV in the AGV group, or a server, etc. This method can also be jointly executed by each AGV in the AGV group, which is not specifically limited here.

[0025] In some embodiments, the server can be a standalone physical server, a server cluster or distributed system consisting of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0026] Terminal devices can be smartphones, tablets, laptops, desktop computers, intelligent voice interaction devices, in-vehicle terminals, etc., but are not limited to these.

[0027] The map update method may specifically include the following steps S110 to S140.

[0028] Step S110: Obtain a reference ground texture map.

[0029] The reference ground texture map refers to a pre-created navigation map that can be used as a navigation map for each automated guided vehicle in a group of automated guided vehicles.

[0030] The reference ground texture map can be obtained by receiving a reference ground texture map uploaded by any automated guided vehicle in the automated guided vehicle group, or by obtaining the reference ground texture automated guided transport map from the memory of the electronic device or an associated storage device.

[0031] Among them, automated guided vehicles can refer to any vehicle that can locate the ground texture by referring to the ground texture map. Specifically, they can be applied to scenarios such as logistics handling and production line handling to locate the ground texture by referring to the ground texture map, thereby realizing logistics handling, production line handling, etc.

[0032] The automated guided vehicle (AGV) group includes multiple AGVs, such as at least one AGV with ground texture image acquisition and environmental image data acquisition functions, and multiple AGVs with only ground texture image acquisition functions and AGVs.

[0033] In one possible implementation, referring to Figure 2, the reference ground texture map can be created based on the following:

[0034] Step S111: Acquire multiple initial ground texture images and multiple sets of initial environment image data collected during the driving process of the first vehicle.

[0035] The first vehicle refers to an automated guided vehicle with ground texture image acquisition function and environmental image data acquisition function. For example, the first vehicle can be an automated guided vehicle including a downward-looking imaging device (such as a camera) and a LiDAR (Light Detection and Ranging) device, or it can be an automated guided vehicle including a downward-looking imaging device and an environmental imaging device.

[0036] It is worth noting that LiDAR devices construct a 3D point cloud map of the surrounding environment by emitting lasers and measuring the reflection time, while environmental imaging devices and downward-looking imaging devices typically refer to color or monochrome cameras used to capture visual images. When the first vehicle is an automated guided vehicle (AGV) including both downward-looking imaging devices and LiDAR devices, the initial environmental image data is specifically point cloud data; when the first vehicle is an AGV including both downward-looking imaging devices and environmental imaging devices, the initial environmental image data is an RGB image.

[0037] Step S112: Construct an initial ground texture map based on multiple initial ground texture images.

[0038] In one possible implementation, multiple initial ground texture images can be processed using image processing and computer vision techniques (such as image stitching, feature matching, etc.) to integrate the processed images into a continuous ground texture map.

[0039] Specifically, each initial ground texture image can be preprocessed (e.g., denoising, correction, and enhancement, among other preprocessing methods). Corner detection is performed on the preprocessed initial ground texture images to find stable feature points in the images, and a descriptor is calculated for each feature point. Then, based on the descriptors of each feature point, matching feature points are found between each pair of adjacent images. Geometric transformation parameters (rotation angle, scaling ratio, translation distance, etc.) between each pair of adjacent images are obtained based on the matching feature points. Multiple preprocessed initial ground texture images are then transformed to the same coordinate system and aligned and fused according to the geometric transformation parameters between each pair of adjacent images to obtain the final initial ground texture image.

[0040] Step S113: Construct an initial environment map based on multiple sets of initial environment image data.

[0041] In one possible implementation, if the initial environment image data is an RGB image, preprocessing is performed on each set of initial environment image data (e.g., using one or more preprocessing methods such as denoising, color correction, and image correction) to obtain a preprocessed image. Subsequently, object detection can be performed on the preprocessed image to obtain objects in the preprocessed image, or semantic segmentation can be performed on the image to obtain the category corresponding to each pixel in the preprocessed image. Then, visual techniques can be used to estimate the depth information of each pixel to construct a 3D environment model. The 3D environment models corresponding to multiple sets of initial environment data are aligned and fused to obtain an initial environment map.

[0042] In another possible implementation, if the initial environmental image data is point cloud data, preprocessing is performed on each set of initial environmental image data (e.g., denoising, downsampling, and ground segmentation) to obtain processed point cloud data. Then, local and global features are extracted from the processed point cloud data. Based on the local and global features corresponding to each set of preprocessed point cloud data, the multiple sets of preprocessed point cloud data are aligned to ensure that the aligned sets of point cloud data are in the same coordinate system. The aligned sets of point cloud data are then fused to obtain a 3D environmental model. Machine learning or deep learning methods are used to classify the point cloud data in the 3D environmental model, and the semantics of objects in the 3D environmental model (e.g., roads, buildings, vegetation) are labeled according to the classification results, thereby obtaining a 3D environmental map.

[0043] Step S114: Fuse the initial environment map with the initial ground texture map to obtain a reference ground texture map.

[0044] In one possible implementation, the initial environment map and the initial ground texture map can be aligned in the same coordinate system to merge the aligned initial environment map with the initial ground texture map.

[0045] Specifically, the initial environment map and the initial ground texture map are transformed into the same global coordinate system. Then, a registration algorithm is used to align the feature points in the initial environment map and the initial ground texture map in the global coordinate system. For point cloud data, the registration algorithm used can be the iterative nearest point algorithm; for image data, the registration algorithm used can be feature matching and homography matrix. Afterwards, a fusion algorithm, such as Bayesian fusion or Kalman filtering, can be used to fuse the registered initial environment map and the registered initial ground texture map to obtain a reference ground texture map.

[0046] By employing steps S111-S114 above, the initial environmental map and the initial ground texture map are aligned and fused in the same coordinate system to achieve the fusion of image data acquired by the image acquisition device, forming a unified map model. This not only improves map accuracy but also enhances the system's flexibility and adaptability. Furthermore, if the initial environmental map is generated based on point cloud data acquired by a LiDAR device or RGB images acquired by an environmental imaging device, it can effectively overcome the influence of external factors such as lighting changes and weather conditions, improving the robustness of map construction. Even under extreme conditions, the accuracy and integrity of the map can be guaranteed.

[0047] Figure 3 shows a schematic diagram of the driving direction on a reference ground texture map and a schematic diagram of the initial environment map, as well as a texture image of the reference ground texture map at a specified location. It can be seen that the spatial coverage of the initial environment map and the reference ground texture map is consistent. The initial environment map typically contains more macroscopic information, such as building outlines and road directions, etc., representing large-scale structures. The reference ground texture map, on the other hand, provides more detailed surface texture information, reflecting minute details such as ground materials, colors, and patterns.

[0048] In another possible implementation, the reference ground texture map can be created based on the following:

[0049] Multiple initial ground texture images are acquired during the driving of the first vehicle. Each initial texture image is acquired under conditions where there are no interfering objects (such as stains, shadows, puddles, obstacles, or any other objects that affect the texture map effect) on the corresponding ground. A reference ground texture map is constructed based on the multiple initial ground texture images.

[0050] For a detailed description of the above implementation method, please refer to the previous detailed description of steps S111 and S112, which will not be repeated here.

[0051] By employing the above implementation method, initial ground texture images are acquired without interference (such as stains, shadows, puddles, and obstacles), ensuring higher quality and clearer texture details. Subsequently, feature extraction and matching are easier performed on the interference-free ground texture images, allowing for more accurate calculation of geometric transformation parameters between the initial texture images, thereby improving the accuracy of the final reference ground texture map.

[0052] Step S120: If it is determined that the target area in the reference ground texture map needs to be updated, control the first vehicle in the automated guided vehicle group to move to the first position in the target area and control the second vehicle in the automated guided vehicle group to move to the second position in the target area. The first position and the second position are located at opposite ends of the target area. The first vehicle is an automated guided vehicle with ground texture map acquisition function and environmental image data acquisition function. The second vehicle serves as a temporary reference when the first vehicle acquires image data.

[0053] This can be achieved by determining that the target area in the reference ground texture map needs updating when it is determined to be invalid (e.g., during feature point matching between ground texture maps collected by multiple automated guided vehicles (AGVs) during their operation and a reference ground texture map). Alternatively, it can be determined when the time elapsed since the last update reaches a preset threshold; or it can be determined when an update command for the target area in the reference ground texture map is received from a user via a client.

[0054] The reasons for the failure of the target area may include physical changes to the ground (such as the appearance of new cracks, potholes, repair marks, etc.), the appearance of temporary obstacles, stains, cleaning and maintenance, or changes due to natural factors (such as rain or snow making the ground slippery, or prolonged exposure to sunlight causing the ground color to change).

[0055] It is worth noting that the position of the second vehicle, which serves as a reference point for the first vehicle, remains unchanged during the image data acquisition process. Furthermore, the first vehicle can move towards the second vehicle, which acts as a temporary reference point, and acquire environmental image data during this movement. It should be understood that the first vehicle can also acquire ground texture images during its movement.

[0056] Considering that the target area may cover a wide road surface, there can be one or more second vehicles, such as at least two. In this case, at least two second vehicles are positioned at the edge of the target area, with the edge direction perpendicular to the navigation direction of the reference ground texture map. As shown in Figure 4, Figure 4(a) shows the case where there is one second vehicle, and Figure 4(b) shows the case where there are two second vehicles.

[0057] It should be understood that the target area mentioned above can be one or multiple. When there are multiple target areas, the above S120 can be executed for each target area.

[0058] Step S130: Acquire environmental image data collected by the first vehicle when the second vehicle serves as a temporary reference point.

[0059] The electronic device can receive environmental image data collected and uploaded by the first vehicle in real time, or it can receive data packets containing the collected environmental image data after the first vehicle has completed the collection of environmental image data and upload them. The settings can be configured according to actual needs.

[0060] Step S140: Update the map data of the target area in the reference ground texture map based on the environmental image data to obtain the target ground texture map.

[0061] The environmental image data can be one set or multiple sets. The environmental image data mentioned above can be environmental point cloud data or environmental visual image data (e.g., RGB image).

[0062] In one feasible implementation, if the environmental image data is a set, and the environmental image data is specifically environmental point cloud data, the above step S140 may be: according to the positions of the first vehicle and the second vehicle in the environmental point cloud data, extract the point cloud data of the target area from the environmental point cloud data; extract the ground texture information of the target area from the point cloud data of the target area; map the ground texture information of the target area to the target area in the reference ground texture map to obtain the target ground texture map.

[0063] In some scenarios, before extracting the target area's point cloud data from the environmental point cloud data based on the positions of the first and second vehicles, a registration algorithm can be used to align the environmental image data (environmental point cloud data) with a reference ground texture map. Specifically, this process involves extracting the target area's point cloud data from the aligned environmental image data (environmental point cloud data). Then, semantic classification is performed on the target area's point cloud data. Based on the semantic classification results, ground points are separated from the target area's point cloud data, and texture information is extracted from these ground points. This texture information is then mapped onto the target area in the reference ground texture map to replace the texture information in the target area of ​​the reference ground texture map, thus obtaining the target ground texture map.

[0064] In another possible implementation, if the environmental image data consists of multiple sets, and the environmental image data is specifically environmental point cloud data, the above step S140 may be: aligning the multiple sets of environmental image data to ensure that the multiple sets of environmental image data are located in the same coordinate system; then, fusing the multiple sets of aligned environmental image data (environmental point cloud data) to generate a complete three-dimensional environment model; then, performing semantic classification on the three-dimensional environment model; separating ground points within the target area from the three-dimensional environment model based on the semantic classification result, the first position, and the second position; and extracting texture information from the ground points, and mapping the texture information to the target area in the reference ground texture map to replace the texture information in the target area of ​​the reference ground texture map, thereby obtaining the target ground texture map.

[0065] In another possible implementation, if the environmental image data is a set, and the environmental image data is specifically environmental visual image data (RGB image), the above step S140 may be: aligning the RGB image with the reference ground texture map according to the position information of the first vehicle, preprocessing the aligned RGB image (e.g., using one or more preprocessing methods such as denoising, correction, and enhancement) to obtain a preprocessed RGB image; determining the image region in the RGB image corresponding to the target region according to the positions of the first vehicle and the second vehicle, extracting texture information from the image region, and mapping the texture information to the target region of the reference ground texture map to replace the texture information in the target region of the reference ground texture map to obtain the target ground texture map.

[0066] In another possible implementation, if the environmental image data consists of multiple sets, specifically environmental visual image data (RGB images), then step S140 may be as follows: Align the multiple sets of RGB images to ensure they are in the same coordinate system. Preprocess each aligned RGB image (e.g., using one or more preprocessing methods such as denoising, correction, and enhancement) to obtain a preprocessed RGB image. Then, stitch the preprocessed RGB images together to obtain scene map data. Determine the image region in the scene map data corresponding to the target region based on the first and second positions. Extract texture information from the image region and map the texture information to the target region of the reference ground texture map to replace the texture information in the reference ground texture map, thus obtaining the target ground texture map.

[0067] In some scenarios, if the first vehicle collects both environmental image data (i.e., the aforementioned environmental point cloud data or environmental visual image data) and ground texture images simultaneously, then step S140 can be: updating the map data of the target area in the reference ground texture map based on the environmental image data and the ground texture image to obtain the target ground texture map. In this case, a method similar to steps S111-S113 in the aforementioned embodiments can be used to align the environmental image data and the ground texture image, extract the first ground texture information corresponding to the target area from the aligned environmental image data, and extract the second ground texture information corresponding to the target area from the aligned ground texture image; then, the first ground texture information and the second ground texture information are respectively mapped to the target area of ​​the reference ground texture map, and a fusion algorithm (e.g., Bayesian fusion or Kalman filtering) is used to fuse the first ground texture information and the second ground texture information mapped to the target area and replace the texture information in the target area of ​​the reference ground texture map to obtain the target ground texture map.

[0068] Through the above implementation methods, it is possible to combine environmental image data and ground texture images to provide more comprehensive and accurate texture information for target areas in reference ground texture maps.

[0069] By employing the map update method of this application, when a local area in the reference ground texture map used for navigation in an automated guided vehicle (AGV) fleet becomes unavailable, only one AGV (first vehicle) with ground texture map acquisition and environmental image data acquisition capabilities and a second vehicle serving as a temporary reference are needed. The map data for the target area in the reference ground texture map can be updated based on the environmental image data acquired by the first vehicle while the second vehicle serves as a temporary reference. This not only reduces the required manpower and material resources but also accelerates the map update speed. Compared to related technologies that involve manual on-site surveying or complex operations using multiple devices, the map update speed of this application is more efficient. Furthermore, during the map update process, there is no need for a large-scale pause or replanning of the entire AGV fleet's workflow. Only a small number of AGVs (i.e., the first and second vehicles) need to have their tasks adjusted to complete the local area map update. Therefore, the reference ground texture map can be kept up-to-date while minimizing impact on the overall operational efficiency of the AGV fleet.

[0070] To enable each automated guided vehicle (AGV) in the AGV group to navigate and drive in a timely manner based on the updated ground texture map, please refer to Figure 5. In one possible implementation, after performing step S140, the method further includes: sending the target ground texture map to each AGV in the AGV group, so that each AGV replaces the target ground texture map with the new reference ground texture map.

[0071] In one possible implementation, the method further includes steps S160-S180 before performing step S120.

[0072] Step S160: Obtain texture maps in real time from multiple automated guided vehicles (AGVs) in the AGV group as they drive according to the reference ground texture map.

[0073] Specifically, the texture maps collected in real time by multiple automated guided vehicles (AGVs) while driving according to the reference ground texture map are ground texture maps. The method for obtaining the texture maps collected by each AGV can be either receiving texture maps periodically uploaded by each AGV or collecting and uploading texture maps in real time.

[0074] Step S170: Compare the texture maps collected in real time by each of the automated guided vehicles with the reference ground texture map to obtain the comparison results of the texture maps collected in real time by each of the automated guided vehicles and the reference ground texture map. The comparison results are used to indicate whether each texture map collected in real time by the automated guided vehicle matches the reference ground texture map.

[0075] In one possible implementation, each automated guided vehicle (AGV) collects texture maps in real time while driving according to the reference ground texture map, and each texture map has a corresponding collection location. Step S170 includes: for each texture map, aligning the texture map with the reference ground texture map according to the collection location corresponding to the texture map; determining the map region corresponding to the aligned texture map from the reference ground texture map; comparing the map region with the texture map to obtain a comparison result of each texture map of each navigation transport vehicle and its corresponding map region in the reference ground texture map.

[0076] Specifically, the process of comparing the map region with the texture map can be as follows: for each texture map, feature points are extracted from the texture map and the map region corresponding to the texture map, and the nearest neighbor algorithm is used to match the feature points of the texture map with the feature points of the map region corresponding to the texture map to obtain multiple sets of matching point pairs. Each set of matching point pairs includes one feature point of the texture map and one feature point of the map region corresponding to the texture map. The similarity between the texture map and the map region corresponding to the texture map is obtained according to the proportion of the matching point pairs to the total feature points or the average distance between multiple matching point pairs. If the similarity is lower than a preset threshold, it is determined that the map region has changed.

[0077] The process of comparing the map region with the texture map described above can also be as follows: For each texture map, compare the texture map with the corresponding map region pixel by pixel to obtain a pixel difference matrix. Calculate the proportion of pixels in the difference matrix whose pixel differences are greater than a preset difference threshold. If the pixel proportion is greater than the preset proportion threshold, it is determined that the map region has changed.

[0078] Step S180: Based on the comparison results between the texture map collected in real time by each of the automated guided vehicles and the reference ground texture map, determine whether there is a target area in the reference ground texture map that needs to be updated.

[0079] In one possible implementation, if a texture map that fails to match is determined based on the comparison results between the texture map collected in real time by each of the automated guided vehicles and the reference ground texture map, then the area in the reference ground texture map corresponding to that texture map is taken as the target area.

[0080] In another possible implementation, a heat map method can be used to identify areas that are all deemed invalid by comparison results of multiple automated guided vehicles within a certain time period, and these areas will be designated as target areas.

[0081] Specifically, the reference ground texture map includes multiple pre-divided grid regions, and the comparison result includes the comparison result of each texture map with its corresponding map region in the reference ground texture map. The process of determining the target region using the heatmap method specifically includes the following steps: based on the comparison results of the texture maps collected in real time by multiple automated guided vehicles with their corresponding map regions in the reference ground texture map, and the grid regions covered by each map region, the number of comparison failures corresponding to each grid region is counted; if there is a target grid region with more than a preset number of failures among the corresponding failures in multiple grid regions, it is determined that there is a target region in the reference ground texture map that needs to be updated, and the target region is composed of the target grid region; if there is no target grid region with more than a preset number of failures among the corresponding failures in multiple map regions, it is determined that there is no target region in the reference ground texture map that needs to be updated.

[0082] For example, if N texture maps (e.g., 5 texture maps) fail to match their corresponding map regions in a reference ground texture map in the real-time comparison results of the texture maps collected by multiple automated guided vehicles, the grid regions covered by the map region corresponding to texture map 1 are K1, K2, and K3; the grid regions covered by the map region corresponding to texture map 2 are K2, K3, and K4; the grid regions covered by the map region corresponding to texture map 3 are K1, K2, and K3; the grid regions covered by the map region corresponding to texture map 4 are K3, K4, and K5; and the grid regions covered by the map region corresponding to texture map 5 are... Based on the above statistics, the number of alignment failures for grid regions K1 is 2, K2 is 3, K3 is 4, K5 is 2, K6 is 1, and K7 is 1. If the preset number of failures is 1, then the target grid regions can be determined as K1, K2, K3, K4, and K5, that is, the target region is the area formed by K1, K2, K3, K4, and K5.

[0083] It is worth mentioning that the above steps S160-S180 can be executed by electronic devices, or by electronic devices and each automated guided vehicle in the automated guided vehicle group in coordination.

[0084] For example, as shown in Figure 6, if steps S160-S180 are executed collaboratively by the electronic device and each Automated Guided Vehicle (AGV) in the AGV group, each AGV can perform steps S160-S170 to determine the validity of the reference ground texture map. If it is determined that any of its collected texture maps do not match the reference ground texture map (i.e., the reference ground texture map is invalid), it sends the texture map, the corresponding map area (invalid area), and the matching result to the electronic device, so that the electronic device can execute the aforementioned step S180. When it is determined that the target area needs to be updated, steps S120-S140 are executed to invoke the first and second vehicles. The first vehicle, which has a fusion positioning function (with ground texture map acquisition and environmental image data acquisition functions), collects environmental image data when the second vehicle serves as a temporary reference. The second vehicle uploads the collected environmental image data to the electronic device, so that the electronic device updates the map data of the target area in the reference ground texture map based on the environmental image data to obtain the target ground texture map and distributes it to each AGV group.

[0085] By employing steps S160-S180 above, real-time texture maps can be collected and compared with a reference ground texture map during the operation of the automated guided vehicle (AGV) fleet. This allows for the timely detection of changes in ground texture, quickly identifying which areas require updating, and thus helping to maintain the accuracy and timeliness of the ground texture map. Furthermore, by comparing real-time texture maps from multiple AGVs with the reference texture map, and uploading the map area corresponding to the real-time texture map in the reference texture map when a comparison fails, heatmap methods can be used to identify the target area where the map is invalid. This reduces misjudgments caused by errors or anomalies in data collected by a single AGV, thereby improving the accuracy of ground texture map updates.

[0086] For example, as shown in Figure 7, the description focuses on an automated guided vehicle (AGV) fleet comprising at least one AGV (first vehicle) with fusion positioning and multiple AGVs (second vehicles) with only texture positioning. Texture positioning refers to having downward-looking texture positioning capability, i.e., the AGV has a downward-looking imaging device; fusion positioning refers to simultaneously possessing LiDAR positioning and texture positioning capabilities, and also having the ability to fuse texture and LiDAR data for mapping and update texture maps based on LiDAR positioning information, i.e., the AGV is equipped with both LiDAR equipment and downward-looking imaging equipment.

[0087] Before the automated guided vehicle fleet begins transportation, a map can be created using the first vehicle. That is, the first vehicle is controlled to collect multiple initial ground texture images and multiple sets of initial environment image data during operation. Then, the aforementioned steps S111-S114 are executed to complete the creation of the initial ground texture map and the initial environment map (LiDAR map), and the initial ground texture map and the initial environment map are aligned and fused to obtain a reference ground texture map.

[0088] Subsequently, a reference ground texture map can be distributed to each Automated Guided Vehicle (AVR) in the AVR group, enabling each AVR to operate according to the reference ground texture map. During operation, each AVR can execute the aforementioned steps S160-S180 to determine if there is a target area failure that requires updating. If a target area failure is determined, a local update of the reference ground texture map is triggered. At this time, an AVR capable of fusion positioning (the first vehicle) is used in conjunction with any other AVR to update the map of the failed area. Specifically, a first vehicle and a second vehicle are respectively set at both ends of the failed area. The second vehicle can be equipped with a reflector as a temporary reference marker and execute the aforementioned steps S120-S140 to update the map of the target area. Based on the above, it can be seen that the solution provided by this application embodiment has the characteristics of automatic detection and automatic updating of the reference ground texture map, enabling this solution to automatically and quickly respond to the problem of texture positioning and navigation system failure caused by ground pollution, avoiding the need for manual intervention in the long-term use of texture positioning systems in related technologies, and greatly reducing maintenance costs.

[0089] Please refer to Figure 8, which shows a block diagram of a map updating device provided in an embodiment of this application. The process shown in Figure 8 will be described in detail below. The map updating device 200 is applied to the aforementioned electronic device. The map updating device 200 includes: a reference map acquisition module 210, a motion control module 220, an image data acquisition module 230, and a map updating module 240, wherein:

[0090] The reference map acquisition module 210 is used to acquire a reference ground texture map; the movement control module 220 is used to control a first vehicle in the automated guided vehicle group to move to a first position in the target area and a second vehicle in the automated guided vehicle group to move to a second position in the target area when it is determined that the target area in the reference ground texture map needs to be updated, wherein the first position and the second position are located at opposite ends of the target area, the first vehicle is an automated guided vehicle with ground texture map acquisition function and environmental image data acquisition function, and the second vehicle serves as a temporary reference when the first vehicle acquires image data; the image data acquisition module 230 is used to acquire environmental image data acquired by the first vehicle when the second vehicle serves as a temporary reference; the map update module 240 is used to update the map data of the target area in the reference ground texture map based on the environmental image data to obtain a target ground texture map.

[0091] Furthermore, the map updating device 200 also includes a texture map acquisition module, a comparison module, and an update determination module. The texture map acquisition module is used to acquire texture maps collected in real time by multiple automated guided vehicles (AGVs) in the AGV group while driving according to the reference ground texture map. The comparison module is used to compare the texture maps collected in real time by each AGV with the reference ground texture map to obtain a comparison result between the texture maps collected in real time by each AGV and the reference ground texture map. The comparison result is used to indicate whether each texture map collected in real time by the AGV matches the reference ground texture map. The update determination module is used to determine whether there is a target area in the reference ground texture map that needs to be updated based on the comparison result between the texture maps collected in real time by each AGV and the reference ground texture map.

[0092] Furthermore, the reference ground texture map includes multiple map regions, and the comparison result includes the comparison result of each texture map with its corresponding map region in the reference ground texture map. The update determination module is also used to calculate the number of comparison failures for each grid region based on the comparison results of the texture maps collected in real time by the multiple automated guided vehicles with their corresponding map regions in the reference ground texture map, and the grid regions covered by each map region. If there is a target grid region with more than a preset number of comparison failures among the corresponding numbers in the multiple grid regions, it is determined that there is a target region in the reference ground texture map that needs to be updated, and the target region is composed of the target grid region. If there is no target grid region with more than a preset number of comparison failures among the corresponding numbers in the multiple map regions, it is determined that there is no target region in the reference ground texture map that needs to be updated.

[0093] In one possible implementation, each automated guided vehicle (AGV) acquires texture maps in real time while traveling according to the reference ground texture map, each map having a corresponding acquisition location. The comparison module includes an alignment submodule, a region determination submodule, and a comparison submodule. The alignment submodule aligns each texture map with the reference ground texture map based on its acquisition location. The region determination submodule determines the map region corresponding to the aligned texture map from the reference ground texture map. The comparison submodule compares the map region with the texture map to obtain a comparison result of each texture map of each navigation vehicle and its corresponding map region in the reference ground texture map.

[0094] In one possible implementation, the reference map acquisition module includes a data acquisition submodule, a texture map creation submodule, an environment map creation submodule, and a reference map acquisition submodule. The data acquisition submodule is used to acquire multiple initial ground texture images and multiple sets of initial environment image data collected during the first vehicle's movement. The texture map creation submodule is used to construct an initial ground texture map based on the multiple initial ground texture images. The environment map creation submodule is used to construct an initial environment map based on multiple sets of initial environment image data. The reference map acquisition submodule is used to fuse the initial environment map with the initial ground texture map to obtain a reference ground texture map.

[0095] In one possible implementation, the environmental image data collected by the first vehicle includes environmental point cloud data or environmental visual image data.

[0096] In one possible implementation, the map updating device 200 further includes a map sending module for sending the target ground texture map to each of the automated guided vehicles in the group of automated guided vehicles, so that each of the automated guided vehicles replaces the target ground texture map with a new reference ground texture map.

[0097] In one possible implementation, if the environmental image data is point cloud data, the map update module is further configured to: extract point cloud data of a target area from the point cloud data according to the positions of the first vehicle and the second vehicle in the point cloud data; extract ground texture information of the target area from the point cloud data of the target area; and map the ground texture information of the target area onto the target area in the reference ground texture map to obtain a target ground texture map.

[0098] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described device and module can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0099] In the several embodiments provided in this application, the coupling between modules can be electrical, mechanical, or other forms of coupling.

[0100] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0101] Please refer to Figure 9, which shows a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device 300 may include the following components: a memory 310, one or more processors 320, and one or more application programs, wherein the one or more application programs are stored in the memory 310 and are used to cause the electronic device 400 to execute a map update method applied to the electronic device when invoked by one or more processors 320.

[0102] The processor 320 may include one or more processing cores. The processor 320 connects to various parts within the electronic device 400 using various interfaces and lines, and performs various functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory, and by calling data stored in memory. Optionally, the processor may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. It is understood that the aforementioned modem may also not be integrated into the processor 320, but may be implemented separately through a communication chip.

[0103] The memory 310 may include random access memory (RAM) or read-only memory (ROM). The memory 310 may be used to store instructions, programs, code, code sets, or instruction sets. The memory 310 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for implementing at least one function, instructions for implementing the various method embodiments described above, etc. The data storage area may store data created during the use of the electronic device.

[0104] Please refer again to Figure 7. This application embodiment also provides a transport vehicle system, including electronic equipment and an automated guided vehicle group. The electronic equipment is communicatively connected to multiple automated guided vehicles in the automated guided vehicle group. The multiple automated guided vehicles include at least one automated guided transport vehicle with ground texture map acquisition function and environmental image data acquisition function, as well as multiple automated guided transport vehicles with only ground texture map acquisition function.

[0105] Please refer to Figure 10, which shows a structural block diagram of a computer-readable storage medium provided in an embodiment of this application. The computer-readable medium 400 stores program code that can be called by a processor to execute the methods described in the above method embodiments.

[0106] The computer-readable storage medium 400 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Optionally, the computer-readable storage medium 400 includes a non-transitory computer-readable storage medium. The computer-readable storage medium 400 has storage space for program code 410 that performs any of the method steps described above. This program code can be read from or written to one or more computer program products. The program code 410 may be compressed, for example, in a suitable form.

[0107] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A map update method characterized by comprising: The method includes: Obtain a reference ground texture map; If it is determined that the target area in the reference ground texture map needs to be updated, the first vehicle in the automated guided vehicle group is controlled to move to a first position in the target area and the second vehicle in the automated guided vehicle group is controlled to move to a second position in the target area. The first position and the second position are located at opposite ends of the target area. The first vehicle is an automated guided transport vehicle with ground texture map acquisition function and environmental image data acquisition function. The second vehicle serves as a temporary reference when the first vehicle acquires image data. Acquire environmental image data collected by the first vehicle when the second vehicle serves as a temporary reference point; The target ground texture map is obtained by updating the map data of the target area in the reference ground texture map based on the environmental image data.

2. The method of claim 1, wherein, Before controlling the first vehicle in the automated guided vehicle group to move to a first end position of the target area and controlling the second vehicle in the automated guided vehicle group to move to a second end position of the target area, the method further includes: Obtain texture maps in real time from multiple automated guided vehicles (AGVs) in the AGV group as they drive according to the reference ground texture map; The texture maps collected in real time by each of the automated guided vehicles are compared with the reference ground texture map to obtain the comparison results of the texture maps collected in real time by each of the automated guided vehicles and the reference ground texture map. The comparison results are used to indicate whether each texture map collected in real time by the automated guided vehicle matches the reference ground texture map. Based on the comparison results between the texture map collected in real time by each of the automated guided vehicles and the reference ground texture map, it is determined whether there is a target area in the reference ground texture map that needs to be updated.

3. The method of claim 2, wherein, The reference ground texture map includes multiple pre-divided grid regions, and the comparison result includes the comparison result of each texture map and its corresponding map region in the reference ground texture map; The determination of whether there is a target area in the reference ground texture map that needs to be updated, based on the comparison results between the texture map collected in real time by each of the automated guided vehicles and the reference ground texture map, includes: Based on the comparison results of the texture maps collected in real time by multiple automated guided vehicles and their corresponding map regions in the reference ground texture map, as well as the grid regions covered by each map region, the number of comparison failures corresponding to each grid region is counted. If there is a target grid region with a number of times greater than a preset number among the multiple grid regions, it is determined that there is a target region in the reference ground texture map that needs to be updated, and the target region is composed of the target grid regions; If there is no target grid region with a number greater than a preset number of occurrences in the corresponding number of occurrences in the multiple map regions, it is determined that there is no target region in the reference ground texture map that needs to be updated.

4. The method of claim 2, wherein, Each automated guided vehicle collects texture maps in real time while driving according to the reference ground texture map, and the collection location corresponds to the collection location. The step of comparing the texture maps collected in real time by each of the automated guided vehicles with the reference ground texture map to obtain the comparison results of the texture maps collected in real time by each of the automated guided vehicles and the reference ground texture map includes: For each texture map, the texture map is aligned with the reference ground texture map according to the acquisition location corresponding to the texture map; Determine the map region corresponding to the aligned texture map from the reference ground texture map; The map region is compared with the texture map to obtain the comparison result of each texture map of each navigation transport vehicle and its corresponding map region in the reference ground texture map.

5. The method of claim 1, wherein, Obtain a reference ground texture map, including Acquire multiple initial ground texture images and multiple sets of initial environment image data collected during the driving process of the first vehicle; Construct an initial ground texture map based on multiple initial ground texture images; Construct an initial environment map based on multiple sets of initial environment image data; The initial environment map and the initial ground texture map are fused to obtain a reference ground texture map.

6. The method according to any one of claims 1 to 5, characterized in that, The environmental image data collected by the first vehicle includes environmental point cloud data or environmental visual image data.

7. The method according to any one of claims 1 to 5, characterized in that, After updating the map data of the target area in the reference ground texture map based on the image data to obtain the target ground texture map, the method further includes: The target ground texture map is sent to each of the automated guided vehicles in the group, so that each of the automated guided vehicles replaces the target ground texture map with a new reference ground texture map.

8. The method of claims 1-5, wherein, If the environmental image data is point cloud data; The step of updating the map data of the target area in the reference ground texture map based on the environmental image data to obtain the target ground texture map includes: Based on the positions of the first vehicle and the second vehicle in the point cloud data, the point cloud data of the target area is extracted from the point cloud data; Extract ground texture information of the target area from the point cloud data of the target area; The ground texture information of the target area is mapped onto the target area in the reference ground texture map to obtain the target ground texture map.

9. A map update device characterized by comprising: The device includes: The reference map acquisition module is used to acquire a reference ground texture map; The movement control module is used to control a first vehicle in the automated guided vehicle group to move to a first position in the target area and a second vehicle in the automated guided vehicle group to move to a second position in the target area when it is determined that the target area in the reference ground texture map needs to be updated. The first position and the second position are located at opposite ends of the target area. The first vehicle is an automated guided transport vehicle with ground texture map acquisition function and environmental image data acquisition function. The second vehicle serves as a temporary reference when the first vehicle acquires image data. The image data acquisition module is used to acquire environmental image data collected by the first vehicle when the second vehicle is used as a temporary reference. The map update module is used to update the map data of the target area in the reference ground texture map based on the environmental image data, so as to obtain the target ground texture map.

10. An electronic device, comprising: include: One or more processors; Memory; One or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, the one or more applications being configured to perform the method as described in any one of claims 1-8.

11. A transport cart system characterized by, The device includes the electronic device and automated guided vehicle group as described in claim 10, wherein the electronic device is communicatively connected to multiple automated guided vehicles in the automated guided vehicle group, and the multiple automated guided vehicles include at least one automated guided transport vehicle with ground texture map acquisition function and environmental image data acquisition function, as well as multiple automated guided transport vehicles with only ground texture map acquisition function.

12. A computer readable storage medium, characterized in that, The computer-readable storage medium contains program code that can be invoked by a processor to execute the method as described in any one of claims 1-8.

Images