Road construction area map layer construction method and device, equipment, storage medium and product

Abstract

The invention relates to the technical field of intelligent driving, in particular to a road construction area map layer construction method and device, equipment, a storage medium and a product. According to the scheme, vehicle end road environment perception data uploaded by crowdsourcing vehicles are used for recognizing and extracting construction related elements, and spatial clustering is carried out on the construction related elements to obtain a recognition perception result of a construction area; meanwhile, road width abrupt change areas are detected in combination with road reference data, construction areas are determined through the spatial incidence relation of the road width abrupt change areas and related attribute parameters are calculated, and then a standardized construction layer of each construction area is constructed and accurately pushed. According to the scheme, through a dual verification mechanism of construction element space aggregation and road width mutation, single feature misjudgment is effectively eliminated, the construction area recognition accuracy is improved, the dynamic layer is generated through the standardized attribute parameter set, structural description and efficient transmission of construction information are achieved, and the construction efficiency is improved. And a real-time and reliable road construction situation sensing capability is provided for a terminal user.

Description

Technical Field

[0001] This invention relates to the field of intelligent driving technology, and in particular to methods, apparatus, equipment, storage media, and products for constructing map layers in road construction areas. Background Technology

[0002] With the continuous evolution of intelligent transportation and autonomous driving technologies, high-precision maps, as an important supplement to vehicle perception systems, are crucial in terms of timeliness and accuracy. Traditional map update methods mainly rely on professional data collection fleets to conduct periodic road surveys. This approach is costly and has a long update cycle, making it difficult to meet the real-time requirements of scenarios with frequent changes in road facilities. In particular, the information lag is even more pronounced for dynamic and unpredictable temporary construction events.

[0003] To overcome the limitations of traditional map update mechanisms, the industry has begun exploring the use of crowdsourced vehicles equipped with environmental perception sensors as dynamic data sources. These vehicles can continuously generate massive amounts of road environment perception data during their daily driving, theoretically forming a wide-coverage and highly effective macroscopic sensor network. However, in practice, crowdsourced data is fragmented, asynchronous, and noisy, leading to inaccurate positioning of construction elements; secondly, scattered construction elements are easily confused with temporary road conditions, resulting in a high false alarm rate based on this single feature; and thirdly, crowdsourced data struggles to overcome perception errors and lacks a reliable fusion and identification mechanism.

[0004] Therefore, how to achieve accurate real-time identification of road construction areas based on scattered crowdsourced vehicle perception data has become a technical problem that urgently needs to be solved in this industry. Summary of the Invention

[0005] The main objective of this invention is to provide a method, apparatus, device, storage medium, and product for constructing a road construction area map layer, aiming to solve the technical problem of how to achieve accurate real-time identification of road construction areas based on scattered crowdsourced vehicle perception data in the prior art.

[0006] To achieve the above objectives, the present invention provides a method for constructing a map layer for a road construction area, the method comprising the following steps: Based on the vehicle-side road environment perception data, a set of road features is obtained; Based on the set of road features, the spatial clustering area of ​​construction elements is obtained; Based on the road baseline data and the road feature set, the road width abrupt change region is obtained; Based on the spatial aggregation area of ​​the construction elements and the abrupt change area of ​​the road width, the construction area and the set of attribute parameters of the construction area are determined. Based on the set of attribute parameters of the construction area, a construction layer is constructed and the construction layer is pushed to the user terminal.

[0007] Optionally, before obtaining the road feature set based on the vehicle-side road environment perception data, the method further includes: The received raw asynchronous data packets from the vehicle are cleaned to obtain initial road perception data; The coordinates of each sensor coordinate system in the initial road perception data are transformed to a unified global coordinate system to obtain road perception data with global coordinate alignment. The road perception data after global coordinate alignment is subjected to time-series alignment processing to obtain the vehicle-side road environment perception data.

[0008] Optionally, the original vehicle-side asynchronous data packet includes at least: Local map data, which includes lightweight map fragments obtained by reconstructing a local environment that meets preset local mapping conditions from the vehicle end; The raw vehicle perception data includes road images, point cloud data, obstacle results, and road semantic recognition results. Vehicle status data, which includes vehicle identity information, vehicle positioning coordinates, vehicle heading angle, vehicle speed, and upload timestamp.

[0009] Optionally, obtaining the road feature set based on the vehicle-side road environment perception data includes: Target recognition is performed on the vehicle-side road environment perception data to obtain road element recognition results; The road element recognition results are associated with the identification information of the perception data source to obtain road element instance data; The road element instance data is divided according to element type to obtain element subsets of each feature category; The road feature set is obtained by summing the subsets of elements from each feature category.

[0010] Optionally, obtaining the spatial clustering area of ​​construction elements based on the road feature set includes: Based on the road feature set, feature points and their coordinates that are identified as construction-related elements are extracted to obtain the construction element set. Based on the density clustering strategy, the construction elements in the set of construction elements are spatially clustered to obtain candidate clustering regions; The candidate clustering region is subjected to boundary generation processing to obtain the spatial clustering region of the construction elements.

[0011] Optionally, the density-based clustering strategy, which spatially clusters the construction elements in the set of construction elements to obtain candidate clustering regions, includes: Based on the preset neighborhood radius parameter and cluster point number parameter, cluster analysis is performed on the coordinate points of each element in the construction element set to obtain the initial clustering result; The coordinate points that are density-connected in the initial clustering results are assigned to the same cluster, and the coordinate points that cannot be assigned to any cluster are marked as noise points, thus obtaining the set of classification points; The set of coordinate points of each cluster formed by the classification point set is taken as the candidate clustering region.

[0012] Optionally, the boundary generation process performed on the candidate clustering region to obtain the spatial clustering region of the construction elements includes: Based on the set of coordinate points contained in the candidate cluster area, calculate the minimum bounding geometry of the set of coordinate points to obtain the initial geographical range; Based on the initial geographical range, generate the regional boundaries corresponding to the candidate clustering areas; Based on the area boundary and its geographic coordinates, the spatial clustering area of ​​the construction elements is obtained.

[0013] Optionally, obtaining the road width abrupt change region based on the road reference data and the road feature set includes: Based on the aforementioned road feature set, road width perception data is obtained; Based on the geographic location coordinates in the road width sensing data, the corresponding road reference width data is obtained by querying the road reference data. Based on the road width sensing data and the road reference width data, the road width abrupt change area is obtained.

[0014] Optionally, obtaining road width sensing data based on the road feature set includes: Based on the road feature set, extract the lane line features and road edge features; Based on the lane line features and road edge features, the geographical coordinates of the corresponding road boundary are obtained; The perceived width value is calculated based on the geographical coordinates of the road boundary. The road width perception data is obtained based on the perceived width value and its corresponding geographical coordinates.

[0015] Optionally, obtaining the road width abrupt change region based on the road width sensing data and the road reference width data includes: Based on the perceived width value in the road width perception data and the reference width value in the road reference width data, calculate the road width change rate for each data point; The road width change rate is compared with a preset change threshold to determine the data points where a sudden change in road width occurs. The data points where road width changes abruptly are clustered according to their geographical coordinates to obtain the areas where road width changes abruptly.

[0016] Optionally, determining the construction area based on the spatial clustering area of ​​the construction elements and the area of ​​abrupt changes in road width includes: The deterministic construction area set is obtained by the spatial intersection of the spatial clustering area of ​​construction elements and the area of ​​abrupt change in road width; The areas where the spatial intersection is excluded from the road width abrupt change areas are selected as the candidate construction area set; The candidate construction area set is then verified a second time based on the spatial relationship between the road closure query results and the candidate construction area set to obtain the verified construction area set. The construction area is determined by merging the set of deterministic construction areas with the set of verified construction areas.

[0017] Optionally, obtaining a deterministic set of construction areas based on the spatial intersection of the spatial clustering area of ​​construction elements and the area of ​​abrupt change in road width includes: The overlapping area ratio is obtained based on the geographical coordinate distribution of the spatial clustering area of ​​construction elements and the area of ​​abrupt change in road width; Based on the comparison between the overlapping region ratio and the preset overlap threshold, the target region that satisfies the dual characteristics is determined; The target areas are summarized to obtain the set of deterministic construction areas.

[0018] Optionally, after merging the deterministic construction area set with the verified construction area set to determine the construction area, the method further includes: Based on the spatial outline coordinates of the construction area, the geographical range parameters are obtained; Based on the distribution characteristics of construction element types within the construction area, construction type parameters are obtained; Confidence parameters are calculated based on the statistical values ​​of the number of construction elements in the construction area, the spatial overlap index, and the type of data acquisition equipment. Based on the geographical range parameter, construction type parameter, and confidence level parameter, the set of attribute parameters for the construction area is obtained.

[0019] Optionally, constructing a construction layer based on the set of attribute parameters of the construction area includes: Based on the set of attribute parameters, standardized description information corresponding to each construction area is generated; The standardized description information is encapsulated into a dynamic layer data object according to a preset data encapsulation format; The construction layer is obtained by summarizing the dynamic layer data objects of each construction area.

[0020] Optionally, the standardized description information includes construction type information, construction area information, confidence level information, time parameter information, and traffic behavior guidance information; The step of generating standardized description information for each construction area based on the set of attribute parameters includes: The construction type information is obtained based on the construction element type composition information in the attribute parameter set; The construction area information is obtained based on the location range information and regional geometric feature information in the attribute parameter set; The confidence information is obtained by detecting the confidence parameters of the construction area in the set of attribute parameters. The time parameter information is obtained based on the construction start timestamp and the expected duration parameter in the attribute parameter set; Based on the construction type information and the construction area information, and combined with the set of traffic behavior rules, the traffic behavior guidance information is obtained; The standardized description information is obtained by summarizing the construction type information, construction area information, confidence level information, time parameter information, and traffic behavior guidance information.

[0021] Optionally, pushing the construction layer to the user terminal includes: The distance determination result is obtained based on the real-time relative distance between the current location of the user terminal and the geographical area covered by the construction layer; When the distance determination result meets the preset warning distance threshold condition, a data distribution instruction is generated; The data distribution instruction drives the geographic message subscription system to push the construction layer to the target user terminal to complete the push of the real-time construction layer.

[0022] Furthermore, to achieve the above objectives, the present invention also proposes a road construction area map layer construction device, the road construction area map layer construction device comprising: Based on the vehicle-side road environment perception data, a set of road features is obtained; Based on the set of road features, the spatial clustering area of ​​construction elements is obtained; Based on the road baseline data and the road feature set, the road width abrupt change region is obtained; Based on the spatial aggregation area of ​​the construction elements and the abrupt change area of ​​the road width, the construction area and the set of attribute parameters of the construction area are determined. Based on the set of attribute parameters of the construction area, a construction layer is constructed and the construction layer is pushed to the user terminal.

[0023] Furthermore, to achieve the above objectives, the present invention also proposes a road construction area map layer construction device, which includes: a memory, a processor, and a road construction area map layer construction program stored in the memory and executable on the processor. The road construction area map layer construction program is configured to implement the steps of the road construction area map layer construction method described above.

[0024] In addition, to achieve the above objectives, the present invention also proposes a storage medium storing a road construction area map layer construction program, wherein when the road construction area map layer construction program is executed by a processor, it implements the steps of the road construction area map layer construction method described above.

[0025] In addition, to achieve the above objectives, the present invention also proposes a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the road construction area map layer construction method described above.

[0026] The proposed technical solutions (one or more) have at least the following technical effects: This solution preprocesses and unifies the coordinates of raw perception data from the vehicle to obtain standardized road environment perception data. By performing target recognition and classification on this data, a road feature set containing various features such as construction elements and lane lines is extracted. Then, based on a density clustering algorithm, spatial clustering areas are identified from the construction elements. Simultaneously, by comparing the real-time perceived road width with baseline data, areas of abrupt changes in road width are detected. Then, by combining the spatial overlap and correlation analysis of the above two types of areas, the extent of the construction area is accurately determined and its various attribute parameters are calculated. Finally, a standardized construction layer is generated based on the attribute parameters and dynamically pushed according to the terminal location. The above solution improves the accuracy of identification by fusing multiple types of road features for joint inference, which avoids misjudgments caused by relying solely on a single feature to identify construction areas. It utilizes crowdsourced data from the vehicle to achieve extensive and continuous detection of construction areas, overcoming the limitations of slow updates in traditional methods and significantly enhancing the timeliness of information. Encapsulating construction information into a standardized dynamic layer and actively pushing it allows end users to promptly obtain road dynamics, providing effective support for route planning and driving safety. Attached Figure Description

[0027] 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.

[0028] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 This is a flowchart illustrating the first embodiment of the method for constructing a map layer for a road construction area according to the present invention. Figure 2 This is a schematic diagram of the original sensing data preprocessing process for the road construction area map layer construction method of the present invention; Figure 3 This is a flowchart illustrating the second embodiment of the method for constructing a map layer for a road construction area according to the present invention. Figure 4 This is a flowchart illustrating the third embodiment of the road construction area map layer construction method of the present invention; Figure 5 This is a schematic diagram of the construction element aggregation process of the road construction area map layer construction method of the present invention; Figure 6 This is a flowchart illustrating the fourth embodiment of the road construction area map layer construction method of the present invention; Figure 7 This is a schematic diagram illustrating the road width calculation process of the road construction area map layer construction method of the present invention; Figure 8 This is a flowchart illustrating the fifth embodiment of the road construction area map layer construction method of the present invention; Figure 9 This is a schematic diagram illustrating the generation of attribute parameter sets for the road construction area map layer construction method of the present invention; Figure 10 This is a flowchart illustrating the sixth embodiment of the method for constructing a map layer for a road construction area according to the present invention. Figure 11 This is a schematic diagram of the construction layer push process in the road construction area map layer construction method of the present invention; Figure 12 This is a structural block diagram of the first embodiment of the road construction area map layer construction device of the present invention; Figure 13 This is a schematic diagram of the structure of a road construction area map layer construction device in the hardware operating environment involved in the embodiments of the present invention.

[0030] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0031] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0032] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0033] The main solution of this application embodiment is as follows: Based on the vehicle-side road environment perception data, a road feature set is obtained; based on the road feature set, a spatial aggregation area of ​​construction elements is obtained; based on the road reference data and the road feature set, a road width abrupt change area is obtained; based on the spatial aggregation area of ​​construction elements and the road width abrupt change area, a construction area and a set of attribute parameters of the construction area are determined; based on the set of attribute parameters of the construction area, a construction layer is constructed, and the construction layer is pushed to the user terminal.

[0034] Currently, the updating and maintenance of high-precision maps mainly rely on periodic road data collection by specialized vehicles. This centralized collection model is costly, has a long update cycle, and struggles to capture temporary or sudden road construction events in a timely manner, resulting in map data lag. With the widespread adoption of intelligent connected vehicles, utilizing the massive, real-time crowdsourced perception data they generate to dynamically update maps has become a promising technological approach. However, crowdsourced data is inherently fragmented, asynchronous, and heterogeneous, posing significant challenges to directly identifying construction areas from it: sporadically detected construction elements (such as traffic cones) lack contextual relevance, easily leading to false alarms; local geometric changes in roads (such as lane narrowing) are difficult to directly link to construction activities; more importantly, there is a lack of effective methods to fuse these isolated pieces of evidence to accurately delineate the overall scope and boundaries of construction areas. Therefore, how to achieve accurate real-time identification of road construction areas based on fragmented crowdsourced vehicle perception data is a pressing technical problem that needs to be solved.

[0035] This solution effectively eliminates misjudgments based on single features through a dual verification mechanism of spatial aggregation of construction elements and sudden changes in road width, thereby improving the accuracy of construction area identification. By generating dynamic layers through standardized attribute parameter sets, it achieves structured description and efficient transmission of construction information, providing end users with real-time and reliable road construction situation awareness capabilities.

[0036] It should be noted that the executing entity of this invention can be a road construction area map layer construction device, or a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or a thermal management device capable of realizing the above functions of a road construction area map layer construction device, etc. This embodiment does not specifically limit it in this way. The following uses a road construction area map layer construction device as the executing entity as an example to describe this embodiment and the following embodiments.

[0037] Based on this, this application provides a method for constructing a map layer for a road construction area, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the method for constructing a map layer for a road construction area according to this application.

[0038] In this embodiment, the method for constructing a road construction area map layer includes steps S10 to S50: Step S10: Obtain a set of road features based on the vehicle-side road environment perception data.

[0039] It should be noted that the vehicle-side road environment perception data in this step refers to the standardized data set obtained after preprocessing processes such as data cleaning, coordinate system one and time sequence alignment. It contains the structured environmental information perceived by the vehicle through various sensors during driving. The road feature set refers to the integrated result of various road environment elements extracted from the above perception data for subsequent analysis. The road feature set constitutes the basic data layer for identifying construction areas.

[0040] Understandably, obtaining a road feature set is essentially a transformation and organization process from raw perceived information to structured feature information. Specifically, firstly, a target recognition algorithm needs to be executed on the standardized vehicle-side road environment perception data to identify various road elements, such as construction-related elements like traffic cones, construction signs, and temporary barriers, as well as basic road infrastructure like lane lines, curbs, and traffic lights. The output of this step is the road element recognition result. Next, to ensure data traceability and the accuracy of subsequent analysis, each identified road element instance needs to be bound to the identity information of its source data packet, thus forming road element instance data with traceability information. Then, all these instance data are categorized according to their physical attributes and functions, such as into subsets of construction elements, lane line elements, and road boundary elements. Finally, by summarizing these categorized element subsets, a well-structured and clearly categorized road feature set is formed.

[0041] like Figure 2 As shown, Figure 2This is a schematic diagram of the raw sensing data preprocessing flow of the road construction area map layer construction method of the present invention.

[0042] In one feasible implementation, before obtaining the road feature set based on the vehicle-side road environment perception data, the method further includes steps A10 to A30: Step A10: Clean the received raw vehicle-side asynchronous data packets to obtain initial road perception data.

[0043] It should be noted that the original vehicle-side asynchronous data packets refer to the data units that are independently collected and uploaded by the sensors (such as cameras, LiDAR, GPS, etc.) on the vehicle during the driving process. These data packets are not continuous in time, are discretely distributed in space, and may contain various noises, outliers, and format differences.

[0044] Understandably, data cleaning typically involves several key aspects. First, invalid data is removed, such as empty data packets caused by brief sensor malfunctions or data frames with completely lost signals. Second, outliers are detected and processed, such as drift points where GPS coordinates significantly deviate from the road network, or noise points in the LiDAR point cloud caused by reflections with extremely high or low distances. Through this series of processes, the originally disordered raw data stream is refined into initial road perception data of controllable quality.

[0045] Step A20: Transform the coordinates of each sensor coordinate system of the initial road perception data to a unified global coordinate system to obtain road perception data after global coordinate alignment.

[0046] Step A30: Perform time-series alignment processing on the road perception data after global coordinate alignment to obtain the vehicle-side road environment perception data.

[0047] Understandably, because each vehicle's sensors, such as cameras or lidar, have their own independent local coordinate system, the coordinate representations of the same object perceived by different vehicles, or even by the same vehicle at different times, can vary. This step transforms the coordinates of all data points into a unified global coordinate system by applying a series of coordinate transformation matrices, such as a latitude, longitude, and altitude coordinate system referenced to the Earth, or a planar coordinate system for a specific region. This allows sensing data from different perspectives and locations to be overlaid and compared on the same "map." Based on this, these spatially aligned, discrete data points on the time axis are aligned as much as possible to multiple shared timestamps or the same short time window through interpolation and correlation methods. This allows previously scattered and isolated instantaneous observations to be integrated into a relatively complete description of a continuous space within a specific time period.

[0048] It should be understood that after the aforementioned steps are completed, the data is formatted and standardized to ensure that data from different vehicle models and batches are consistent in coordinate system, timestamp format, and data accuracy.

[0049] In one feasible implementation, the original vehicle-side asynchronous data packet includes at least local map data, original vehicle perception data, and vehicle status data.

[0050] It should be noted that the raw vehicle perception data includes road images, point cloud data, obstacle results, and road semantic recognition results; vehicle status data includes vehicle identity information, vehicle positioning coordinates, vehicle heading angle, vehicle speed, and upload timestamp.

[0051] Understandably, the raw vehicle perception data constitutes the vehicle's real-time observation and record of its environment. Road images refer to visual images captured by onboard cameras, intuitively presenting the texture, color, and various objects on the road surface; point cloud data is usually generated by sensors such as LiDAR, and this data is used to depict the three-dimensional geometric contours of the surrounding environment; obstacle results are structured information output by the onboard computing unit after preliminary analysis of the above raw data, identifying the location, size, and type of dynamic or static obstacles such as vehicles and pedestrians; road semantic recognition results are the result of further in-depth understanding and analysis of the aforementioned data, provided that the vehicle's computing power is sufficient, such as the identification and classification of road infrastructure elements such as lane lines, traffic signs, and curbs.

[0052] It should be understood that vehicle status data provides spatiotemporal context and identification labels for all perceived data. Vehicle identification information is a unique identifier, ensuring data traceability to a specific source; vehicle positioning coordinates and vehicle heading angle together establish the vehicle's precise pose in global space at the moment of data acquisition, which is the basis for transforming all local perceived data into a unified coordinate system; vehicle speed reflects the vehicle's motion state, helping to determine the dynamic changes in the vehicle scene; and upload timestamps record the time when data packets arrive at the cloud, serving as the basis for time-series alignment and correlation of multi-vehicle data. For example, knowing the precise location, orientation, and data acquisition time of each vehicle allows the cloud system to determine whether images or point clouds from different vehicles depict the same road segment, thus enabling effective fusion analysis.

[0053] In a preferred embodiment, the local map data includes lightweight map fragments obtained by reconstructing a local environment that meets preset local mapping conditions on the vehicle side.

[0054] It should be noted that local map data refers to a dynamic map representing the vehicle's surrounding environment, which is perceived and constructed in real time during vehicle operation. Preset local mapping conditions refer to specific logic that triggers map construction. For example, this computationally intensive process is initiated locally on the vehicle when the vehicle first enters a road segment not previously used by other vehicles or when the sensors detect significant changes in the surrounding environment relative to existing map data, indicating a need to refresh the existing map. This process requires fusing data from multiple vehicle sensors and using simultaneous localization and mapping (SLAM) or other environmental modeling algorithms to generate map fragments from raw observations. This process does not aim to generate high-precision local maps but focuses on the necessary features of the current local area, resulting in lightweight map fragments with small data volumes, facilitating real-time processing and rapid uploading.

[0055] Step S20: Based on the road feature set, obtain the spatial clustering area of ​​construction elements.

[0056] It should be noted that this step aims to transform the previously identified road feature information into a construction area with clear geographical significance. The input is a set of road features, which refers to various elements extracted from the raw data that characterize the existence of construction activities, such as the location of construction vehicles, the spatial distribution of isolation facilities, or the coordinates of specific signs. The output is a spatially concentrated area of ​​construction elements, which refers to a spatially continuous area with densely distributed internal elements, identified as a unified construction activity zone.

[0057] Understandably, in the specific implementation, the elements in the road feature set are first grouped through cluster analysis to form several preliminary candidate clustering regions, each representing a potential construction core area. Subsequently, the minimum bounding geometry of these point sets is calculated and the boundaries are further refined to generate a precise contour for each candidate region. For example, the system may identify multiple traffic cones and construction vehicle coordinates scattered along a main road. Through this step, these points are ultimately aggregated and delineated into a polygonal area that completely covers the maintenance work area of ​​that road section; this polygonal area is the spatial clustering region of the construction elements.

[0058] It should be understood that the regional boundary data generated in this step can be seamlessly integrated into a geographic information system or traffic management platform for the automated issuance of construction reminders, planning of vehicle detour routes, or implementation of regional traffic restrictions, thereby significantly improving the intelligence level and response efficiency of road construction management.

[0059] Step S30: Based on the road reference data and the road feature set, obtain the road width abrupt change region.

[0060] It should be noted that by comparing and analyzing real-time perceived road features with pre-set road baseline data, continuous areas in the road space where the width changes significantly can be located. Here, road baseline data refers to a pre-stored data source reflecting the standard geometric information of the road under normal or design conditions, such as the number of lanes and design width contained in a high-precision map. The road feature set contains a summary of various road element information collected and identified in real time by the vehicle-mounted sensing devices, including dynamic and static targets such as lane lines, curbs, vehicles, and pedestrians. The core task of this step is to filter out key features that can accurately define road boundaries from this complex real-time information, quantitatively compare them with baseline values, and ultimately delineate abnormal areas.

[0061] Understandably, the implementation process first requires extracting the most representative lane line features and road edge features from the road feature set. These two types of features are clearly defined by traffic rules and are physically relatively stable road boundary markers, effectively eliminating dynamic interference from temporarily parked vehicles or pedestrians, thus obtaining accurate road width perception data reflecting the actual passable space. Subsequently, the system queries the standard width information for that location in the road reference database based on the geographic coordinates corresponding to the perception data. Next, by calculating the rate of change between the perceived width and the reference width point by point and comparing it with a preset reasonableness threshold, data points with excessive changes are filtered out.

[0062] It should be understood that genuine road anomalies (such as road closures due to construction or geological hazards) are usually continuous. For example, when a lane on a road is occupied for an extended period due to pipeline maintenance, the width change rate calculated from multiple monitoring points along that section will consistently exceed a threshold. Clustering can then clearly delineate the start and end points of this occupied area. This method effectively avoids misjudgments caused by instantaneous errors in individual sensors or temporary obstacles, ensuring that the identified abrupt change areas are real and reliable.

[0063] Step S40: Determine the construction area and the set of attribute parameters of the construction area based on the spatial aggregation area of ​​the construction elements and the abrupt change area of ​​the road width.

[0064] It should be noted that the spatial clustering area of ​​construction elements refers to the spatial range formed by the dense distribution of typical construction-related objects such as traffic cones and construction barriers in image or point cloud data; the abrupt change area of ​​road width refers to the area where the originally continuous lane lines or boundaries of the road show obvious inward contraction or interruption. This step uses spatial analysis techniques to determine whether there is a significant geographical overlap or spatial adjacency between these two types of areas. When it is confirmed that they both point to the same geographical location, it can be determined that there is a construction activity there, and the spatial boundary of the construction area can be delineated accordingly. The attribute parameter set refers to the dataset used to describe a series of characteristics of the construction area, and its construction process involves extracting various quantitative or categorical parameters from the above-mentioned areas.

[0065] Understandably, if a large number of traffic cones are found arranged along the shoulder of a road (construction element clustering), and the lane markings at that location show the outermost lane disappearing or changing position (abrupt change in road width), then it can be confirmed with high confidence that lane-occupying construction is underway. Based on this identified area, the system will further extract its attributes from the raw data. These attributes include the geographical location and geometry of the construction, the road area affected by the construction, the nature of the construction inferred from the type and distribution of the clustered elements (such as road repair, pipeline laying), and a confidence parameter representing the reliability of the judgment. This set of attributes constitutes a set of all valid information about the construction area.

[0066] It should be understood that this multi-source information cross-validation method significantly improves the accuracy of construction area determination and effectively reduces false alarms caused by misjudgments based on single features, such as clusters caused by temporarily parked vehicles. The final output of the construction area and its attribute parameter set provides a structured digital representation of a temporary, dynamic road event, enabling it to be accurately stored, processed, and transmitted by computer systems.

[0067] Step S50: Construct a construction layer based on the set of attribute parameters of the construction area, and push the construction layer to the user terminal.

[0068] It should be noted that the core task of this step is to transform the descriptive data (i.e., the set of attribute parameters) of each construction area obtained in the preceding steps into a structured information layer that can be recognized and displayed by the map system, and finally distribute it to end users who need this information. The construction layer refers to an independent data layer specifically used to overlay and display construction-related information on a geographic information system or electronic map. It is composed of data objects from multiple construction areas organized according to a unified standard. The process of constructing this layer first requires encapsulating the attribute parameters of each construction area, such as location range, construction type, and estimated duration, according to a preset standardized format to form a dynamic data object containing complete information. Then, these independent objects representing different construction points are collected and integrated to ultimately form a comprehensive layer that fully reflects the construction status of all current road network construction. Pushing to the user terminal refers to sending the constructed construction layer data to terminals such as vehicle navigation systems or mobile smart devices via a wireless communication network.

[0069] Understandably, constructing a construction layer is a crucial step in transferring information from the data processing backend to practical applications. Since the attribute parameter set itself is a raw list of data, the construction layer endows this data with spatial attributes and visualization capabilities. After constructing the construction layer, the system can encapsulate attribute information such as "located on a certain avenue, heading north, undergoing road repairs, and expected to last three days" into an object conforming to map data specifications. When numerous such objects are aggregated, the map on the terminal device can clearly display all construction sections to the driver through highlights, icons, or text labels.

[0070] It should be understood that the push process is not a simple one-time data transmission. It generally follows certain specific triggering mechanisms. For example, when the system detects that a user's vehicle is about to enter the influence range of a construction site, the relevant layer information is accurately pushed to achieve the early warning function and avoid information overload.

[0071] This embodiment first cleans, unifies coordinates, and aligns time for various raw perception data from vehicles, forming standardized road environment perception data. From this data, a set of road features containing various road elements is extracted. Based on this set, on the one hand, spatial clustering areas of construction elements are delineated by identifying the aggregation of construction-related elements; on the other hand, areas of abrupt changes in road width are located by comparing real-time perceived road width with baseline data. These two spatially related areas are then fused to accurately determine the extent of the construction area and extract its attribute parameters. Finally, these parameters are encapsulated into a standardized construction layer, and based on the real-time distance relationship between the user terminal and the construction area, accurate early warning information is pushed out.

[0072] In summary, this technical solution combines two key pieces of evidence—the aggregation of construction elements and abrupt changes in road geometry—to determine the area, forming a dual verification mechanism that significantly improves the accuracy and reliability of construction area identification and reduces the risk of misjudgment based on single features. Secondly, by integrating scattered construction information into unified, spatially attributed layer data, it provides an intuitive and comprehensive dynamic information view for route planning and traffic management, effectively ensuring the driving safety of core users.

[0073] Based on the first embodiment of this application, in the second embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 3 In the method for constructing a map layer for a road construction area, step S10 includes steps S101 to S104: Step S101: Perform target recognition on the vehicle-side road environment perception data to obtain road element recognition results.

[0074] It should be noted that the purpose of this step is to use a pre-trained computer vision or deep learning model to automatically analyze the vehicle-side road environment perception data obtained in the previous steps, detect targets of a preset type, and classify them. The road element recognition result is the output of this step, which specifically includes a series of structured data with semantic labels and location information, such as identifying a traffic cone within a rectangle in an image, or determining that a set of points in a point cloud belongs to a lane line.

[0075] Understandably, this step relies on mature object detection and semantic segmentation algorithms. For image data in vehicle-side road environment perception data, the model scans the entire image, locates the bounding boxes of targets such as vehicles, pedestrians, traffic signs, and construction barriers, and provides their category and confidence level. For point cloud data, the algorithm analyzes the distribution of points in three-dimensional space, classifying points into different categories such as ground, buildings, trees, streetlights, and traffic cones. For example, when a construction vehicle appears within the perception range, the algorithm can not only identify that it is a truck, but also further identify the excavator loaded on the vehicle and the warning sign placed next to it, outputting these elements as road element recognition results.

[0076] Step S102: Associate the road element recognition result with the perception data source identification information to obtain road element instance data.

[0077] It should be noted that the sensing data source identification information refers to the metadata that identifies the source of the data. It includes at least the vehicle identification number that collected the data, the precise timestamp of the data collection time, and the vehicle's global positioning coordinates and heading angle at that time. Combining these two information allows for the establishment of a clear correspondence in the data structure, ensuring that each identified road element instance clearly records which vehicle observed it, when, and where.

[0078] It is understandable that the road element instance data obtained in this step is a complete data entity. For example, a road closure warning sign, after association, will not only contain the category "warning sign" and its pixel coordinates in the image in its data record, but will also be associated with the license plate number of the vehicle collecting the data, the collection time, and the latitude and longitude of the warning sign in the actual geographic coordinate system, converted from vehicle positioning data. This allows the same warning sign to be identified and recorded multiple times by different passing vehicles, forming multiple observation instances with a time series.

[0079] It should be understood that this step lays a solid foundation for subsequent data aggregation, redundancy removal, change detection, and long-term trajectory tracking. By associating source information, the system can determine whether identification results from different vehicles and at different times point to the same physical entity in the real world, thereby avoiding duplicate counting and analyzing the changes in the entity's state over time. This step greatly enhances the traceability and value of the data, transforming a single, instantaneous observation into a data asset that can be continuously managed and analyzed.

[0080] Step S103: Divide the road element instance data according to element type to obtain element subsets of each feature category.

[0081] It should be noted that the purpose of this step is to categorize and archive the mixed instance data according to semantics. Since road element instance data is a diverse collection of road objects already associated with source information, it may include many different types of entities such as traffic signs, traffic cones, construction vehicles, and temporary barriers. In practice, the objects can be categorized based on their function and form within the road environment. For example, all instances identified as "construction warning signs" can be grouped into one category, and all instances of "roadblocks" into another. That is, based on the category label recorded in each instance data, it is assigned to the corresponding group.

[0082] Understandably, different categories of road elements play different roles and have different spatial distribution patterns in construction area identification. Through pre-classification, the system can apply the most suitable analysis algorithm to different feature categories subsequently. For example, density clustering can be used for point-distributed traffic cones, while contour extraction algorithms may be needed for linearly distributed temporary barriers.

[0083] Step S104: Summarize the element subsets of each feature category to obtain the road feature set.

[0084] It should be understood that the aggregated data will form a total set of road features. While elements within this set can still be distinguished by labels, it as a whole represents the complete picture of all relevant features within the currently analyzed road area. This ensures that in subsequent analyses, the algorithm can simultaneously consider all types of features and their spatial relationships.

[0085] This embodiment constructs a road feature set containing complete spatiotemporal context information by sequentially performing target identification, data source association, element type classification, and final aggregation and integration on the perceived data, providing a structured and traceable data foundation for accurately determining the construction area.

[0086] In summary, this scheme integrates discrete identification results from multiple vehicles and time periods into a feature set with clear sources and classifications through a systematic pre-processing workflow. This allows subsequent analysis to comprehensively utilize the spatiotemporal information and distribution patterns of various elements. Therefore, compared to identification results that rely on a single data source or are not integrated, this method can identify construction areas more comprehensively and reliably, effectively improving the accuracy and robustness of identification.

[0087] Based on the first embodiment of this application, in the third embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 4 In the method for constructing a map layer for a road construction area, step S20 includes steps S201 to S203: Step S201: Based on the road feature set, extract the feature points and their coordinates that are identified as construction-related elements to obtain the construction element set.

[0088] It should be noted that the road feature set refers to a comprehensive dataset containing all identified and classified road element instances. Construction-related elements refer to object types in the road environment that typically indicate construction, maintenance, or temporary traffic control activities, such as construction warning signs, traffic cones, guardrails, construction vehicles, and construction workers. Feature points and their coordinates refer to the core spatial positioning information contained in each selected construction element instance, which can be the pixel position in the image coordinate system or the latitude and longitude coordinates in the transformed geographic coordinate system.

[0089] Understandably, the system iterates through every element instance in the road feature set, checking if its type label matches a predefined list of construction-related categories. For example, the system will filter out all instances labeled "road cone" and "construction guardrail," but exclude instances labeled "ordinary private car" or "street tree." For each matching instance, the system reads and records its location information. The resulting set of construction elements is a finer-grained subset of data specifically composed of these construction-related element instances and their spatial coordinates.

[0090] Step S202: Based on the density clustering strategy, spatial clustering is performed on the construction elements in the set of construction elements to obtain candidate clustering regions.

[0091] It should be noted that density clustering is a machine learning method based on the spatial distribution density of data points. Its core idea is to identify clusters in the data space where the point density is higher than that of the surrounding areas. In practice, by calculating the distance relationship between elements, elements that are close in location are automatically grouped into the same group.

[0092] Understandably, candidate clusters are the output of the clustering process, referring to one or more potential areas formed by the close aggregation of multiple construction elements in physical space. For example, at a road maintenance site, numerous traffic cones and warning signs are scattered around the work area, forming a high-density area, while a few scattered warning signs further away have a lower density. Density clustering algorithms can automatically discover these naturally formed clusters and identify high-density areas as independent candidate clusters, while filtering out construction elements that may be misidentified or caused by non-construction reasons.

[0093] It should be understood that each candidate region represents a physical space where construction activities may occur. This provides a foundation for further verification and precise delineation of construction area boundaries. For example, the algorithm might identify densely packed facilities at the construction entrance of a road segment as one candidate region, while identifying scattered directional signs in the middle of the road segment as another, thus helping the system distinguish between different functional zones such as the main work area and the traffic safety guidance area. This method does not rely on a preset number of clusters and can adaptively discover clusters of arbitrary shapes, thus better reflecting the complex layout of actual construction scenarios.

[0094] In one feasible implementation, the density-based clustering strategy for spatially clustering construction elements in the construction element set to obtain candidate clustering regions includes: performing cluster analysis on the coordinate points of each element in the construction element set based on preset neighborhood radius parameters and cluster point number parameters to obtain initial clustering results; classifying density-connected coordinate points in the initial clustering results into the same cluster, and marking coordinate points that cannot be assigned to any cluster as noise points to obtain a set of classified points; and using the set of coordinate points of each cluster formed by the set of classified points as the candidate clustering region.

[0095] It should be noted that this embodiment provides a classic density clustering process, the core of which is to separate spatially densely distributed construction elements from sparse noise points and form clusters.

[0096] Understandably, the preset neighborhood radius parameter refers to the distance threshold used to measure whether two points are "nearby," while the cluster point count parameter refers to the minimum number of points required within the neighborhood radius. Only areas that meet this requirement will be identified as the core of a dense region. Clustering analysis of element coordinate points refers to the process by which the algorithm traverses each point and determines whether it belongs to a core point, a boundary point, or a noise point based on the above two parameters.

[0097] It should be understood that the initial clustering result is an intermediate state, identifying the preliminary category attributes of each point. Assigning points to the same cluster means grouping all points that can be connected by density relationships into an independent set, thus forming a candidate clustering region. Density connectivity refers to a set of points connected by a series of directly or indirectly density-reachable core points. Noise points, on the other hand, refer to isolated points whose surrounding point density is too low to belong to any cluster, such as a single misidentified object or scattered facilities far from the main construction area; these will be excluded from the candidate region.

[0098] It is understood that this embodiment effectively overcomes the dependence of traditional methods on the shape and number of preset areas by automatically identifying and integrating spatially closely related elements, transforming discrete observation points into spatially meaningful clusters, and finally obtaining a candidate cluster area that accurately yields a quantifiable distribution area of ​​construction activities.

[0099] Step S203: Perform boundary generation processing on the candidate clustering region to obtain the spatial clustering region of the construction elements.

[0100] It should be noted that the candidate clustering area refers to the core area identified by density clustering, which is formed by the close spatial aggregation of construction elements, but its boundaries are not yet clear. Therefore, this step needs to delineate a closed and continuous external outline for these core areas, specifically a polygonal area with clear geometric boundaries, which visually defines the actual ground area occupied by the construction facilities.

[0101] Understandably, generating clearly defined spatial clusters of construction elements has significant practical application value. This information layer with geometric boundaries transforms the identification of construction areas from an abstract set of points into concrete objects that can be overlaid on electronic maps for spatial analysis and decision-making. For example, autonomous vehicles can plan detours based on these polygonal boundaries, and traffic management systems can issue precise warnings about restricted areas to vehicles behind them. Therefore, this step is crucial for the transformation of structured geographic information in downstream tasks and is a key link in generating practical benefits from the entire method.

[0102] like Figure 5 As shown, Figure 5 This is a schematic diagram of the construction element aggregation process in the road construction area map layer construction method of the present invention.

[0103] In one feasible implementation, step S203 includes steps B10 to B30: Step B10: Based on the set of coordinate points contained in the candidate cluster area, calculate the minimum bounding geometry of the set of coordinate points to obtain the initial geographical range.

[0104] It should be noted that this embodiment details the specific process for generating clear geographical boundaries for candidate cluster areas. Calculating the minimum bounding geometry refers to finding a regular geometric shape, such as a rectangle or circle, that contains all discrete points from a set of points and has the smallest area. The purpose of this step is to initially define the approximate coverage area of ​​the construction element cluster with a concise outline. The resulting initial geographical range provides a general description of the spatial occupancy of the cluster area, offering a basic framework for the subsequent generation of fine-grained boundaries.

[0105] Step B20: Generate the regional boundary corresponding to the candidate clustering area based on the initial geographical range.

[0106] Understandably, the initial geographical area may be relatively regular and cannot perfectly match the irregular contours of the actual distribution of construction elements. Therefore, it is necessary to further transform the initial regular shape into a polygonal boundary that better matches the actual spatial distribution of the point clusters, even if it is irregular. For example, for a construction area set along a curved road, its minimum bounding rectangle may contain a large number of non-construction areas, while a boundary generation algorithm can form a continuous boundary line that closely follows the outermost construction elements and is consistent with the road direction.

[0107] Step B30: Based on the area boundary and the geographic coordinates of the area boundary, obtain the spatial clustering area of ​​the construction elements.

[0108] Understandably, a spatially clustered area of ​​construction elements is a complete geographic entity containing clearly defined geometric boundaries and its geographic coordinates. This entity can be accurately located on a map and correlated with the real world using its geographic coordinates. For example, the boundary of a generated polygonal region can clearly indicate the closed construction area of ​​a certain section of a highway. This information can be directly used for navigation route replanning or traffic control, realizing the final transformation from discrete point cloud data to structured spatial information with practical application value.

[0109] This embodiment selects construction-related elements from comprehensive data containing various road elements and obtains their coordinates to form a set of construction elements. Then, it applies a density clustering algorithm to analyze this set, automatically grouping spatially adjacent construction elements into several candidate clustering regions. Finally, it calculates the minimum bounding graph of these regions and optimizes their boundaries to generate spatial clustering regions of construction elements with accurate geographic coordinates.

[0110] In summary, this embodiment, through a density clustering strategy, can adaptively discover clusters of construction elements in the original data, thereby more accurately reflecting the actual spatial distribution of construction activities. By transforming discrete construction element coordinates into clear geographical boundaries, the identification results can be directly applied to electronic maps and traffic management systems, providing structured and operable spatial information support for autonomous driving path planning and traffic control, and improving the automation and practicality of construction area identification.

[0111] Based on the first embodiment of this application, in the fourth embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 6 In the method for constructing a map layer for a road construction area, step S30 includes steps S301 to S303: Step S301: Obtain road width perception data based on the road feature set.

[0112] It should be noted that this step aims to transform a set of road features containing multiple elements into road width perception data specifically used to measure the spatial scale of roads. Road width perception data refers to the raw data summary consisting of the identification results of various road elements such as vehicles, pedestrians, traffic signs, lane lines, and curbs.

[0113] like Figure 7 As shown, Figure 7 This is a schematic diagram illustrating the road width calculation process of the road construction area map layer construction method of the present invention.

[0114] In one feasible implementation, step S301 includes steps C10 to C40: Step C10: Extract lane line features and road edge features from the road feature set.

[0115] It should be noted that the core elements used to define the road space in this step are lane lines and road edges.

[0116] Step C20: Based on the lane line features and road edge features, obtain the geographical coordinates of the corresponding road boundary.

[0117] Understandably, both types of data exhibit significant regularity and stability in the road environment, most clearly and consistently defining the physical boundaries of vehicular traffic areas. However, since the road feature set may include dynamic or temporary objects such as vehicles, pedestrians, and temporary obstacles, the appearance of these elements is accidental and time-sensitive. Using them as the boundary basis would lead to drastic and frequent meaningless fluctuations in perceived road width. Conversely, lane lines are clearly defined driving guide lines according to traffic rules, while road edges (such as curbs and roadbed boundaries) are permanent physical separations between the road and its surrounding environment (such as sidewalks and green belts). These two types of features together constitute the "skeleton" of the road, providing a reliable and unchanging reference benchmark for accurately and consistently measuring width.

[0118] Step C30: Calculate the perceived width value based on the geographical coordinates of the road boundary.

[0119] Step C40: Obtain the road width sensing data based on the perceived width value and its corresponding geographical coordinates.

[0120] Understandably, based on the aforementioned data, geometric calculations can yield a numerical value that characterizes the width of the road at that location—the perceived width value. For example, if the left and right boundaries are two parallel lane lines, the system can calculate the vertical distance between corresponding points on the two lines as the width value; if the boundary is an irregular curb, feature points can be selected to calculate the Euclidean distance between the lines connecting them. This process realizes the conversion from spatial positioning information to physical measurement.

[0121] It should be understood that since the perceived width value is just an isolated numerical value, it only has meaning when bound to its specific location information on the road to be practical. Road width perception data is a structured data set composed of geographic location coordinates and corresponding perceived width values. Through the above process, a width profile map continuously distributed along the road space can be generated, providing a standardized and easy-to-process data format for the next step of comparison with benchmark data.

[0122] Step S302: Query the road reference data based on the geographical location coordinates in the road width sensing data to obtain the corresponding road reference width data.

[0123] It should be noted that road baseline data refers to pre-stored databases or high-precision map data containing standard road geometry information (such as the number of lanes, design width, etc.). The query process involves using the geographic coordinates in the sensing data as an index to retrieve the road width information that the location should have under normal or design conditions from the baseline database; that is, the road baseline width data. This step is a prerequisite for comparing the "current state" with the "desired state."

[0124] Understandably, the essence of this query operation is to correlate real-time, dynamically sensed information with static, historical baseline data. Road baseline width data represents the expected width of a road under normal conditions, derived from road design drawings, verified high-precision maps, or standard values ​​obtained through long-term monitoring and statistics. For example, when the sensing system locates a specific latitude and longitude of a major urban road, a query can retrieve the standard three-lane total width data for that road segment. This baseline value serves as an objective benchmark for judging whether the current sensed width is normal.

[0125] Step S303: Based on the road width sensing data and the road reference width data, obtain the road width abrupt change area.

[0126] It should be noted that abrupt changes in road width refer to continuous road segments where the real-time perceived road width data differs significantly from pre-stored baseline road width data. This step identifies the starting and ending points of these abrupt changes by comparing the degree of difference between the two sets of data, thereby delineating the scope of the abnormal area.

[0127] Understandably, the system compares the perceived width value with the baseline width value at each location or segment along the road trajectory. When the difference exceeds a preset reasonable tolerance range, the point is marked as a potential anomaly. Subsequently, the algorithm aggregates spatially adjacent potential anomalies to form continuous abnormal road segments. For example, if the system detects that the baseline width of a road segment should be three lanes, but multiple consecutive sampling points show only two lanes, the system will classify this road segment as a region of abrupt width change caused by construction.

[0128] In one embodiment, obtaining the road width abrupt change region based on the road width sensing data and the road reference width data includes: calculating the road width change rate of each data point based on the sensing width value in the road width sensing data and the reference width value in the road reference width data; comparing the road width change rate with a preset change threshold to determine the data points where road width abrupt changes occur; and performing cluster analysis on the data points where road width abrupt changes occur according to their geographic coordinates to obtain the road width abrupt change region.

[0129] It should be noted that calculating the road width change rate refers to comparing the perceived width value at the same geographical location with a reference width value. The degree of deviation from the normal state is typically quantified as the ratio of the difference to the reference value. A preset change threshold defines what level of change is considered a significant and noteworthy abrupt change. The threshold comparison process effectively filters out minor fluctuations such as measurement noise, ensuring the reliability of the detection results.

[0130] Understandably, since road anomalies are often not isolated points but rather extend over a distance, identifying only a single abrupt change is insufficient. Cluster analysis refers to grouping adjacent or nearby points into the same group based on their geographic coordinates, thus forming a continuous spatial region. For example, suppose a road is partially occupied due to construction. The width variation rate of a series of continuous monitoring points along this road segment may consistently exceed a threshold. Through clustering, these originally discrete points are integrated to collectively identify a complete construction-occupied area. It's important to note that while an isolated, transient data point exceeding the threshold might be due to momentary interference, a continuous or adjacent data point exceeding the threshold more strongly suggests a genuine abrupt change in road width. Therefore, the application logic of the change threshold aims to filter out areas exhibiting sustained abnormal characteristics in a spatial sequence, rather than transient, isolated anomalies. This ensures that the system's diagnostic results reflect substantial changes in road structure. Furthermore, the change threshold can be set in stages, such as a two-level incremental threshold, with each level using half a standard lane width. Triggering the second level would then use a full standard lane width, allowing for more precise quantification of the construction area's extent.

[0131] It should be understood that the core of this step is not only to determine "whether" a sudden change has occurred, but more importantly, to quantify "where" the change occurred and "how large its impact range is." The resulting road width change area is a geographical entity with clear spatial boundaries, providing accurate data support for subsequent route planning adjustments, traffic safety early warnings, and facility maintenance decisions.

[0132] In this embodiment, the lane lines and road edge features that can most stably represent the road boundary are first selected from a complex set of road features, and the accurate perceived width data is calculated accordingly. Then, the perceived data is compared with the preset road reference width data to quantify the degree of width change at each point. Finally, based on a preset threshold and combined with spatial clustering analysis, discrete significant outliers are aggregated into a road width mutation region with spatial continuity.

[0133] In summary, this solution utilizes stable features such as lane lines and curbs to calculate road width, effectively avoiding measurement interference caused by temporary obstacles and other factors, thus ensuring the accuracy and reliability of the perceived data. Furthermore, by comparing with baseline data and setting thresholds for spatial clustering, it can intelligently filter out instantaneous and isolated abnormal signals, accurately identifying road segments with truly continuous width changes. This processing flow results in clear boundaries and high reliability of the finally identified abrupt change areas.

[0134] Based on the first embodiment of this application, in the fifth embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 8 In the method for constructing a map layer for a road construction area, step S40 includes steps S401 to S404: Step S401: Based on the spatial intersection of the spatial clustering area of ​​construction elements and the area of ​​abrupt change in road width, a deterministic construction area set is obtained.

[0135] It should be noted that the basic principle of this step is to cross-validate based on spatial correlation. The core logic is that if a region has both the characteristics of "abnormal road width in physical space" and "clustering of construction elements in visual perception", then the possibility that the road conditions in that region have changed due to construction activities will be significantly increased.

[0136] Understandably, areas of spatial clustering of construction elements are visual cues identified from perceptual data such as images or point clouds, indicating the presence of construction-related objects or activities. Conversely, areas of abrupt changes in road width are physical spatial anomalies discovered through geometric measurements, reflecting objective changes in the road's actual traffic capacity. Performing spatial intersection operations on these two types of information essentially involves matching and coupling the visual evidence of "construction in progress" with the physical evidence of "road space being occupied."

[0137] In one embodiment, obtaining a deterministic construction area set based on the spatial intersection of the spatial clustering area of ​​construction elements and the abrupt change area of ​​road width includes: obtaining an overlapping area ratio value based on the geographical coordinate distribution of the spatial clustering area of ​​construction elements and the abrupt change area of ​​road width; determining a target area that satisfies dual characteristics based on the comparison result of the overlapping area ratio value and a preset overlap threshold; and summarizing the target areas to obtain the deterministic construction area set.

[0138] It should be noted that this embodiment describes a specific method for accurately determining construction areas by quantitatively assessing the degree of spatial overlap. The core of this method lies in introducing a quantitative indicator called the overlap area ratio. This value is used to accurately measure the percentage of overlapping area between areas where construction elements are concentrated and areas where road width changes abruptly in two-dimensional geographic space. This ratio can objectively reflect the consistency or correlation strength of the spatial distribution of these two types of characteristic areas.

[0139] Understandably, the preset overlap threshold is crucial for classification decisions. This threshold, as an adjustable parameter, represents the minimum spatial overlap standard that two types of feature regions need to achieve, filtering out accidental cases where there is partial overlap but weak correlation. The system compares the calculated overlap ratio with this preset threshold to determine whether the region simultaneously satisfies both visual element and physical deformation characteristics. Only when the ratio is greater than or equal to the threshold is the region considered a reliable target region, meaning there is a high degree of certainty that the road anomaly is caused by construction activities. For example, if a region is marked with construction elements such as cones and engineering vehicles, and more than 80% of the area within this region is also detected as having significantly narrowed road width, then the overlap ratio reaches 0.8. Based on this, if the preset overlap threshold is set to 0.7, then the calculated ratio of this region meets the condition of being greater than or equal to the threshold, and is thus judged by the system as a target region satisfying both characteristics.

[0140] It should be understood that summarizing all target areas that have passed the above dual verification constitutes the final set of deterministic construction areas.

[0141] Step S402: Select the areas in the road width abrupt change region that exclude the spatial intersection as the candidate construction area set.

[0142] It should be noted that after spatial intersection calculations, a number of areas within the original construction element spatial clusters did not overlap with any other construction element spatial clusters. These areas were categorized as candidate construction areas because they exhibited abnormal changes in road morphology, suggesting potential alterations in traffic conditions and thus construction possibilities. However, due to the lack of direct visual construction elements such as traffic cones, barriers, or construction machinery as corroborating evidence, their certainty as construction areas is low. For example, sudden changes in road width could also be caused by temporarily parked large vehicles, accident scenes, or illegally piled debris.

[0143] Step S403: Based on the spatial correlation between the road segment closure query results and the candidate construction area set, perform a second verification of the candidate construction area set to obtain the verified construction area set.

[0144] It should be noted that the road closure query results refer to external data obtained through access to official or authoritative traffic information systems regarding whether a specific road segment is under planned or temporary closure. The core purpose of this step is to introduce this independent source of external evidence to conduct secondary verification of candidate construction areas identified solely based on visual and sensor data. The basic logic is that a real, especially large-scale, road construction project is usually reflected in the traffic management department's filings or announcements. Therefore, if a candidate construction area happens to be located within an officially announced road closure area, the probability that it is caused by construction will significantly increase, thus elevating the confidence level of the area from "candidate" to "verified." This process essentially improves the accuracy and reliability of the system's judgment through cross-validation of external information.

[0145] Step S404: Merge the set of deterministic construction areas with the set of verified construction areas to determine the construction area.

[0146] It should be noted that the two categories of construction areas with different levels of confidence identified in the early stages were ultimately summarized and confirmed. The set of definitive construction areas refers to those areas directly identified as being under construction through dual verification of visual and physical characteristics; their internal chain of evidence is complete, and their confidence level is the highest. The set of verified construction areas refers to those areas that, although lacking direct visual evidence, have road physics anomalies corroborated by external, official, closed-off information; their confidence level has also been significantly improved through secondary verification.

[0147] Understandably, merging these two sets allows for the construction of a complete and hierarchical dynamic construction area map. This map includes both clearly defined construction points supported by visual evidence and high-probability construction points endorsed by authoritative information. For example, in a complex traffic network, the system might simultaneously identify a definite area under an overpass filled with barriers and machinery, and a verified construction area with only simple markings due to temporary pipeline repairs, but with an officially issued road closure notice. By merging these two sets, the system obtains a comprehensive intelligence report containing all the aforementioned key information, thus providing a more comprehensive obstacle data foundation for route planning.

[0148] like Figure 9 As shown, Figure 9 This is a schematic diagram illustrating the generation of attribute parameter sets in the method for constructing a map layer for road construction areas according to the present invention.

[0149] In one feasible implementation, after step S404, steps D10~D40 are further included: Step D10: Obtain the geographical range parameters based on the spatial outline coordinates of the construction area.

[0150] It should be noted that spatial contour coordinates refer to a series of geographic coordinate points acquired through a sensing system that can delineate the boundary of the construction area. Geographic range parameters are a set of feature values ​​extracted from these raw coordinates that can clearly define the location, shape, and scale of the construction area, such as the latitude and longitude of the center point of its circumscribed rectangle, the length of its boundary, and its area.

[0151] Understandably, geographic extent parameters help the system accurately determine the relative position of the vehicle to the construction area, thereby assessing the scope of impact. For example, a long, narrow road repair area will have a large aspect ratio in its geographic extent parameters, allowing the system to determine that it is a closed lane that needs to be detoured, rather than a localized pothole that can be avoided.

[0152] Step D20: Obtain construction type parameters based on the distribution characteristics of construction element types within the construction area.

[0153] It should be noted that the distribution characteristics of construction element types refer to the types, quantities, and spatial arrangement patterns of various construction-related objects (such as cones, fences, excavators, and warning signs) identified within a defined construction area. Construction type parameters are parameters derived from these characteristics and used to describe the category of construction activities, such as "road paving," "pipeline repair," or "building construction."

[0154] Understandably, different types of construction impact traffic in vastly different ways and require drastically different response strategies. Analyzing the distribution characteristics of construction elements to determine the construction type enables the system to make smarter decisions. For example, identifying a large number of cones arranged linearly and guiding vehicles to merge into another lane indicates "lane closure" construction; while identifying large excavators and deep pits may indicate "underground work," which may require more careful detours to avoid the risk of vibration or landslides.

[0155] It should be understood that extracting construction type parameters enhances the system's situational awareness. It no longer simply identifies obstacles, but begins to understand the real-world events they represent, thus achieving a leap from perception to cognition. This provides a deeper basis for predicting dynamic changes in the construction area (such as work time and potential expansion directions) and selecting the optimal interaction strategy.

[0156] Step D30: Calculate the confidence parameters based on the statistical values ​​of the number of construction elements in the construction area, the spatial overlap index, and the type of data acquisition equipment.

[0157] It should be noted that the statistical value of the number of construction elements reflects the quantity of visual evidence supporting the judgment of "construction"; the spatial overlap index measures the consistency of the location of construction areas identified from different sources or with different timestamps; and the data acquisition device type takes into account the impact of the performance differences of various sensors (such as LiDAR and cameras) on the recognition quality. The confidence parameter is a quantitative value that represents the credibility of the judgment result, calculated by integrating these factors.

[0158] Understandably, when calculating confidence parameters, the type of data acquisition device can refer not only to single sensors such as LiDAR and cameras, but also to the vehicle itself as a "device" integrating perception, computing, and communication capabilities. Since different models or configurations of vehicles possess different levels of perception capabilities, an expensive luxury vehicle typically features a higher-performance sensor suite and a more powerful onboard computing unit, while the system capabilities of an economy vehicle are relatively limited.

[0159] It should be understood that high-performance systems can identify construction elements such as cones and barriers with greater accuracy and robustness at greater distances and in more severe weather conditions, and can more accurately delineate the outline of the construction area. Therefore, data from high-performance vehicles will be given higher confidence weights, and this differentiated assessment is a key foundation for achieving reliable fusion of multi-source heterogeneous data.

[0160] Step D40: Based on the geographical range parameter, construction type parameter, and confidence level parameter, obtain the set of attribute parameters for the construction area.

[0161] Understandably, based on the geographical range parameters, construction type parameters, and confidence level parameters determined in the aforementioned steps, a set of attribute parameters for describing the construction area is generated. This step aims to integrate scattered, multi-dimensional feature information into a unified, structured dataset, providing a data foundation for subsequent advanced applications such as route planning or navigation prompts.

[0162] In this embodiment, the spatial intersection of two types of internal perception features—construction element clusters and sudden changes in road width—is used to filter out construction areas with high certainty. Then, physical anomaly areas not covered by visual evidence are used as candidates, and external authoritative road closure information is introduced for secondary verification, thereby forming verified construction areas. Finally, the two types of areas are merged, and key attributes such as their geographical range, construction type, and comprehensive confidence level are further extracted to form a complete set of attribute parameters, providing structured data support for intelligent decision-making.

[0163] In summary, this solution effectively overcomes the limitations of relying on a single sensor or judgment criterion by integrating multi-dimensional evidence such as visual perception, geometric measurement, and external information, and by conducting a progressive verification process to double-check the recognition results. This significantly reduces false alarms caused by non-construction factors such as temporary road occupancy or debris accumulation. Furthermore, by analyzing the characteristics of construction elements to infer the construction type and combining this with the confidence level of the data source quality assessment results, this method achieves a leap from low-level perception to high-level scenario understanding. This provides more accurate and reliable dynamic information about construction areas for autonomous driving systems or navigation services, enabling them to make safer and more efficient path planning decisions.

[0164] Based on the first embodiment of this application, in the sixth embodiment of this application, the content that is the same as or similar to that in the first embodiment described above can be referred to the above description, and will not be repeated hereafter. Based on this, please refer to... Figure 10 In the method for constructing a map layer for a road construction area, step S50 includes steps S501 to S503: Step S501: Generate standardized description information for each construction area based on the set of attribute parameters.

[0165] It should be noted that the purpose of this step is to generate a standardized set of descriptive information based on the set of attribute parameters of the construction area obtained in the previous steps. Since the set of attribute parameters is a structured data package that integrates core features such as the geographical extent of the construction area, construction type parameters, and confidence level parameters, it is necessary to convert or encapsulate these parameter data, which are originally designed for machine processing, into a standardized and semantically clear descriptive form to facilitate their use, storage, or cross-system transmission in subsequent map layer generation stages.

[0166] Understandably, standardized descriptive information is used to ensure consistency in the understanding of construction areas across different subsystems or vehicle platforms. For example, a complex geographical range parameter containing precise boundary coordinates and area information might be translated into a standard text description such as "There is a construction section approximately 300 meters from the intersection of Road B and Road C on road segment A, running from east to west." Similarly, construction type parameters might be mapped to industry-standard terminology such as "road milling operation." This ensures that both the path planning module and the human-machine interface can understand and process construction information based on the same standard language, avoiding misunderstandings or errors caused by differences in data formats or definitions.

[0167] In one embodiment, the standardized description information includes construction type information, construction area information, confidence level information, time parameter information, and traffic behavior guidance information.

[0168] Furthermore, generating standardized description information for each construction area based on the attribute parameter set includes: obtaining construction type information based on the construction element type composition information in the attribute parameter set; obtaining construction area information based on the location range information and regional geometric feature information in the attribute parameter set; obtaining confidence information based on the construction area detection confidence parameter in the attribute parameter set; obtaining time parameter information based on the construction start timestamp and expected duration parameters in the attribute parameter set; obtaining traffic behavior guidance information based on the construction type information and the construction area information, combined with a set of traffic behavior rules; and summarizing the construction type information, construction area information, confidence information, time parameter information, and traffic behavior guidance information to obtain the standardized description information.

[0169] It should be noted that this process is an information extraction and reorganization process. First, it extracts specific information such as construction category, location range, reliability, and time element from the attribute set. Then, based on traffic rules, it combines the construction type with the regional location to derive specific traffic behavior guidance suggestions such as "suggest changing lanes in advance" or "no passage". Finally, all these processed independent pieces of information are summarized and packaged to form a complete standardized description containing all key dimensions.

[0170] Understandably, the resulting standardized descriptive information is a systematic summary of all the aforementioned information elements. It integrates the essential attributes of construction, spatial and temporal characteristics, reliability assessments, and behavioral recommendations into a structurally complete and semantically clear data entity. This achievement enables downstream applications, such as vehicle control systems or navigation platforms, to directly and accurately understand the overall situation at the construction site without parsing the underlying raw parameters, and to make safe and efficient decisions accordingly.

[0171] Step S502: Encapsulate the standardized description information into a dynamic layer data object according to a preset data encapsulation format.

[0172] It should be noted that the preset data encapsulation format refers to a predefined, unified digital envelope or packaging specification that defines which fields the data should include, the type and length of each field, and the organizational relationships between them. The dynamic layer data object, on the other hand, is the product of this encapsulation; it is a self-contained data entity carrying complete information about a specific construction area. This object not only clearly describes the construction content but also possesses good interoperability due to its standardized structure, making it easy for different map services or vehicle terminals to accurately parse.

[0173] Understandably, through encapsulation, text or key-value pair information that might have existed in a loose form is organized into an independent data object with clear boundaries and internal structure. This object can be added to an electronic map as a complete logical unit, serving as an independent, dynamically displayable and hideable information layer—a dynamic layer.

[0174] Step S503: Summarize the dynamic layer data objects of each construction area to obtain the construction layer.

[0175] It should be noted that the dynamic layer data objects generated from multiple independent construction events obtained in the aforementioned steps are integrated and merged to construct a comprehensive construction layer. This layer fully depicts the spatial distribution and details of all identified construction points within the target area, forming an overall view that can comprehensively reflect all dynamic construction conditions in the current road network.

[0176] Understandably, this aggregation process ensures that the system can view road conditions from a global perspective, rather than focusing solely on isolated construction sites. For example, a vehicle navigation system can obtain comprehensive information on all construction sites along the entire planned route at once, rather than requesting and parsing them one by one, thus significantly improving data processing efficiency.

[0177] It should be understood that this layer, as a standardized data interface, enables map rendering engines to efficiently visualize all construction zones and allows path planning algorithms to perform optimization calculations based on complete environmental constraints. By providing a unified and integrated data view, the construction layer effectively supports the collaborative perception and decision-making of temporary and frequent road events in intelligent transportation systems.

[0178] like Figure 11 As shown, Figure 11 This is a schematic diagram of the construction layer push process in the road construction area map layer construction method of the present invention.

[0179] In one feasible implementation, step S50 further includes steps E10 to E30: Step E10: Obtain the distance judgment result based on the real-time relative distance between the current location of the user terminal and the geographical area covered by the construction layer.

[0180] Step E20: When the distance judgment result meets the preset warning distance threshold condition, a data distribution instruction is generated.

[0181] Step E30: Drive the geographic message subscription system to push the construction layer to the target user terminal according to the data distribution instruction, so as to complete the push of the real-time construction layer.

[0182] It's important to note that the core of this step lies in the real-time assessment of the spatial proximity between the user terminal and potential road construction risks. The user terminal's current location is typically provided by modules such as the Global Positioning System (GPS), while the construction layer contains the geographic boundary information of all known construction areas. This step calculates the real-time relative distance between the two to obtain a quantified distance judgment result, thus providing a numerical representation of the spatial relationship between the current user and one or more nearest construction points. The preset warning distance threshold refers to a pre-set critical distance value used to trigger a warning action. This threshold can be dynamically adjusted based on factors such as road grade, speed limits, or construction type.

[0183] Understandably, when the calculated actual distance is less than or equal to the preset warning threshold, it indicates that the user is about to enter the construction impact area that needs to be alerted. At this time, the system automatically generates a data distribution instruction, the purpose of which is to notify the downstream system to start the information push process, ensuring the timeliness and accuracy of the warning, avoiding the information being pushed too early or too late, preventing information interference and eliminating the failure of the warning.

[0184] It should be understood that a geographic message subscription system is a mechanism that broadcasts messages to subscribers within a specific area based on geographic location tags. Data distribution instructions drive the system to accurately push construction information associated with the target user from the construction layer to their terminal. In this way, users can know the road conditions ahead in advance, thus completing a closed loop from construction information perception to the final provision of early warning services to users. The entire process realizes intelligent perception, real-time judgment, and on-demand distribution of dynamic road information, effectively improving driving safety and efficiency.

[0185] In this embodiment, the identified construction area attribute parameters are converted into standardized descriptive information, then encapsulated into independent dynamic layer data objects and aggregated to form a construction layer. Finally, based on the real-time distance judgment result between the user terminal and the construction layer, the accurate distribution of early warning information is triggered as needed.

[0186] In summary, this solution achieves the transformation and integration of raw data into standardized information, ensuring information standardization and interoperability. Through a distance-triggered mechanism, it enables precise, on-demand information delivery, effectively avoiding interference from irrelevant information. Integrating scattered construction information into a unified layered view provides comprehensive and real-time data support for route planning and decision-making, thereby helping to improve the response efficiency and driving safety of assisted driving or navigation systems to dynamic road events.

[0187] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the method for constructing road construction area map layers in this application. Any simple transformations based on this technical concept are within the protection scope of this application.

[0188] This application also provides a device for constructing a map layer for a road construction area. Please refer to [reference needed]. Figure 12 The road construction area map layer construction device includes: The perception feature extraction module 10 is used to obtain a set of road features based on the vehicle-side road environment perception data; The construction element identification module 20 is used to obtain the spatial aggregation area of ​​construction elements based on the road feature set; The road anomaly detection module 30 is used to obtain the road width abrupt change region based on the road reference data and the road feature set; The regional attribute determination module 40 is used to determine the construction area and the set of attribute parameters of the construction area based on the spatial aggregation area of ​​the construction elements and the abrupt change area of ​​the road width. The layer construction and push module 50 is used to construct a construction layer based on the set of attribute parameters of the construction area and push the construction layer to the user terminal.

[0189] In one embodiment, the perception feature extraction module 10 is further configured to perform data cleaning on the received original vehicle-side asynchronous data packets to obtain initial road perception data; transform the coordinates of each sensor coordinate system of the initial road perception data to a unified global coordinate system to obtain globally coordinate aligned road perception data; and perform time-series alignment processing on the globally coordinate aligned road perception data to obtain the vehicle-side road environment perception data.

[0190] In one embodiment, the perception feature extraction module 10 is further configured to perform target recognition on the vehicle-side road environment perception data to obtain road element recognition results; associate the road element recognition results with the perception data source identification information to obtain road element instance data; divide the road element instance data according to element type to obtain element subsets of each feature category; and summarize the element subsets of each feature category to obtain the road feature set.

[0191] In one embodiment, the construction element identification module 20 is further configured to extract feature points and their coordinates identified as construction-related elements based on the road feature set to obtain a construction element set; perform spatial clustering of the construction elements in the construction element set based on a density clustering strategy to obtain candidate clustering regions; and perform boundary generation processing on the candidate clustering regions to obtain the spatial clustering region of the construction elements.

[0192] In one embodiment, the construction element identification module 20 is further configured to perform cluster analysis on the coordinate points of each element in the construction element set based on preset neighborhood radius parameters and cluster point number parameters to obtain initial clustering results; divide the density-connected coordinate points in the initial clustering results into the same cluster, and mark the coordinate points that cannot be assigned to any cluster as noise points to obtain a set of classified points; and use the set of coordinate points of each cluster formed by the set of classified points as the candidate clustering region.

[0193] In one embodiment, the construction element identification module 20 is further configured to calculate the minimum bounding geometry of the set of coordinate points contained in the candidate clustering area to obtain an initial geographical range; generate a regional boundary corresponding to the candidate clustering area based on the initial geographical range; and obtain the spatial clustering area of ​​the construction elements based on the regional boundary and the geographical coordinate information of the regional boundary.

[0194] In one embodiment, the road anomaly detection module 30 is further configured to obtain road width sensing data based on the road feature set; query road reference data based on the geographical location coordinates in the road width sensing data to obtain corresponding road reference width data; and obtain the road width abrupt change area based on the road width sensing data and the road reference width data.

[0195] In one embodiment, the road anomaly detection module 30 is further configured to extract lane line features and road edge features from the road feature set; obtain the geographical coordinates of the corresponding road boundary based on the lane line features and road edge features; calculate the perceived width value based on the geographical coordinates of the road boundary; and obtain the road width perceived data based on the perceived width value and its corresponding geographical coordinates.

[0196] In one embodiment, the road anomaly detection module 30 is further configured to calculate the road width change rate of each data point based on the perceived width value in the road width perception data and the reference width value in the road reference width data; compare the road width change rate with a preset change threshold to determine the data points where road width abrupt changes occur; and perform cluster analysis on the data points where road width abrupt changes occur according to their geographical coordinates to obtain the road width abrupt change region.

[0197] In one embodiment, the region attribute determination module 40 is further configured to: obtain a deterministic construction region set based on the spatial intersection of the construction element spatial clustering region and the road width abrupt change region; exclude the spatial intersection region from the road width abrupt change region as a candidate construction region set; perform secondary verification on the candidate construction region set based on the spatial correlation between the road segment closure query result and the candidate construction region set to obtain a verified construction region set; and merge the deterministic construction region set and the verified construction region set to determine the construction region.

[0198] In one embodiment, the regional attribute determination module 40 is further configured to obtain an overlapping area ratio value based on the geographical coordinate distribution of the spatial clustering area of ​​construction elements and the abrupt change area of ​​road width; determine the target area that satisfies the dual characteristics based on the comparison result of the overlapping area ratio value and the preset overlap threshold; and summarize the target areas to obtain the deterministic construction area set.

[0199] In one embodiment, the layer construction and push module 50 is further configured to generate standardized description information corresponding to each construction area based on the attribute parameter set; encapsulate the standardized description information into dynamic layer data objects according to a preset data encapsulation format; and summarize the dynamic layer data objects of each construction area to obtain the construction layer.

[0200] In one embodiment, the layer construction and push module 50 is further configured to: obtain the construction type information based on the construction element type composition information in the attribute parameter set; obtain the construction area information based on the location range information and regional geometric feature information in the attribute parameter set; obtain the confidence information based on the construction area detection confidence parameter in the attribute parameter set; obtain the time parameter information based on the construction start timestamp and expected duration parameter in the attribute parameter set; obtain the traffic behavior guidance information based on the construction type information and the construction area information, combined with the traffic behavior rule set; and summarize the construction type information, construction area information, confidence information, time parameter information, and traffic behavior guidance information to obtain the standardized description information.

[0201] In one embodiment, the layer construction and push module 50 is further configured to obtain a distance judgment result based on the real-time relative distance between the current location of the user terminal and the geographical area covered by the construction layer; when the distance judgment result meets the preset warning distance threshold condition, generate a data distribution instruction; and drive the geographic message subscription system to push the construction layer to the target user terminal according to the data distribution instruction to complete the push of the real-time construction layer.

[0202] This embodiment identifies and extracts construction-related elements from vehicle-mounted road environment perception data uploaded by crowdsourced vehicles, and performs spatial clustering to obtain the identification and perception results of construction areas. Simultaneously, it combines road baseline data to detect areas of abrupt changes in road width, determines construction areas through the spatial correlation between the two, calculates relevant attribute parameters, and then constructs a standardized construction layer for each construction area for accurate data delivery. This solution effectively eliminates misjudgments based on single features through a dual verification mechanism of spatial clustering of construction elements and abrupt changes in road width, improving the accuracy of construction area identification. By generating dynamic layers through a standardized set of attribute parameters, it achieves structured description and efficient transmission of construction information, providing end users with real-time and reliable road construction situation awareness capabilities.

[0203] The road construction area map layer construction device provided in this application, employing the road construction area map layer construction method described in the above embodiments, can solve the technical problem of how to achieve accurate real-time identification of road construction areas based on scattered crowdsourced vehicle perception data. Compared with the prior art, the beneficial effects of the road construction area map layer construction device provided in this application are the same as those of the road construction area map layer construction method described in the above embodiments, and other technical features in the road construction area map layer construction device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0204] This application provides a road construction area map layer construction 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 to enable the at least one processor to perform the road construction area map layer construction method in the first embodiment described above.

[0205] The following is for reference. Figure 13This document illustrates a structural schematic diagram of a road construction area map layer construction device suitable for implementing embodiments of this application. The road construction area map layer construction device in this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 13 The road construction area map layer construction device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0206] like Figure 13 As shown, the road construction area map layer construction device may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM) 1004. The RAM 1004 also stores various programs and data required for the operation of the road construction area map layer construction device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the road construction area map layer building device to communicate wirelessly or wiredly with other devices to exchange data. Although the figure shows a road construction area map layer building device with various systems, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems can be implemented alternatively.

[0207] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0208] The road construction area map layer construction device provided in this application, employing the road construction area map layer construction method described in the above embodiments, can solve the technical problem of how to achieve accurate real-time identification of road construction areas based on scattered crowdsourced vehicle perception data. Compared with the prior art, the beneficial effects of the road construction area map layer construction device provided in this application are the same as those of the road construction area map layer construction method described in the above embodiments, and other technical features in this road construction area map layer construction device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0209] It should be understood that the various parts disclosed in this application can be implemented using 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 suitable manner in one or more embodiments or examples.

[0210] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0211] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the road construction area map layer construction method described in the above embodiments.

[0212] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. 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 conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0213] The aforementioned computer-readable storage medium may be included in the road construction area map layer construction device; or it may exist independently and not be assembled into the road construction area map layer construction device.

[0214] The aforementioned computer-readable storage medium carries one or more programs. When these programs are executed by a road construction area map layer construction device, the road construction area map layer construction device: obtains a set of road features based on vehicle-side road environment perception data; obtains a spatial clustering area of ​​construction elements based on the set of road features; obtains a road width abrupt change area based on road reference data and the set of road features; determines a construction area and a set of attribute parameters for the construction area based on the spatial clustering area of ​​construction elements and the road width abrupt change area; constructs a construction layer based on the set of attribute parameters for the construction area, and pushes the construction layer to a user terminal.

[0215] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0216] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0217] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0218] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described method for constructing a road construction area map layer. This solves the technical problem of how to achieve accurate real-time identification of road construction areas based on scattered crowdsourced vehicle perception data. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the road construction area map layer construction method provided in the above embodiments, and will not be elaborated upon here.

[0219] The computer program product provided in this application can solve the technical problem of constructing a map layer for road construction areas. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the road construction area map layer construction method provided in the above embodiments, and will not be repeated here.

[0220] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A method for constructing a map layer for a road construction area, characterized in that, The method for constructing the map layer of the road construction area includes: Based on the vehicle-side road environment perception data, a set of road features is obtained; Based on the set of road features, the spatial clustering area of ​​construction elements is obtained; Based on the road baseline data and the road feature set, the road width abrupt change region is obtained; Based on the spatial aggregation area of ​​the construction elements and the abrupt change area of ​​the road width, the construction area and the set of attribute parameters of the construction area are determined. Based on the set of attribute parameters of the construction area, a construction layer is constructed and the construction layer is pushed to the user terminal.

2. The method for constructing a road construction area map layer according to claim 1, characterized in that, Before obtaining the road feature set based on the vehicle-side road environment perception data, the process also includes: The received raw asynchronous data packets from the vehicle are cleaned to obtain initial road perception data; The coordinates of each sensor coordinate system in the initial road perception data are transformed to a unified global coordinate system to obtain road perception data with global coordinate alignment. The road perception data after global coordinate alignment is subjected to time-series alignment processing to obtain the vehicle-side road environment perception data.

3. The method for constructing a road construction area map layer according to claim 2, characterized in that, The original vehicle-side asynchronous data packet includes at least: Local map data, which includes lightweight map fragments obtained by reconstructing a local environment that meets preset local mapping conditions from the vehicle end; The raw vehicle perception data includes road images, point cloud data, obstacle results, and road semantic recognition results. Vehicle status data, which includes vehicle identity information, vehicle positioning coordinates, vehicle heading angle, vehicle speed, and upload timestamp.

4. The method for constructing a road construction area map layer according to claim 1, characterized in that, The road feature set obtained based on vehicle-side road environment perception data includes: Target recognition is performed on the vehicle-side road environment perception data to obtain road element recognition results; The road element recognition results are associated with the identification information of the perception data source to obtain road element instance data; The road element instance data is divided according to element type to obtain element subsets of each feature category; The road feature set is obtained by summing the subsets of elements from each feature category.

5. The method for constructing a road construction area map layer according to claim 1, characterized in that, The step of obtaining the spatial clustering area of ​​construction elements based on the road feature set includes: Based on the road feature set, feature points and their coordinates that are identified as construction-related elements are extracted to obtain the construction element set. Based on the density clustering strategy, the construction elements in the set of construction elements are spatially clustered to obtain candidate clustering regions; The candidate clustering region is subjected to boundary generation processing to obtain the spatial clustering region of the construction elements.

6. The method for constructing a road construction area map layer according to claim 5, characterized in that, The density-based clustering strategy spatially clusters the construction elements in the set of construction elements to obtain candidate clustering regions, including: Based on the preset neighborhood radius parameter and cluster point number parameter, cluster analysis is performed on the coordinate points of each element in the construction element set to obtain the initial clustering result; The coordinate points that are density-connected in the initial clustering results are assigned to the same cluster, and the coordinate points that cannot be assigned to any cluster are marked as noise points, thus obtaining the set of classification points; The set of coordinate points of each cluster formed by the classification point set is taken as the candidate clustering region.

7. The method for constructing a road construction area map layer according to claim 5, characterized in that, The process of generating boundaries for the candidate clustering regions to obtain the spatial clustering regions of construction elements includes: Based on the set of coordinate points contained in the candidate cluster area, calculate the minimum bounding geometry of the set of coordinate points to obtain the initial geographical range; Based on the initial geographical range, generate the regional boundaries corresponding to the candidate clustering areas; Based on the area boundary and its geographic coordinates, the spatial clustering area of ​​the construction elements is obtained.

8. The method for constructing a road construction area map layer according to claim 1, characterized in that, The process of obtaining abrupt changes in road width based on road reference data and the road feature set includes: Based on the aforementioned road feature set, road width perception data is obtained; Based on the geographic location coordinates in the road width sensing data, the corresponding road reference width data is obtained by querying the road reference data. Based on the road width sensing data and the road reference width data, the road width abrupt change area is obtained.

9. The method for constructing a road construction area map layer according to claim 8, characterized in that, The process of obtaining road width perception data based on the road feature set includes: Based on the road feature set, extract the lane line features and road edge features; Based on the lane line features and road edge features, the geographical coordinates of the corresponding road boundary are obtained; The perceived width value is calculated based on the geographical coordinates of the road boundary. The road width perception data is obtained based on the perceived width value and its corresponding geographical coordinates.

10. The method for constructing a road construction area map layer according to claim 8, characterized in that, The step of obtaining the road width abrupt change region based on the road width sensing data and the road reference width data includes: Based on the perceived width value in the road width perception data and the reference width value in the road reference width data, calculate the road width change rate for each data point; The road width change rate is compared with a preset change threshold to determine the data points where a sudden change in road width occurs. The data points where road width changes abruptly are clustered according to their geographical coordinates to obtain the areas where road width changes abruptly.

11. The method for constructing a road construction area map layer according to claim 1, characterized in that, The determination of the construction area based on the spatial clustering area of ​​the construction elements and the abrupt change area of ​​road width includes: The deterministic construction area set is obtained by the spatial intersection of the spatial clustering area of ​​construction elements and the area of ​​abrupt change in road width; The areas where the spatial intersection is excluded from the road width abrupt change areas are selected as the candidate construction area set; The candidate construction area set is then verified a second time based on the spatial relationship between the road closure query results and the candidate construction area set to obtain the verified construction area set. The construction area is determined by merging the set of deterministic construction areas with the set of verified construction areas.

12. The method for constructing a road construction area map layer according to claim 11, characterized in that, The deterministic construction area set is obtained by the spatial intersection of the spatial clustering area of ​​construction elements and the abrupt change area of ​​road width, including: The overlapping area ratio is obtained based on the geographical coordinate distribution of the spatial clustering area of ​​construction elements and the area of ​​abrupt change in road width; Based on the comparison between the overlapping region ratio and the preset overlap threshold, the target region that satisfies the dual characteristics is determined; The target areas are summarized to obtain the set of deterministic construction areas.

13. The method for constructing a road construction area map layer according to claim 11, characterized in that, After merging the deterministic construction area set with the verified construction area set to determine the construction area, the method further includes: Based on the spatial outline coordinates of the construction area, the geographical range parameters are obtained; Based on the distribution characteristics of construction element types within the construction area, construction type parameters are obtained; Confidence parameters are calculated based on the statistical values ​​of the number of construction elements in the construction area, the spatial overlap index, and the type of data acquisition equipment. Based on the geographical range parameter, construction type parameter, and confidence level parameter, the set of attribute parameters for the construction area is obtained.

14. The method for constructing a road construction area map layer according to claim 1, characterized in that, The step of constructing a construction layer based on the set of attribute parameters of the construction area includes: Based on the set of attribute parameters, standardized description information corresponding to each construction area is generated; The standardized description information is encapsulated into a dynamic layer data object according to a preset data encapsulation format; The construction layer is obtained by summarizing the dynamic layer data objects of each construction area.

15. The method for constructing a road construction area map layer according to claim 14, characterized in that, The standardized descriptive information includes construction type information, construction area information, confidence level information, time parameter information, and traffic behavior guidance information; The step of generating standardized description information for each construction area based on the set of attribute parameters includes: The construction type information is obtained based on the construction element type composition information in the attribute parameter set; The construction area information is obtained based on the location range information and regional geometric feature information in the attribute parameter set; The confidence information is obtained by detecting the confidence parameters of the construction area in the set of attribute parameters. The time parameter information is obtained based on the construction start timestamp and the expected duration parameter in the attribute parameter set; Based on the construction type information and the construction area information, and combined with the set of traffic behavior rules, the traffic behavior guidance information is obtained; The standardized description information is obtained by summarizing the construction type information, construction area information, confidence level information, time parameter information, and traffic behavior guidance information.

16. The method for constructing a road construction area map layer according to claim 1, characterized in that, The step of pushing the construction layer to the user terminal includes: The distance determination result is obtained based on the real-time relative distance between the current location of the user terminal and the geographical area covered by the construction layer; When the distance determination result meets the preset warning distance threshold condition, a data distribution instruction is generated; The data distribution instruction drives the geographic message subscription system to push the construction layer to the target user terminal to complete the push of the real-time construction layer.

17. A driving experience-driven map building device, characterized in that, The driving experience-driven map building device includes: The perception feature extraction module is used to obtain a set of road features based on the vehicle-side road environment perception data; The construction element identification module is used to obtain the spatial aggregation area of ​​construction elements based on the road feature set. The road anomaly detection module is used to obtain areas of sudden changes in road width based on road reference data and the road feature set; The regional attribute determination module is used to determine the construction area and the set of attribute parameters of the construction area based on the spatial aggregation area of ​​the construction elements and the abrupt change area of ​​the road width. The layer construction and push module is used to construct a construction layer based on the set of attribute parameters of the construction area, and push the construction layer to the user terminal.

18. A device for constructing a map layer for a road construction area, characterized in that, The road construction area map layer construction device includes: a memory, a processor, and a road construction area map layer construction program stored in the memory and executable on the processor, wherein the road construction area map layer construction program is configured to implement the road construction area map layer construction method as described in any one of claims 1 to 16.

19. A storage medium, characterized in that, The storage medium stores a road construction area map layer construction program, which, when executed by a processor, implements the road construction area map layer construction method as described in any one of claims 1 to 16.

20. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the road construction area map layer construction method as described in any one of claims 1 to 16.

Images