Construction scene automatic excavation and data processing method and device, equipment and medium

Abstract

Embodiments of the present application provide a construction scene automatic mining and data processing method, device, equipment and medium. The method comprises: acquiring scene mining data, the scene mining data comprising at least one of vehicle trajectory data, crowd-sourced image data, network public opinion information and after-sales reports; based on the scene mining data, at least one candidate construction scene is identified; scene verification is performed on each candidate construction scene, and the map data is updated according to the verification result. The method is used to achieve high-reliability automatic mining of construction scenes, thereby improving the accuracy of map data.

Description

Technical Field

[0001] This application relates to the field of data processing, and in particular to a method, apparatus, equipment and medium for automatic mining and data processing of construction scenarios. Background Technology

[0002] Against the backdrop of the rapid development of intelligent transportation and autonomous driving technologies, the dynamic management of road construction scenarios has become a crucial aspect of ensuring traffic safety and improving the reliability of autonomous driving systems. Road construction is typically accompanied by static or semi-static elements such as construction barriers, temporary traffic signs, and construction vehicles, which significantly alter road traffic rules and driving behavior.

[0003] However, existing technologies have significant shortcomings in construction scenario management: the updating of construction information relies on manual reporting or preset markings, and cannot automatically discover newly emerging construction areas, resulting in map data lagging behind the actual construction progress. Summary of the Invention

[0004] This application provides a method, apparatus, equipment, and medium for automatic excavation and data processing of construction scenes, in order to achieve highly reliable automatic excavation of construction scenes, thereby improving the accuracy of map data.

[0005] In a first aspect, embodiments of this application provide a method for automatic mining and data processing of construction scenes, including:

[0006] Acquire scene mining data, which includes at least one of vehicle trajectory data, crowdsourced image data, online public opinion information, and after-sales reports;

[0007] Based on the scene mining data, at least one candidate construction scene is identified;

[0008] Each candidate construction scenario is validated, and the map data is updated based on the validation results.

[0009] In one possible implementation, identifying at least one candidate construction scenario based on the scenario mining data includes:

[0010] For the vehicle trajectory data in the scene mining data, the abnormal detour behavior of vehicles is analyzed based on the vehicle trajectory data, and the road sections with abnormal detour behavior are identified as candidate construction scenes.

[0011] For the crowdsourced image data in the scene mining data, feature analysis is performed on the crowdsourced image data based on image recognition data, and road sections with construction features in the images are identified as candidate construction scenes; the crowdsourced image data includes video data and image data of different roads;

[0012] Based on the online public opinion information mined from the scenario data, natural language processing technology is used to mine the road sections indicated by the online public opinion information that are currently under construction, and the road sections are identified as candidate construction scenarios.

[0013] Based on the after-sales reports in the data mining of the aforementioned scenarios, the road sections that caused the after-sales problems were analyzed and identified as candidate construction scenarios.

[0014] In one possible implementation, the step of performing scenario verification for each candidate construction scenario and updating the map data based on the verification results includes:

[0015] A scenario verification task is generated for each candidate construction scenario, and scenario verification is performed on the candidate construction scenario according to the scenario verification task to obtain the verification result of the candidate construction scenario; the scenario verification task is used to verify whether there is indeed a construction problem in the candidate construction scenario;

[0016] The candidate construction scenarios whose verification results indicate that there are indeed construction problems are identified as the target construction scenarios;

[0017] The map data is updated based on the construction information of each target construction scenario.

[0018] In one possible implementation, generating a scenario verification task for each candidate construction scenario, and performing scenario verification on the candidate construction scenarios according to the scenario verification task to obtain the verification result of the candidate construction scenarios, includes:

[0019] A geofence is created for each candidate construction scene, and an image acquisition command is sent to the target vehicle entering the geofence to obtain the actual image data of the candidate construction scene transmitted back by the target vehicle.

[0020] The candidate construction scenarios are verified and analyzed based on the actual image data to obtain verification information for the candidate construction scenarios. The verification information includes indication information indicating whether there are construction problems in the candidate construction scenarios and construction information for the candidate construction scenarios. The construction information includes construction type and construction boundaries.

[0021] In one possible implementation, updating the map data based on the construction information of each target construction scenario includes:

[0022] Event records are generated based on the construction information of each target construction scenario, and the blacklist / whitelist event table of the map data is updated based on the event records.

[0023] In one possible implementation, the method further includes:

[0024] If a vehicle is approaching a target construction site and its map update status is not updated, a speed reduction command is issued to the vehicle.

[0025] In one possible implementation, the method further includes:

[0026] The status of the map fence for each candidate construction scenario is updated, including a first status indicating that the candidate construction scenario has not been verified and a second status indicating that the candidate construction scenario has been verified.

[0027] Secondly, embodiments of this application provide an automatic excavation and data processing device for construction scenes, comprising:

[0028] The acquisition module is used to acquire scene mining data, which includes at least one of vehicle trajectory data, crowdsourced image data, online public opinion information, and after-sales reports.

[0029] The identification module is used to identify at least one candidate construction scenario based on the scenario mining data.

[0030] The verification and update module is used to verify each candidate construction scenario and update the map data based on the verification results.

[0031] In one possible implementation, the identification module is specifically used for:

[0032] For the vehicle trajectory data in the scene mining data, the abnormal detour behavior of vehicles is analyzed based on the vehicle trajectory data, and the road sections with abnormal detour behavior are identified as candidate construction scenes.

[0033] For the crowdsourced image data in the scene mining data, feature analysis is performed on the crowdsourced image data based on image recognition data, and road sections with construction features in the images are identified as candidate construction scenes; the crowdsourced image data includes video data and image data of different roads;

[0034] Based on the online public opinion information mined from the scenario data, natural language processing technology is used to mine the road sections indicated by the online public opinion information that are currently under construction, and the road sections are identified as candidate construction scenarios.

[0035] Based on the after-sales reports in the data mining of the aforementioned scenarios, the road sections that caused the after-sales problems were analyzed and identified as candidate construction scenarios.

[0036] In one possible implementation, the verification and update module includes:

[0037] The verification unit is used to generate a scenario verification task for each candidate construction scenario, and to perform scenario verification on the candidate construction scenario according to the scenario verification task to obtain the verification result of the candidate construction scenario; the scenario verification task is used to verify whether there is indeed a construction problem in the candidate construction scenario;

[0038] The determination unit is used to identify candidate construction scenarios where the verification results indicate that there are indeed construction problems as target construction scenarios;

[0039] The update unit is used to update the map data based on the construction information of each target construction scenario.

[0040] In one possible implementation, the verification unit is specifically used for:

[0041] A geofence is created for each candidate construction scene, and an image acquisition command is sent to the target vehicle entering the geofence to obtain the actual image data of the candidate construction scene transmitted back by the target vehicle.

[0042] The candidate construction scenarios are verified and analyzed based on the actual image data to obtain verification information for the candidate construction scenarios. The verification information includes indication information indicating whether there are construction problems in the candidate construction scenarios and construction information for the candidate construction scenarios. The construction information includes construction type and construction boundaries.

[0043] In one possible implementation, the updating unit is specifically used for:

[0044] Event records are generated based on the construction information of each target construction scenario, and the blacklist / whitelist event table of the map data is updated based on the event records.

[0045] In one possible implementation, the device further includes:

[0046] The sending module is used to send a speed reduction command to a vehicle approaching the target construction scene if the vehicle's map update status is not updated.

[0047] In one possible implementation, the device further includes:

[0048] The status update module is used to update the status of the map fence for each candidate construction scenario. The status includes a first status indicating that the candidate construction scenario has not been verified and a second status indicating that the candidate construction scenario has been verified.

[0049] Thirdly, embodiments of this application provide a cloud server, including: a memory and a processor;

[0050] The memory stores computer-executed instructions;

[0051] The processor executes computer execution instructions stored in the memory, causing the processor to perform the first aspect and / or various possible implementations of the first aspect as described above.

[0052] Fourthly, embodiments of this application provide a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the first aspect and / or various possible implementations of the first aspect.

[0053] Fifthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the first aspect and / or various possible implementations of the first aspect.

[0054] The construction scene automatic mining and data processing method, apparatus, equipment and medium provided in this application embodiment actively discovers construction scenes through multi-source data, ensures data credibility through verification, and finally realizes business value transformation through map data updates. This achieves automated management of the entire life cycle of construction scenes, realizes highly reliable automatic mining of construction scenes, and thus improves the accuracy of map data. Attached Figure Description

[0055] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0056] Figure 1 A flowchart illustrating an automatic construction scene mining and data processing method provided in Embodiment 1 of this application;

[0057] Figure 2 This is a flowchart illustrating a specific method for automatic excavation and data processing of construction scenarios, provided in Embodiment 2 of this application.

[0058] Figure 3 A schematic diagram of the technical framework for an automatic construction scene mining and data processing method provided in this application;

[0059] Figure 4 This is a schematic diagram of the structure of an automatic excavation and data processing device for a construction scene provided in Embodiment 3 of this application;

[0060] Figure 5 This is a schematic diagram of the structure of an automatic excavation and data processing device for a construction scene provided in Embodiment 4 of this application;

[0061] Figure 6 A schematic diagram of the structure of the cloud server provided in this application.

[0062] The accompanying drawings illustrate specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0063] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0064] Based on the aforementioned background technology, the inventors discovered during their research that the problem of data inaccuracy caused by delayed event updates can be solved by constructing a fully automated closed-loop process of "data mining-verification-map update." Specifically, the inventors considered using multi-source data fusion as a foundation, employing relevant verification technologies to achieve reliable scenario confirmation, and transforming construction information into executable business instructions through map data linkage, forming a complete data closed loop from perception to decision-making, which greatly improves the freshness of map data and management efficiency.

[0065] It should be noted that this application applies to the fields of intelligent transportation and autonomous driving, particularly addressing the dynamic management needs of road construction scenarios. Its network architecture includes at least: a cloud server (also referred to as the cloud), in-vehicle terminals (such as intelligent vehicles), a crowdsourced image platform, a high-precision map system, and a communication network based on Vehicle-to-Everything (V2X) technology. The cloud platform mines potential construction scenarios through multi-source data (vehicle trajectories, crowdsourced images, public opinion information, etc.), verifies the authenticity of the construction, and finally updates the map data based on the verification results. The automatic construction scenario mining and data processing provided in this application can be executed on the aforementioned cloud server.

[0066] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will now be described with reference to the accompanying drawings.

[0067] Figure 1 This is a flowchart illustrating an automatic construction scene mining and data processing method provided in Embodiment 1 of this application, as shown below. Figure 1 As shown, the method includes:

[0068] S101. Obtain scene mining data, wherein the scene mining data includes at least one of vehicle trajectory data, crowdsourced image data, online public opinion information, and after-sales reports.

[0069] In this step, the cloud will obtain scene mining data from multiple data sources that can be used for construction scene mining.

[0070] In detail, vehicle trajectory data refers to a collection of driving trajectory data from multiple vehicles. Each vehicle's trajectory data includes information such as its driving path, speed, and time. For example, the cloud can receive driving trajectory data sent by vehicles through in-vehicle terminals such as Telematics Boxes (T-BOX) or V2X technology; alternatively, the cloud can access the trajectory application programming interface (API) of a high-precision map service provider to acquire driving trajectory data from multiple vehicles in batches.

[0071] Crowdsourced video data refers to road-related visual data captured by the general public (such as car owners and citizens) using devices such as mobile phones, cameras, and dashcams. Specifically, this includes road photos and videos with the geographic coordinates of the shooting location. For example, the cloud can use compliant web crawling technology to scrape publicly available video content from social media platforms and filter materials related to road construction; and / or, it can also pre-connect to the automatic upload interfaces of some vehicle dashcams to obtain road images collected by the vehicles along their routes.

[0072] Online public opinion information refers to publicly available information related to road construction published on internet platforms, primarily in text form. This includes official media news reports on construction, construction announcements and traffic control notices issued by the transportation bureau and / or housing and construction bureau, user discussions on social media, and user feedback on car forums or local forums. For example, cloud-based systems can use compliant keyword crawling technology to retrieve relevant publicly available content from relevant internet platforms (such as news websites and forums) based on pre-defined construction-related keywords (such as road construction and road surface repair).

[0073] After-sales report data refers to records related to vehicle maintenance, including sales, parts, after-sales service, and information feedback (Sale, Sparepart, Service, Survey, 4S) repair records collected by automakers regarding vehicle repairs caused by road issues or traffic accidents. For example, the cloud can access the automaker's after-sales management system to obtain relevant after-sales report data.

[0074] It should be noted that all data mined from various scenarios involved in this solution (including but not limited to vehicle trajectory data, crowdsourced image data, online public opinion information, and after-sales report data) are information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, storage, use, processing, transmission, provision, disclosure, and application of related data all comply with the relevant laws, regulations, and standards of the relevant countries and regions, and necessary confidentiality measures have been taken, without violating public order and good morals.

[0075] In practical applications, the cloud can periodically acquire the latest scene mining data according to a preset acquisition cycle to ensure high real-time updates of map data. The preset acquisition cycle can be determined based on the actual application scenario (such as the computing power of the cloud server), and this application does not impose specific restrictions on it.

[0076] In addition, in practical applications, the scene mining data obtained from the cloud includes vehicle trajectory data, crowdsourced image data, online public opinion information, and one, two, or more of the data from after-sales reports. This application does not impose any restrictions on this.

[0077] S102. Based on scene mining data, identify at least one candidate construction scene.

[0078] In this step, the above-mentioned scene mining data will be analyzed to identify potential scenes that may be construction scenarios in the near future, which will be used as candidate construction scenarios.

[0079] The construction scenario refers to a road area where events affect the normal passage of traffic. These events include, for example, construction work on roads, facilities and sites, traffic accidents, traffic jams, obstacles, traffic control, event management, natural disasters and natural phenomena.

[0080] In one possible implementation, this step can be achieved using the methods described in 1) to 4) below:

[0081] 1) Based on the vehicle trajectory data in the scene mining data, analyze the abnormal detour behavior of vehicles and identify the road sections with abnormal detour behavior as candidate construction scenes.

[0082] Specifically, vehicle trajectory data can be analyzed using machine learning models (such as clustering algorithms) to identify road segments with abnormal vehicle detour behavior as candidate construction scenarios. For example, if cluster analysis of vehicle trajectory data confirms that road segment A has frequent vehicle detours and / or multiple sudden speed drops, then the area where road segment A is located is identified as a candidate construction scenario.

[0083] 2) For crowdsourced image data in scene mining data, feature analysis is performed on the crowdsourced image data based on image recognition data, and road sections with construction features in the images are identified as candidate construction scenes; where crowdsourced image data includes video data and image data of different roads.

[0084] Among them, construction features refer to visual elements in the image that can directly or indirectly point to "road construction", such as construction fences, excavators, warning signs, road damage, traffic control facilities, etc.

[0085] Specifically, pre-trained image recognition models can be used to automatically extract and analyze features from image data, identifying road sections where images with construction characteristics are captured as candidate construction scenes. The image recognition model can be a model with road construction scene recognition capabilities, pre-trained using a training set to obtain an initial recognition model. This initial recognition model could be, for example, a YOLO-based object detection model or a semantic segmentation model.

[0086] It should be noted that the selection of image recognition algorithms can be determined based on the actual application of the scheme, and this application does not impose specific restrictions on this.

[0087] 3) Based on the online public opinion information in the scene mining data, natural language processing technology is used to mine the road sections that are currently under construction as indicated in the online public opinion information, and the road sections are identified as candidate construction scenes.

[0088] Specifically, it will use natural language processing technology to perform word segmentation, entity recognition, semantic analysis, and other mining processes on online public opinion information, thereby transforming unstructured public opinion texts into structured and effective information.

[0089] For example, after segmenting the online public opinion information, key entity information such as road segment name, construction time, and construction scope can be identified, and semantic analysis can be performed based on the identified entities. Finally, at least one road segment indicating current construction status in the online public opinion information can be obtained from the semantic analysis results, and each indicated road segment can be identified as a candidate construction scenario.

[0090] 4) Analyze the after-sales reports in the scenario mining data to identify the road sections that caused the after-sales problems and determine the road sections as candidate construction scenarios.

[0091] Specifically, the relevant road sections indicated in the after-sales report that caused the after-sales issues need to be identified as candidate construction scenarios. These after-sales issues include, for example, abnormal tire wear, chassis scrapes, or damage caused by a car accident.

[0092] The method provided in this implementation utilizes multi-dimensional data processing techniques, including analysis of abnormal detour behavior in vehicle trajectory data, image recognition analysis of construction features from crowdsourced image data, natural language mining analysis of construction sections from online public opinion information, and attribution analysis of after-sales problem sections from after-sales report data, to mine candidate construction scenarios. This achieves automated discovery of construction scenarios. Simultaneously, the multi-source heterogeneous scenario mining data achieves synergistic complementarity, accurately identifying candidate construction scenarios with relevant characteristics from multiple perspectives. This enhances the comprehensiveness, accuracy, and timeliness of construction scenario identification, providing a reliable data foundation for subsequent map data updates.

[0093] S103. Perform scenario verification for each candidate construction scenario and update the map data based on the verification results.

[0094] In this step, to ensure the reliability of the data used for map updates, it is necessary to verify the authenticity of each candidate scene (i.e., confirm whether the candidate construction scene is indeed a construction scene where vehicles are driving normally in the image). Then, based on the credible candidate construction scenes indicated by the verification results, the construction events recorded in the map data are updated.

[0095] The application does not impose specific restrictions on the verification of candidate construction scenarios, such as cross-validation using multi-source data or dispatching personnel for on-site inspections.

[0096] It should be understood that the updated map data includes newly added candidate construction scenarios that have been verified and confirmed, which has higher real-time performance and ensures the accuracy of the map data.

[0097] The construction scene automatic mining and data processing method, apparatus, equipment and medium provided in this application actively discovers construction scenes through multi-source data, ensures data credibility through verification, and finally realizes business value transformation through map data updates. It generates accurate and usable construction scene map data from "nothing" to "something", realizes full life cycle automated management of construction scenes, achieves highly reliable automatic mining of construction scenes, and thus improves the accuracy of map data.

[0098] Figure 2 This is a flowchart illustrating a specific method for automatic construction scene mining and data processing provided in Embodiment 2 of this application, as shown below. Figure 2 As shown, based on the above embodiments, this embodiment provides a detailed description of the specific implementation of step S103 in the aforementioned embodiments, specifically including the following:

[0099] S1031. Generate a scenario verification task for each candidate construction scenario, and perform scenario verification on the candidate construction scenario according to the scenario verification task to obtain the verification result of the candidate construction scenario.

[0100] Among them, the scenario verification task is used to verify whether there are indeed construction problems in the candidate construction scenarios.

[0101] In this step, a scenario verification task will be automatically and in batches generated for each candidate construction scenario to ensure the orderliness and efficiency of the verification process.

[0102] In detail, each scenario verification task is used to verify whether there are events affecting normal traffic flow in the road area corresponding to the candidate construction scenario; if there are, it is determined that the candidate construction scenario does have a construction problem; if not, it is determined whether the candidate construction scenario does not have a construction problem.

[0103] In one possible implementation, the method described in steps 1.1 to 1.2 can be used:

[0104] Step 1.1: Create a geofence for each candidate construction scene and send an image acquisition command to the target vehicle entering the geofence to obtain the actual image data of the candidate construction scene transmitted back by the target vehicle.

[0105] In this step, a geofence will be created for each candidate construction scene. For each newly created geofence, an image acquisition command will be sent to at least one target vehicle entering the geofence, so that the vehicle can use its on-board image acquisition tool to take on-site photos of the candidate construction scene and send the acquired actual image data back to the cloud.

[0106] Specifically, the boundaries of the geofence need to be determined based on the geographical area corresponding to each candidate construction scenario. It should be understood that the creation of a geofence is equivalent to defining a precise "target area" for data collection; the actual image data will serve as visual evidence of the candidate construction scenario.

[0107] In practical applications, the cloud typically establishes communication data with multiple vehicles in advance and reports the vehicle location data in real time. Accordingly, the cloud can determine whether there are target vehicles that meet the requirements (such as vehicle type requirements) based on the location data reported by each vehicle and geofence them.

[0108] For example, during the task creation phase, a data acquisition fence task can be created in the geofence table in the cloud, and the geographical area to be collected (i.e., the geofence) and the target vehicle model for actual image data collection can be marked. During the task execution phase, if the cloud and the vehicle communicate via V2X technology, the cloud can determine at least one target vehicle that has entered the geofence based on the location data reported by each vehicle through the V2X link, and automatically send an image acquisition command to the target vehicle through the V2X link. After the vehicle completes the image data collection, it then transmits the collected actual image data back to the vehicle through the V2X link. It should be noted that the cloud and the vehicle can also use other methods for data interaction. This example only uses V2X to illustrate a specific implementation scenario and is not intended to limit the choice of communication method.

[0109] Step 1.2: Analyze and verify the candidate construction scene based on actual image data to obtain the verification information of the candidate construction scene.

[0110] The verification information includes indications of whether there are construction problems in the candidate construction scenario, as well as the construction information of the candidate construction scenario.

[0111] Specifically, construction information includes construction type and construction boundary. Construction type refers to the type of event corresponding to the candidate construction scenario, such as any one of the following: construction, accident, control, or natural disaster. Construction boundary refers to the scope of influence of the event corresponding to the candidate construction scenario, used to define the spatial range of the candidate construction scenario.

[0112] In this step, for each candidate construction scenario, manual or image analysis algorithms will be used to perform image analysis on the actual image data corresponding to the candidate construction scenario to determine whether the candidate construction scenario does indeed have an event that affects the normal passage of the road, and to assess the scope of the event's impact, thereby obtaining the verification information corresponding to the candidate construction scenario.

[0113] Furthermore, it should be understood that the geofences corresponding to each candidate construction scenario are stored in the geofence database. Therefore, before creating a geofence for each candidate construction scenario, it is possible to pre-query the geofence database to see if a corresponding geofence exists for that candidate construction scenario. Similarly, creating a geofence for each candidate construction scenario includes creating a map fence only for each candidate construction scenario for which a corresponding geofence does not exist in the geofence database. This method avoids the problem of redundant task execution caused by repeated task creation, thereby improving the efficiency of verification task execution.

[0114] Optionally, the method provided in this implementation further includes: updating the state of the map fence for each candidate construction scenario, wherein the state includes a first state indicating that the candidate construction scenario has not been verified and a second state indicating that the candidate construction scenario has been verified.

[0115] In detail, for geofences that have not completed verification, their status is marked as "first state" (e.g., unconfirmed); for geofences that have completed verification, their status is marked as "second state" (e.g., confirmed). Specifically, when a map fence is created, its initial status is marked as "first state," and after the verification task corresponding to the geofence is completed, the status of the geofence is changed to "second state."

[0116] In addition, since each geofence has the status label as described above, in practical applications, Apache Flink Change Data Capture (Apache Flink CDC) technology can be used to capture changes in the geofence database in real time, and verification operations can be performed only for tasks with the status label in the first state (including sending image acquisition instructions to target vehicles entering the geofence to obtain actual image data of candidate construction scenes returned by the target vehicles).

[0117] It should be understood that by accurately updating the map fence status of each candidate construction scenario, the fence management can be standardized, regulated and orderly, so that the flow and verification of fence status has a clear logical context and traceable management basis, avoiding problems such as repeated verification and status confusion, and effectively improving the organizational efficiency and management effectiveness of the overall scenario verification work.

[0118] The method provided in this implementation creates a dedicated geofence for each candidate construction scenario, accurately pinpoints the driving range of the target vehicle, and actively invokes the vehicle's onboard image acquisition capabilities. When the vehicle enters the geofence, it automatically issues an image acquisition command, obtains the actual image data of the candidate construction scenario transmitted back by the vehicle in real time, and then conducts targeted verification analysis based on the image data to obtain accurate verification information. This approach enables the verification of candidate construction scenarios to achieve real-time response, accurate coverage, and low-cost efficiency, effectively improving the authenticity and reliability of the verification information.

[0119] S1032. The candidate construction scenario whose verification results indicate that there is indeed a construction problem is identified as the target construction scenario.

[0120] Specifically, the verification results for each candidate construction scenario will indicate whether there are indeed construction problems in that candidate construction scenario.

[0121] S1033. Update the map data based on the construction information of each target construction scenario.

[0122] In this step, the construction information of each target construction scene needs to be added to the map data synchronously in order to update the map data.

[0123] In one possible implementation, this step includes: generating event records based on the construction information of each target construction scenario, and updating the blacklist / whitelist event table of the map data based on the event records.

[0124] Among them, the blacklist and whitelist event table refers to the time rule table for the effectiveness and control of events in the high-precision map, which includes event records bound to different road segments, used to record information such as the time range of various events and event types.

[0125] In detail, for each target construction scenario, the event type can be determined based on the construction type of the target construction scenario, and the time range of the event occurrence can be determined based on the effective time of the construction scenario obtained through mining, etc., to generate an event record corresponding to the target construction scenario; then, the road segment affected by the event can be determined based on the construction boundary of the target construction scenario; finally, the newly generated event record is bound to its corresponding road segment in the blacklist and whitelist event table to complete the update of the blacklist and whitelist event table.

[0126] The method provided in this implementation generates standardized event records based on the construction information of each target construction scenario, and accurately updates the blacklist and whitelist event tables of map data based on these records. This enables the mined construction scenarios to directly serve the maintenance of high-precision maps and the improvement of the safety of intelligent driving systems, thus realizing data-driven business.

[0127] It is foreseeable that for vehicles approaching the target construction site but whose map data has been updated in a timely manner, the navigation route will be dynamically adjusted based on the updated map data, combined with the vehicle's current location and driving status, to avoid detours, congestion, or accidental entry into restricted areas caused by outdated map data, thereby greatly improving traffic efficiency and driving experience in construction sites.

[0128] The automatic construction scenario mining and data processing method provided in this application generates a dedicated scenario verification task for each candidate construction scenario, accurately verifies whether there are construction problems and outputs the verification results, filters out target construction scenarios with construction problems, and accurately updates map data based on their construction information, so that the map data can reflect the actual road traffic status in real time and accurately, greatly improve the dynamic adaptability and practicality of the map, provide reliable data support for navigation planning and traffic control, and effectively ensure road traffic safety and driving experience.

[0129] Furthermore, in one possible implementation, based on the foregoing embodiments, after updating the map data, the method provided in this application may further include:

[0130] If a vehicle is approaching a target construction site and its map update status is not updated, a speed reduction command will be issued to the vehicle.

[0131] It should be understood that after the map data is updated in the cloud, some vehicles approaching the target construction site may not be able to synchronize the map data with their vehicles in a timely manner due to factors such as vehicle hardware capabilities, network conditions, and software compatibility. Therefore, for these vehicles, the cloud will immediately send a speed reduction command to prevent them from entering the construction area due to outdated map data, or from causing collisions, congestion, or other safety risks due to a lack of understanding of road conditions, thus ensuring vehicle safety and road traffic order.

[0132] In practical applications, the cloud can locate vehicles approaching the target construction scene based on the location data reported by each vehicle. In addition, vehicles usually send heartbeat packets to the cloud with which they have established a pre-existing communication connection. The heartbeat packets can carry the version number and update time of the map data they are using. Thus, the cloud can issue a speed reduction command to vehicles approaching the target construction scene that have not been updated in time, based on the location data reported by each vehicle and the version number and update time of the map data.

[0133] The method provided in this implementation effectively avoids the problem of vehicles accidentally entering construction areas due to untimely map data updates by actively issuing speed reduction commands to vehicles approaching the target construction scene and whose map update status is not updated. This effectively ensures vehicle driving safety and road traffic order.

[0134] Figure 3 This application provides a schematic diagram of the technical framework for an automatic construction scene mining and data processing method, as shown below. Figure 3 As shown, after the cloud receives multi-source scene mining data, it identifies candidate construction scenes through a candidate construction scene intelligent mining engine. For each candidate construction scene, a scheduling verification is performed to determine whether the candidate construction scene is the target construction scene. If it is, the task flow for that candidate construction scene is terminated after the map data is updated. If it is not, the task flow for that candidate construction scene is terminated. During the verification task, map fences are generated and stored in the cloud, and the verification results returned based on the map fences are received.

[0135] The technical framework provided in this application can realize the automated management of the entire life cycle of construction scenarios, achieve highly reliable automatic excavation of construction scenarios, and thus improve the accuracy of map data.

[0136] Figure 4 This is a structural schematic diagram of an automatic excavation and data processing device for a construction scene provided in Embodiment 3 of this application, as shown below. Figure 4 As shown, the automatic construction scene excavation and data processing device 20 provided in this embodiment includes:

[0137] The acquisition module 201 is used to acquire scene mining data, which includes at least one of vehicle trajectory data, crowdsourced image data, online public opinion information, and after-sales reports.

[0138] The identification module 202 is used to identify at least one candidate construction scenario based on scene mining data;

[0139] The verification and update module 203 is used to verify each candidate construction scenario and update the map data based on the verification results.

[0140] The automatic construction scene excavation and data processing device 20 provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0141] Figure 5 This is a structural schematic diagram of an automatic excavation and data processing device for a construction scene provided in Embodiment 4 of this application, as shown below. Figure 5 As shown, based on the above embodiments, the automatic construction scene excavation and data processing device 20 provided in this embodiment further includes:

[0142] The sending module 204 is used to send a speed reduction command to a vehicle approaching the target construction scene if the vehicle's map update status is not updated.

[0143] The status update module 205 is used to update the status of the map fence for each candidate construction scenario. The status includes a first status indicating that the candidate construction scenario has not been verified and a second status indicating that the candidate construction scenario has been verified.

[0144] In one possible implementation, the identification module 202 is specifically used for:

[0145] Based on the vehicle trajectory data in the scene mining data, abnormal detour behavior of vehicles is analyzed, and road sections with abnormal detour behavior are identified as candidate construction scenes.

[0146] For crowdsourced image data in scene mining data, feature analysis is performed on the crowdsourced image data based on image recognition data, and road sections with construction features in the images are identified as candidate construction scenes; crowdsourced image data includes video data and image data of different roads;

[0147] Based on the online public opinion information in the scene mining data, natural language processing technology is used to mine the road sections that are currently under construction as indicated in the online public opinion information, and the road sections are identified as candidate construction scenes.

[0148] Based on the after-sales reports in the scenario mining data, we analyze and identify the road sections that cause after-sales problems, and determine these road sections as candidate construction scenarios.

[0149] In one possible implementation, the verification and update module 203 includes:

[0150] The verification unit is used to generate a scenario verification task for each candidate construction scenario, and to perform scenario verification on the candidate construction scenario according to the scenario verification task to obtain the verification result of the candidate construction scenario; the scenario verification task is used to verify whether there are indeed construction problems in the candidate construction scenario;

[0151] The determination unit is used to identify candidate construction scenarios where the verification results indicate that there are indeed construction problems as target construction scenarios;

[0152] The update unit is used to update the map data based on the construction information of each target construction scenario.

[0153] In one possible implementation, the verification unit is specifically used for:

[0154] Create a geofence for each candidate construction scene and send an image acquisition command to the target vehicle that enters the geofence to obtain the actual image data of the candidate construction scene returned by the target vehicle.

[0155] Verification analysis of candidate construction scenarios is performed based on actual image data to obtain verification information for the candidate construction scenarios. The verification information includes indications of whether there are construction problems in the candidate construction scenarios and construction information for the candidate construction scenarios. The construction information includes construction type and construction boundaries.

[0156] In one possible implementation, the updating unit is specifically used for:

[0157] Event records are generated based on the construction information of each target construction scenario, and the blacklist and whitelist event tables of the map data are updated based on the event records.

[0158] The automatic construction scene excavation and data processing device 20 provided in this embodiment can execute the method provided in the above method embodiment. Its implementation principle and technical effect are similar, and will not be described in detail here.

[0159] Figure 6 This is a schematic diagram of the cloud server structure provided in this application. Figure 6As shown, the cloud server 30 provided in this embodiment includes at least one processor 301 and a memory 302. Optionally, the cloud server 30 further includes a communication component 303. The processor 301, memory 302, and communication component 303 are connected via a bus 304.

[0160] In a specific implementation, at least one processor 301 executes computer execution instructions stored in memory 302, causing at least one processor 301 to perform the above-described method.

[0161] The specific implementation process of processor 301 can be found in the above method embodiments, and its implementation principle and technical effect are similar. It will not be repeated here.

[0162] In the above embodiments, it should be understood that the processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), etc. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the method disclosed in this invention can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor.

[0163] The memory may include read-only memory and random access memory. The memory may be volatile or non-volatile, or may include both. Non-volatile memory may include read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. Volatile memory may include random access memory (RAM), which serves as an external cache. Many forms of RAM are available by way of example, but not limitation. Examples include Static Random Access Memory (SRAM), Dynamic Random Access Memory (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), Enhanced Synchronous DRAM (ESDRAM), Sync Link DRAM (SLDRAM), and Direct Rambus RAM (DR RAM).

[0164] The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, the buses shown in the accompanying drawings are not limited to a single bus or a single type of bus.

[0165] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method.

[0166] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the above-described method.

[0167] The aforementioned readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as SRAM, EEPROM, EPROM, PROM, ROM, magnetic storage, flash memory, magnetic disk, or optical disk. The readable storage medium can be any available medium accessible to a general-purpose or special-purpose computer.

[0168] An exemplary readable storage medium is coupled to a processor, enabling the processor to read information from and write information to the readable storage medium. Of course, the readable storage medium can also be a component of the processor. The processor and the readable storage medium can reside within an ASIC. Alternatively, the processor and the readable storage medium can exist as discrete components in a device.

[0169] The division of units is merely a logical functional division; in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices, or units, and may be electrical, mechanical, or other forms.

[0170] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0171] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0172] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0173] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0174] Finally, it should be noted that other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention that follow the general principles of the invention and include common knowledge or customary techniques in the art not disclosed herein, and is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of the invention is limited only by the appended claims.

Claims

1. A method for automatic excavation and data processing of construction scenes, characterized in that, The method includes: Acquire scene mining data, which includes at least one of vehicle trajectory data, crowdsourced image data, online public opinion information, and after-sales reports; Based on the scene mining data, at least one candidate construction scene is identified; Each candidate construction scenario is validated, and the map data is updated based on the validation results.

2. The method according to claim 1, characterized in that, Based on the scene mining data, at least one candidate construction scene is identified, including: For the vehicle trajectory data in the scene mining data, the abnormal detour behavior of vehicles is analyzed based on the vehicle trajectory data, and the road sections with abnormal detour behavior are identified as candidate construction scenes. For the crowdsourced image data in the scene mining data, feature analysis is performed on the crowdsourced image data based on image recognition data, and road sections with construction features in the images are identified as candidate construction scenes; the crowdsourced image data includes video data and image data of different roads; Based on the online public opinion information mined from the scenario data, natural language processing technology is used to mine the road sections indicated by the online public opinion information that are currently under construction, and the road sections are identified as candidate construction scenarios. Based on the after-sales reports in the data mining of the aforementioned scenarios, the road sections that caused the after-sales problems were analyzed and identified as candidate construction scenarios.

3. The method according to claim 1 or 2, characterized in that, The step of verifying each candidate construction scenario and updating the map data based on the verification results includes: A scenario verification task is generated for each candidate construction scenario, and scenario verification is performed on the candidate construction scenario according to the scenario verification task to obtain the verification result of the candidate construction scenario; the scenario verification task is used to verify whether there is indeed a construction problem in the candidate construction scenario; The candidate construction scenarios whose verification results indicate that there are indeed construction problems are identified as the target construction scenarios; The map data is updated based on the construction information of each target construction scenario.

4. The method according to claim 3, characterized in that, The process of generating a scenario verification task for each candidate construction scenario and performing scenario verification on the candidate construction scenarios according to the scenario verification task to obtain the verification results of the candidate construction scenarios includes: A geofence is created for each candidate construction scene, and an image acquisition command is sent to the target vehicle entering the geofence to obtain the actual image data of the candidate construction scene transmitted back by the target vehicle. The candidate construction scenarios are verified and analyzed based on the actual image data to obtain verification information for the candidate construction scenarios. The verification information includes indication information indicating whether there are construction problems in the candidate construction scenarios and construction information for the candidate construction scenarios. The construction information includes construction type and construction boundaries.

5. The method according to claim 3, characterized in that, The step of updating the map data based on the construction information of each target construction scenario includes: Event records are generated based on the construction information of each target construction scenario, and the blacklist / whitelist event table of the map data is updated based on the event records.

6. The method according to claim 1 or 2, characterized in that, The method further includes: If a vehicle is approaching a target construction site and its map update status is not updated, a speed reduction command is issued to the vehicle.

7. The method according to claim 4, characterized in that, The method further includes: The status of the map fence for each candidate construction scenario is updated, including a first status indicating that the candidate construction scenario has not been verified and a second status indicating that the candidate construction scenario has been verified.

8. An automatic excavation and data processing device for construction scenarios, characterized in that, include: The acquisition module is used to acquire scene mining data, which includes at least one of vehicle trajectory data, crowdsourced image data, online public opinion information, and after-sales reports. The identification module is used to identify at least one candidate construction scenario based on the scenario mining data. The verification and update module is used to verify each candidate construction scenario and update the map data based on the verification results.

9. A cloud server, characterized in that, include: Memory, processor; The memory stores computer-executed instructions; The processor executes computer execution instructions stored in the memory, causing the processor to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the method as described in any one of claims 1-7.

Images