Automobile positioning method, device and equipment and storage medium

Abstract

The invention discloses an automobile positioning method, device and equipment and a storage medium, and relates to the technical field of vehicle positioning, the automobile positioning method comprises the steps that target data and a target image of an automobile end are acquired, and the target data comprise position feature data and environment feature data; establishing a target three-dimensional map based on the target data; and matching the target three-dimensional map with the target image, and determining the position of the target image in the target three-dimensional map. According to the invention, through integration of the vehicle end data, construction of the full-scene three-dimensional map, and based on matching of the image and the three-dimensional map, high-precision vehicle positioning in semi-shielded and underground environments is realized, a human-vehicle mutual searching function is supported, and the user experience is improved.

Description

Technical Field

[0001] The present application relates to the technical field of vehicle positioning, and in particular to a method, device, equipment and storage medium for positioning a vehicle. Background Art

[0002] With the acceleration of urbanization, people are increasingly relying on cars as a means of transportation in semi-occluded scenarios such as large commercial complexes and residential communities. However, in these scenarios, due to the obstruction of buildings, traditional GPS and RTK positioning methods often cannot provide accurate positioning services, which makes it difficult for drivers to find parking spaces and vehicles. In addition, the positioning problem in completely obstructed environments such as underground garages is particularly prominent, which not only affects the user experience, but also limits the development of autonomous driving technology.

[0003] At present, automobile positioning mainly relies on GPS and RTK technologies. These technologies can provide relatively accurate positioning services in open outdoor environments. In some specific application scenarios, IMU, vehicle-mounted sensors and vision systems are also combined to assist positioning. However, these methods face the problem of reduced positioning accuracy in environments such as urban canyons and indoor parking lots, because the obstruction and reflection of satellite signals in these environments will seriously affect the accuracy of positioning.

[0004] Although existing positioning technologies can provide effective positioning services in certain environments, their performance will be significantly reduced in semi-occluded or fully occluded environments. This makes it difficult for drivers to accurately find their vehicles in scenarios such as underground garages, and also limits the application of autonomous vehicles in these scenarios. In addition, the current positioning technology still has shortcomings in terms of positioning accuracy, robustness and real-time performance, especially in dynamically changing urban environments. Therefore, how to obtain high-precision positioning of vehicles in various environments has become an urgent problem to be solved. Summary of the invention

[0005] The purpose of this application is to provide a vehicle positioning method, device, equipment and storage medium, aiming to solve the technical problem of how to obtain high-precision positioning of a vehicle in various environments.

[0006] To achieve the above purpose, the present application proposes a vehicle positioning method, the method comprising:

[0007] Acquire target data and target image on the vehicle side, wherein the target data includes position feature data and environment feature data;

[0008] Establishing a target three-dimensional map based on the target data;

[0009] The target three-dimensional map is matched with the target image to determine the position of the target image in the target three-dimensional map.

[0010] In one embodiment, obtaining target data and target images on the vehicle side includes:

[0011] Sending a target data acquisition request so that the vehicle end collects position feature data and environmental feature data based on the acquisition request, wherein the position feature data includes vehicle displacement data, which is obtained based on vehicle chassis data and inertial measurement data; the environmental feature data is obtained when it is determined that the vehicle is in motion or when the vehicle displacement distance exceeds a preset distance threshold based on the vehicle chassis data and the inertial measurement data; the environmental feature data includes parking space feature information, traffic infrastructure information, and positioning information;

[0012] A request for obtaining a target image is sent, where the target image includes environmental information.

[0013] In one embodiment, establishing a target three-dimensional map based on the target data includes:

[0014] Performing position determination based on the target data to obtain environmental feature data of nearby locations;

[0015] Perform feature matching based on the environmental feature data to obtain a feature matching result;

[0016] If the feature matching result is that multiple positions coincide, a target three-dimensional map is obtained.

[0017] In one embodiment, if the feature matching result is that multiple positions coincide, after obtaining the target three-dimensional map, the method further includes:

[0018] Performing a topological connection relationship check based on the target three-dimensional map to obtain a check result of the target three-dimensional map;

[0019] If the inspection result is complete, the map inspection is completed;

[0020] If the inspection result is incomplete, map calibration is performed so that the vehicle obtains corresponding reference map data according to the calibrated map and feeds it back to the cloud;

[0021] The target three-dimensional map is updated according to the reference map data.

[0022] In one embodiment, matching the target three-dimensional map with the target image to determine the position of the target image in the target three-dimensional map includes:

[0023] Extracting image feature information from the target image based on a deep learning model;

[0024] Performing visual matching on the image feature information and the target three-dimensional map to obtain a visual matching result;

[0025] Based on the visual matching result, the position of the target image in the target three-dimensional map is obtained.

[0026] In one embodiment, after determining the position of the target image in the target three-dimensional map, the method further includes:

[0027] Get the reference image of the mobile terminal;

[0028] Image semantic comparison and positioning are performed based on the reference image and the target three-dimensional map to obtain a positioning result.

[0029] In one embodiment, after determining the position of the target image in the target three-dimensional map, the method further includes:

[0030] Obtain positioning request and location information from the mobile terminal;

[0031] Matching the positioning information in the target data based on the position information to obtain a position matching result;

[0032] Based on the position matching result, a corresponding three-dimensional map is sent.

[0033] In addition, to achieve the above-mentioned purpose, the present application also proposes a vehicle positioning device, the device comprising:

[0034] An acquisition module, used to acquire target data and target images on the vehicle side, wherein the target data includes position feature data and environment feature data;

[0035] A building module, used for building a target three-dimensional map based on the target data;

[0036] The determination module is used to match the target three-dimensional map with the target image to determine the position of the target image in the target three-dimensional map.

[0037] In addition, to achieve the above objectives, the present application also proposes a vehicle positioning device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the vehicle positioning method described above.

[0038] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the vehicle positioning method described above are implemented.

[0039] In addition, to achieve the above-mentioned purpose, the present application also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps of the vehicle positioning method described above are implemented.

[0040] One or more technical solutions proposed in this application have at least the following technical effects:

[0041] This application first collects the location feature data and environmental feature data of the vehicle, which can accurately capture the dynamic position of the vehicle and detailed information of the surrounding environment, providing a basis for subsequent data processing and map construction, and ensuring the richness and accuracy of the map information. Then, using the collected data, a detailed three-dimensional map can be constructed, providing a comprehensive reference framework for the vehicle, which can better understand its environment. Finally, by matching the real-time target image with the established three-dimensional map, the exact position of the content in the image on the map can be determined, providing the vehicle with accurate positioning information, and enhancing the accuracy and reliability of navigation. This application uses data to construct a detailed three-dimensional map, which not only records the vehicle position, but also carefully reflects the surrounding environment characteristics. By matching the real-time captured image with this three-dimensional map, the precise positioning of the vehicle is achieved. Especially in urban environments and underground garages where satellite signals are limited, the challenges faced by traditional GPS and RTK positioning methods are effectively solved, greatly improving the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0043] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0044] Figure 1 A schematic diagram of a process flow provided for the first embodiment of the vehicle positioning method of the present application;

[0045] Figure 2 A schematic diagram of a flow chart provided for the second embodiment of the vehicle positioning method of the present application;

[0046] Figure 3 This is a schematic diagram of three-terminal data interaction of the vehicle positioning method according to an embodiment of the present application;

[0047] Figure 4 This is a schematic diagram of the module structure of the vehicle positioning device according to an embodiment of the present application;

[0048] Figure 5 Schematic diagram of the device structure of the hardware operating environment involved in the vehicle positioning method in the embodiment of the present application.

[0049] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0050] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0051] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0052] With the development of urbanization, people increasingly need cars as a means of transportation in semi-shielded areas such as shopping malls and residential areas. However, due to building obstructions in these places, commonly used GPS and RTK positioning are inaccurate, especially in places with no signal at all, such as underground garages, it becomes difficult to find a car, which limits the application of autonomous driving technology. At present, cars are mainly positioned by GPS and RTK, supplemented by IMU and on-board sensors. However, in places such as urban canyons and indoor parking lots, satellite signals are blocked and positioning accuracy is reduced. Existing technologies cannot accurately position in obstructed environments, affecting car finding and autonomous driving.

[0053] The main solution of the embodiment of the present application is: the present application firstly collects the position feature data and environmental feature data of the vehicle, so as to accurately capture the dynamic position of the vehicle and the detailed information of the surrounding environment, provide a basis for subsequent data processing and map construction, and ensure the richness and accuracy of the map information. Then, using the collected data, a detailed three-dimensional map can be constructed, which provides a comprehensive reference framework for the vehicle and enables a better understanding of its environment. Finally, by matching the real-time captured target image with the established three-dimensional map, the precise position of the content in the image on the map can be determined, providing the vehicle with accurate positioning information and enhancing the accuracy and reliability of navigation. The present application uses data to construct a detailed three-dimensional map, which not only records the vehicle position, but also carefully reflects the surrounding environment characteristics. By matching the real-time captured image with this three-dimensional map, the accurate positioning of the vehicle is achieved. Especially in urban environments and underground garages where satellite signals are limited, the challenges faced by traditional GPS and RTK positioning methods are effectively solved, greatly improving the user experience.

[0054] It should be noted that the execution subject of the embodiments of the present application may be a computing service device with data processing, network communication and program running functions, such as a tablet computer, a personal computer, a mobile phone, etc., or an electronic device capable of realizing the above functions. The following takes a computer as an example to illustrate this embodiment and the following embodiments.

[0055] Based on this, the embodiment of the present application provides a method for positioning a vehicle, referring to Figure 1 , Figure 1 This is a flow chart of the first embodiment of the vehicle positioning method of the present application.

[0056] In this embodiment, the vehicle positioning method includes steps S10 to S30:

[0057] Step S10, acquiring target data and target image at the vehicle end, wherein the target data includes position feature data and environment feature data;

[0058] It should be noted that target data refers to data collected from the vehicle for positioning and environmental understanding, which includes location feature data and environmental feature data. Among them, the location feature data describes the precise location of the vehicle, and the environmental feature data provides detailed information about the vehicle's surroundings. The target image refers to the image of the surrounding environment taken by the vehicle's on-board camera, which is used to determine the vehicle's location. The target image usually contains images of roads and lanes, including lane lines, road surface markings, surrounding buildings, landmarks, and natural features.

[0059] It is understandable that the vehicle actively collects two types of key data through its onboard sensor system: target data and target image. Target data consists of location feature data and environmental feature data, which can be collected by a variety of sensors such as on-board cameras, LiDAR, ultrasonic sensors, etc., and is used to identify environmental features around the vehicle. The target image is captured in real time by the on-board camera and contains visual information of the vehicle's surroundings, such as road signs, building outlines, traffic light status, etc. These image data provide intuitive visual references for subsequent image processing and map matching. By integrating these data, the vehicle can accurately perceive its position and status in the environment, providing necessary data support for subsequent three-dimensional map construction and vehicle positioning.

[0060] As an example, obtaining target data and target images on the vehicle side includes: sending a request to obtain target data so that the vehicle side collects position feature data and environmental feature data based on the acquisition request, the position feature data includes vehicle displacement data, the vehicle displacement data is obtained based on vehicle chassis data and inertial measurement data, the environmental feature data is obtained when it is determined that the vehicle is in motion or the vehicle displacement distance exceeds a preset distance threshold based on the vehicle chassis data and the inertial measurement data, the environmental feature data includes parking space feature information, traffic infrastructure information, and positioning information; sending a request to obtain a target image, the target image contains environmental information.

[0061] Among them, vehicle displacement data refers to the relevant information of the vehicle moving from a known position to another position, including the distance, direction, speed, etc. of the vehicle's movement. These data are crucial for determining the current position and path tracking of the vehicle. Vehicle chassis data involves the physical parameters of the vehicle, such as wheelbase, track width, etc., which help to understand the movement trajectory of the vehicle. Inertial measurement data (IMU data) comes from the inertial measurement unit, which provides data on the vehicle's acceleration, angular velocity, and sometimes attitude angles (pitch, roll, yaw), which reflect the vehicle's motion state. The preset distance threshold is a pre-set distance value used to determine when the vehicle's location information needs to be updated. When the vehicle's displacement distance exceeds this threshold, the system triggers new data collection to ensure that the update of the location data can meet the requirements of positioning accuracy. Parking space feature information includes the location, size, number, and whether the parking space is occupied. Traffic infrastructure information refers to traffic-related fixed facilities such as lane lines, traffic signs, signal lights, and road signs on the road. Positioning information refers to the real-time position and speed of the vehicle during movement, which is usually obtained through RTK or other positioning systems, and can also be obtained through Beidou satellite navigation information to obtain position and time information. Environmental information refers to the visual and physical characteristics of the vehicle's surroundings, such as buildings, road signs, natural features, etc. This information can be collected through on-board cameras, laser radar (LiDAR), radar, ultrasonic sensors and other devices to understand the vehicle's environment and assist the vehicle in positioning and navigation.

[0062] Specifically, by sending data acquisition requests, the location feature data and environmental feature data collected by these sensors are obtained. Position feature data mainly refers to the displacement information of the vehicle, which is calculated by analyzing the vehicle chassis data and IMU (inertial measurement unit) data, and reflects the movement of the vehicle. When the vehicle starts to move or its displacement exceeds the preset distance threshold, the system will pay special attention to and record these changes. Environmental feature data includes parking space information, traffic infrastructure, and positioning information, which help build an environmental model around the vehicle. At the same time, the vehicle will also send a target image acquisition request to capture images containing environmental information, such as visual data of surrounding buildings, traffic signs, and other important landmarks.

[0063] Step S20, establishing a target three-dimensional map based on the target data;

[0064] It should be noted that the target three-dimensional map refers to a detailed three-dimensional geospatial model created using location feature data and environmental feature data collected from vehicle sensors. This three-dimensional map can include the vehicle's driving path, including historical trajectories, as well as the three-dimensional structure and transportation infrastructure of the vehicle's surroundings, such as buildings, trees, obstacles, lane lines, traffic signs, traffic lights, sidewalks, intersections, etc. It can also include the location and size of parking spaces in parking lots. The target three-dimensional map not only provides a comprehensive reference framework for the vehicle, enabling it to better understand its environment, but can also be used for precise positioning and navigation of the vehicle. This map combines multiple sensor data and is generated through complex data processing and fusion algorithms to ensure the richness and accuracy of map information.

[0065] It can be understood that the target three-dimensional map established based on the target data is an accurate spatial model constructed through data processing and fusion technology using the location feature data and environmental feature data collected by the vehicle. This model displays in detail the vehicle's geographical location, surrounding environmental structure, transportation infrastructure, parking spaces and other information, which is used to support the vehicle's precise positioning and navigation.

[0066] Step S30: Match the target three-dimensional map with the target image to determine the position of the target image in the target three-dimensional map.

[0067] It can be understood that the 3D map constructed by the data uploaded by the vehicle is compared and matched with the image of the vehicle's surrounding environment captured in real time. Through advanced image processing and computer vision technology, it is possible to find corresponding points on the 3D map that match the characteristics of the target image, thereby accurately determining the specific location of the target image in the 3D map. This process is crucial to improving the positioning accuracy of the vehicle and the performance of the navigation system, especially in environments where the GPS signal is limited or completely blocked.

[0068] As an example, the target three-dimensional map is matched with the target image to determine the position of the target image in the target three-dimensional map, including: extracting image feature information from the target image based on a deep learning model; visually matching the image feature information with the target three-dimensional map to obtain a visual matching result; and obtaining the position of the target image in the target three-dimensional map based on the visual matching result.

[0069] Among them, image feature information refers to information extracted from the target image that helps to identify and understand the image content. This feature information usually includes key visual elements such as corners, edges, textures, shapes, and color distribution in the image. Deep learning models, such as convolutional neural networks (CNNs), can automatically identify and extract these features and convert them into numerical representations that can be used for further processing. The visual matching result refers to the result obtained after comparing and analyzing the extracted image feature information with the data in the target three-dimensional map. This process usually involves searching for landmarks or feature points in the three-dimensional map that match the image features. Through this matching, the position of the features in the image in the three-dimensional space can be determined, thereby achieving alignment between the image and the map. The visual matching result usually includes information such as the confidence of the match, the location of the matching points, and the possible matching error. These results are used to calculate the precise position of the target image in the three-dimensional map.

[0070] Specifically, a deep learning model is used to analyze the target image and extract key image feature information, such as shape, texture, and color. Then, this feature information is used for visual matching with the pre-built 3D map, and the visual matching results are obtained by identifying the correspondence between the image features and the map data. Finally, based on these matching results, the geographic location of the target image in the 3D map is accurately determined to provide accurate positioning information for the vehicle. This process combines advanced image processing technology and computer vision algorithms to improve the accuracy and reliability of positioning, especially in environments with limited GPS signals.

[0071] As an example, after determining the position of the target image in the target three-dimensional map, it also includes: acquiring a reference image of the mobile terminal; performing image semantic comparison positioning based on the reference image and the target three-dimensional map to obtain a positioning result.

[0072] Among them, the reference image refers to the image taken by the user in a specific environment through a mobile device (such as a smartphone or tablet). These images serve as references and contain environmental features around the user's current location, such as buildings, signs, street layouts, etc. The positioning result refers to the result obtained by semantically comparing the reference image taken by the user with the target three-dimensional map. This process usually involves the use of image processing and computer vision technology to identify and match features in the image with corresponding landmarks or feature points in the three-dimensional map. The positioning result provides the specific location of the user on the three-dimensional map, and may also include the user's direction and posture information. This positioning method is based on image semantics, that is, the visual content in the image is used to compare with map data to achieve accurate positioning. The positioning result enables the system to inform users of their location relative to a point or path on the map, which is very useful for applications such as navigation, positioning, and path planning. At the same time, the sensors and cameras on the mobile terminal can be continuously uploaded to the cloud for data interaction to update the target three-dimensional map, and path planning can be performed in the map based on the A* algorithm to navigate to the vehicle location and realize the function of finding the vehicle.

[0073] Specifically, users use mobile devices (such as smartphones) to take reference images of the surrounding environment, and then perform image semantic comparison analysis supported by deep learning algorithms with pre-built three-dimensional maps. The system can identify the correspondence between the features in the image and the map data, thereby accurately determining the user's position on the three-dimensional map. This positioning result provides users with the ability to navigate and locate in complex or obstructed environments. At the same time, users can continuously upload data to the cloud through mobile sensors and cameras for data interaction, and continuously update the three-dimensional map to ensure the real-time nature of the three-dimensional map and improve the accuracy of the three-dimensional map. In addition, users can plan paths in the map based on the A* algorithm, navigate to the vehicle location, and realize the function of finding people and cars.

[0074] As an example, after determining the position of the target image in the target three-dimensional map, it also includes: obtaining a positioning request and position information from a mobile terminal; matching the positioning information in the target data based on the position information to obtain a position matching result; and sending a corresponding three-dimensional map based on the position matching result.

[0075] Among them, a positioning request is a request initiated by a user to determine his current geographic location. Location information refers to data about the user's current location provided by the mobile terminal, which may include GPS coordinates, Wi-Fi signal strength, Bluetooth beacon information, mobile network signals, etc., which can assist the positioning process. The location matching result is the result obtained by comparing and matching the location information provided by the mobile terminal with the positioning information in the target data. The positioning information can be the positioning data provided by the Beidou satellite navigation system of China, including longitude and latitude, timestamp, etc. By matching the location information of the mobile terminal with the positioning information, the system can calibrate and confirm the actual location of the user and improve the accuracy of positioning. Once the location matching result is determined, the system will send three-dimensional map data matching the user's current location to the mobile terminal based on this result. In this way, the user can view a detailed map of his location on the mobile device, including a three-dimensional view of the surrounding environment, so as to navigate or locate. This process enables the user to obtain the corresponding map information and obtain accurate positioning information based on the three-dimensional map.

[0076] Specifically, when the user initiates a positioning request and provides location information, the system will match this information with the positioning information stored in the target data to obtain an accurate location matching result. Subsequently, based on this matching result, the system will send a three-dimensional map of a specific area that matches the user's current location to the mobile terminal, allowing the user to view the three-dimensional geographical environment around him on the mobile device, thereby performing effective navigation and positioning, realizing automated processing from positioning request to map data transmission, and improving positioning accuracy and navigation convenience.

[0077] This embodiment provides a method for positioning a car. The present application firstly collects the position feature data and environmental feature data of the vehicle, and can accurately capture the dynamic position of the vehicle and the detailed information of the surrounding environment, which provides a basis for subsequent data processing and map construction, and ensures the richness and accuracy of the map information. Then, using the collected data, a detailed three-dimensional map can be constructed, which provides a comprehensive reference framework for the vehicle and can better understand its environment. Finally, by matching the real-time captured target image with the established three-dimensional map, the precise position of the content in the image on the map can be determined, providing the vehicle with accurate positioning information, and enhancing the accuracy and reliability of navigation. The present application uses data to construct a detailed three-dimensional map, which not only records the vehicle position, but also carefully reflects the surrounding environment characteristics. By matching the real-time captured image with this three-dimensional map, the accurate positioning of the vehicle is achieved. Especially in urban environments and underground garages where satellite signals are limited, the challenges faced by traditional GPS and RTK positioning methods are effectively solved, and the user experience is greatly improved.

[0078] Based on the first embodiment of the present application, in the second embodiment of the present application, the same or similar contents as those in the above-mentioned embodiment 1 can refer to the above introduction, and will not be repeated later. Figure 2 , Figure 2 This is a flow chart of the second embodiment of the vehicle positioning method of the present application. Step S20 of the vehicle positioning method includes steps S21 to S23:

[0079] Step S21, performing position determination based on the target data to obtain environmental feature data of a nearby location;

[0080] It should be noted that position determination refers to the use of target data provided by the vehicle, such as IMU data, GPS / Beidou satellite navigation information, etc., to determine the current precise position and posture of the vehicle. This usually involves data fusion technology, combining data from different sensors to improve the accuracy and reliability of positioning. Environmental feature data of nearby locations refers to information about the surrounding environment related to the current position of the vehicle that is screened from a large number of sensors. These data describe the characteristics of the environment near the vehicle in detail. Position determination and obtaining environmental feature data of nearby locations are key steps in establishing accurate three-dimensional maps and performing effective positioning. With this information, the system can build a detailed environmental model around the vehicle and use it for subsequent path planning, navigation, and vehicle positioning.

[0081] It is understandable that position determination based on the target data refers to the use of various sensor data collected by the vehicle, such as GPS, IMU, on-board cameras, etc., to comprehensively analyze and determine the precise position and posture of the vehicle. Subsequently, the system extracts environmental feature data related to the current position of the vehicle, including important landmarks, traffic signs, lane lines and other information near the vehicle. These data provide key details for building a local three-dimensional environmental model and are used for vehicle navigation and positioning services. This step is to convert the vehicle's perception data into a deep understanding of the surrounding environment, laying the foundation for further positioning and navigation decisions.

[0082] Step S22, performing feature matching based on the environmental feature data to obtain a feature matching result;

[0083] It should be noted that feature matching refers to the technology in the field of computer vision and image processing that determines the relationship between two different images by identifying and corresponding significant feature points (such as corners, edges, textures, etc.) between them. The best matching point pair is found by comparing two sets of feature descriptors using matching algorithms (such as brute force matching, FLANN matching, KNN matching, etc.). The feature matching result refers to the output of the feature matching process, which usually includes matching point pairs, matching scores, and matching confidence. Matching point pairs refer to matching feature point pairs between two images. Each point pair contains a feature point in one image and a matching feature point in another image. The matching score refers to the similarity score of each matching point pair, which indicates the reliability of the match. The higher the score, the more similar the two feature points are, and the more reliable the match is. The matching confidence refers to the overall reliability of the matching process, which is usually evaluated by counting the number, distribution, and quality of matching point pairs. The feature matching result can be used to construct a map, and the location of the vehicle on the map is determined by the environmental feature data captured by the vehicle sensor.

[0084] It is understandable that feature matching based on the environmental feature data refers to the process of comparative analysis of environmental feature information such as lane lines, traffic signs, parking spaces, etc. collected by vehicle sensors. Through this process, the system can identify similar feature points between sensor data to determine the feature matching results. The feature matching results include information such as matching point pairs, matching scores, and matching confidence levels. These results are crucial for evaluating the accuracy and reliability of the match. They can be used for vehicle positioning, navigation, and path planning, especially in environments where GPS signals are limited or completely blocked. This process combines advanced computer vision and pattern recognition technologies to improve the positioning accuracy and navigation performance of vehicles in complex environments.

[0085] Step S23: If the feature matching result is that multiple positions overlap, a target three-dimensional map is obtained.

[0086] It should be noted that multi-position overlap means that during the feature matching process, the system finds that multiple sets of feature point pairs correspond to the same position or very close positions in the 3D map. When the feature matching results show multi-position overlap, the system can splice the overlapping data, and through splicing, a target 3D map can be established, and subsequent positioning and navigation calculations can be performed based on this.

[0087] It is understandable that when the feature matching process obtains multiple possible matching positions, that is, multiple positions overlap, this usually means that the vehicle's sensor data is very close in real position. In this case, the system can use this matching information to comprehensively analyze the relative position and spatial relationship of these sensor data points through data splicing and fusion technology to construct a target three-dimensional map. This map accurately reflects the characteristics of the vehicle's environment, provides precise positioning information for the vehicle, solves the position ambiguity problem caused by the association of sensor data with multiple potential matching positions, and thus provides a solid foundation for autonomous driving and navigation systems.

[0088] As an example, if the feature matching result is that multiple positions overlap, then after obtaining the target three-dimensional map, it also includes: performing a topological connection relationship check based on the target three-dimensional map to obtain an inspection result of the target three-dimensional map; if the inspection result is complete, completing the map detection; if the inspection result is incomplete, performing map calibration, so that the vehicle side obtains the corresponding reference map data according to the calibrated map and feeds it back to the cloud; and updating the target three-dimensional map according to the reference map data.

[0089] Among them, the topological connection relationship check refers to the verification process of the connection relationship between each element in the three-dimensional map. This includes checking whether roads, paths, parking spaces, etc. are correctly connected according to the logical and actual layout, and whether the topological structure between them is reasonable. For example, check whether a road is correctly connected to an intersection, or whether a parking space is correctly placed in the layout of the parking lot. The inspection result refers to the conclusion obtained after the topological connection relationship check, which indicates whether the topological structure of the map is complete, consistent and error-free. If the inspection result is "complete", it means that the map has passed all checks and there is no problem with its topological structure. If the inspection result is "incomplete", it means that there are missing or wrong connections in the map, and further correction is required. Map localization refers to the process of determining the precise location and attributes of specific elements on the map. When the inspection result shows that the map is incomplete, map localization is required to identify and correct the errors in the map. This usually involves using additional data or manual intervention to correct the position of map elements to match the actual environment or reference map. Reference map data refers to known accurate map data used for comparison and calibration with the current target three-dimensional map. This data can come from previously verified map versions, other sensor data, user feedback, or other reliable map services. Reference map data is used to assist the map calibration process, helping the vehicle side to obtain more accurate map information and feed it back to the cloud for map updates and corrections.

[0090] Specifically, the target 3D map is first used to check the topological connection relationship, analyzing whether the connections between elements such as roads, paths, and areas in the map are correct and complete. The inspection results indicate whether the map has passed the inspection. If the map is complete, the inspection is completed; if the map is incomplete, the map will be calibrated, which means that the vehicle will use the calibrated map to collect information corresponding to the reference map data and feed it back to the cloud. Then, use this reference map data to update and improve the target 3D map to ensure the accuracy and latestness of the map data, thereby providing reliable navigation and positioning services for the vehicle. This step ensures that the topological structure of the map is correct and can be corrected and updated in a timely manner when problems are found.

[0091] This embodiment first analyzes the location feature data (such as GPS, IMU data) and environmental feature data (such as parking space, traffic infrastructure information) provided by the vehicle, and can determine the approximate location of the vehicle and collect detailed environmental feature data near the location. This provides the necessary basic information for subsequent precise positioning and map construction. Then, using environmental feature data (such as lane lines, traffic signs, etc.), the system can perform feature matching on data obtained from different sensors, etc. When the feature matching results show that multiple positions overlap, the system uses data splicing and fusion technology to combine these location information to construct a target three-dimensional map. This map can accurately reflect the environmental characteristics of the vehicle and provide accurate positioning information for the vehicle. This embodiment realizes the whole process from data retrieval to the establishment of a three-dimensional map. This process not only improves the positioning accuracy of the vehicle in a complex environment, but also enhances the robustness of the system, enabling it to work normally in an environment where the GPS signal is limited or completely blocked. Through this continuous and iterative map construction process, this technical solution provides strong support for autonomous driving, intelligent transportation systems and vehicle navigation.

[0092] For example, in order to help understand the implementation process of the vehicle positioning method obtained by combining this embodiment with the above-mentioned embodiment 1, please refer to Figure 3 , Figure 3 A schematic diagram of three-terminal data interaction of a vehicle positioning method is provided, specifically:

[0093] The vehicle uses sensors such as cameras and IMUs to collect location feature data (such as vehicle displacement data) and environmental feature data (such as parking space feature information and traffic infrastructure information). These data are generated based on vehicle chassis data and IMU data, especially when the vehicle's motion state or displacement exceeds a preset threshold. The initially collected data is fused and positioned on the vehicle to generate a preliminary positioning result. These data and the generated preliminary positioning result are uploaded to the cloud. The cloud receives data from the vehicle or mobile end and performs splicing processing, splicing discrete data together, generating or updating the target three-dimensional map. The cloud saves and maintains this map to ensure its accuracy and latest status. The mobile end (such as the user's mobile phone) takes multiple pictures of the current scene and uploads these pictures to the cloud. The map sent from the cloud can be obtained to determine its own position.

[0094] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the vehicle positioning method of the present application. More simple transformations based on this technical concept are all within the protection scope of the present application.

[0095] This application also provides a car positioning device, please refer to Figure 4 , the vehicle positioning device comprises:

[0096] An acquisition module 10 is used to acquire target data and target images on the vehicle side, wherein the target data includes position feature data and environment feature data;

[0097] A building module 20, used to build a target three-dimensional map based on the target data;

[0098] The determination module 30 is used to match the target three-dimensional map with the target image to determine the position of the target image in the target three-dimensional map.

[0099] The automobile positioning device provided by the present application adopts the automobile positioning method in the above embodiment, and can solve the technical problem of how to obtain high-precision positioning of the automobile in various environments. Compared with the prior art, the beneficial effects of the automobile positioning device provided by the present application are the same as the beneficial effects of the automobile positioning method provided by the above embodiment, and other technical features in the automobile positioning device are the same as the features disclosed in the above embodiment method, which will not be repeated here.

[0100] The present application provides a car positioning device, which includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the car positioning method in the above-mentioned embodiment 1.

[0101] Reference below Figure 5, which shows a schematic diagram of the structure of a car positioning device suitable for implementing the embodiment of the present application. The car positioning device in the embodiment of the present application may include but is not limited to mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Descriptions), PMPs (Portable Media Players), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 5 The illustrated automobile positioning device is merely an example and should not limit the functions and scope of use of the embodiments of the present application.

[0102] like Figure 5 As shown, the automobile positioning device may include a processing device 1001 (e.g., a central processing unit, a graphics processor, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. In RAM1004, various programs and data required for the operation of the automobile positioning device are also stored. The processing device 1001, ROM1002, and RAM1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems can be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the vehicle positioning device to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows a vehicle positioning device with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can be implemented or have alternatively.

[0103] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0104] The automobile positioning device provided by the present application adopts the automobile positioning method in the above embodiment, and can solve the technical problem of how to obtain high-precision positioning of the automobile in various environments. Compared with the prior art, the beneficial effects of the automobile positioning device provided by the present application are the same as the beneficial effects of the automobile positioning method provided by the above embodiment, and other technical features in the automobile positioning device are the same as the features disclosed in the method of the previous embodiment, which will not be repeated here.

[0105] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0106] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0107] The present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the vehicle positioning method in the above-mentioned embodiment.

[0108] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with 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. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination of the above.

[0109] The computer-readable storage medium may be included in the vehicle positioning device; or may exist independently without being assembled into the vehicle positioning device.

[0110] The above-mentioned computer-readable storage medium carries one or more programs. When the above-mentioned one or more programs are executed by the automobile positioning device, the automobile positioning device: obtains target data and target image on the vehicle side, wherein the target data includes location feature data and environmental feature data; establishes a target three-dimensional map based on the target data; matches the target three-dimensional map with the target image, and determines the position of the target image in the target three-dimensional map.

[0111] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0112] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0113] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0114] The readable storage medium provided by the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned vehicle positioning method, and can solve the technical problem of how to obtain high-precision positioning of vehicles in various environments. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as the beneficial effects of the vehicle positioning method provided by the above-mentioned embodiment, and will not be repeated here.

[0115] The present application also provides a computer program product, including a computer program, wherein when the computer program is executed by a processor, the steps of the above-mentioned vehicle positioning method are implemented.

[0116] The computer program product provided by this application can solve the technical problem of how to obtain high-precision positioning of a car in various environments. Compared with the prior art, the beneficial effects of the computer program product provided by this application are the same as the beneficial effects of the car positioning method provided by the above embodiment, which will not be repeated here.

[0117] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for positioning a vehicle, characterized in that: The method comprises: Acquire target data and target image on the vehicle side, wherein the target data includes position feature data and environment feature data; Establishing a target three-dimensional map based on the target data; The target three-dimensional map is matched with the target image to determine the position of the target image in the target three-dimensional map.

2. The method according to claim 1, characterized in that The obtaining of the target data and the target image of the vehicle side includes: Sending a target data acquisition request so that the vehicle end collects position feature data and environmental feature data based on the acquisition request, wherein the position feature data includes vehicle displacement data, which is obtained based on vehicle chassis data and inertial measurement data; the environmental feature data is obtained when it is determined that the vehicle is in motion or when the vehicle displacement distance exceeds a preset distance threshold based on the vehicle chassis data and the inertial measurement data; the environmental feature data includes parking space feature information, traffic infrastructure information, and positioning information; A request for obtaining a target image is sent, where the target image includes environmental information.

3. The method according to claim 1, characterized in that The establishing of a target three-dimensional map based on the target data comprises: Performing position determination based on the target data to obtain environmental feature data of nearby locations; Perform feature matching based on the environmental feature data to obtain a feature matching result; If the feature matching result is that multiple positions coincide, a target three-dimensional map is obtained.

4. The method according to claim 3, characterized in that If the feature matching result is that multiple positions coincide, after obtaining the target three-dimensional map, the method further includes: Performing a topological connection relationship check based on the target three-dimensional map to obtain a check result of the target three-dimensional map; If the inspection result is complete, the map inspection is completed; If the inspection result is incomplete, map calibration is performed so that the vehicle obtains corresponding reference map data according to the calibrated map and feeds it back to the cloud; The target three-dimensional map is updated according to the reference map data.

5. The method according to claim 1, characterized in that The matching of the target three-dimensional map with the target image to determine the position of the target image in the target three-dimensional map includes: Extracting image feature information from the target image based on a deep learning model; Performing visual matching on the image feature information and the target three-dimensional map to obtain a visual matching result; Based on the visual matching result, the position of the target image in the target three-dimensional map is obtained.

6. The method according to claim 1, characterized in that After determining the position of the target image in the target three-dimensional map, the method further includes: Get the reference image of the mobile terminal; Image semantic comparison and positioning are performed based on the reference image and the target three-dimensional map to obtain a positioning result.

7. The method according to claim 1, characterized in that After determining the position of the target image in the target three-dimensional map, the method further includes: Obtain positioning request and location information from the mobile terminal; Matching the positioning information in the target data based on the position information to obtain a position matching result; Based on the position matching result, a corresponding three-dimensional map is sent.

8. A vehicle positioning device, characterized in that: The device comprises: An acquisition module, used to acquire target data and target images on the vehicle side, wherein the target data includes position feature data and environment feature data; A building module, used for building a target three-dimensional map based on the target data; The determination module is used to match the target three-dimensional map with the target image to determine the position of the target image in the target three-dimensional map.

9. A vehicle positioning device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the vehicle positioning method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the vehicle positioning method according to any one of claims 1 to 7 are implemented.

Images