Internet of vehicles intelligent navigation system based on Internet of Things

Abstract

The invention relates to the technical field of intelligent navigation, in particular to an Internet of Vehicles intelligent navigation system based on the Internet of Things, which comprises a multi-source data fusion module, a traffic situation complementation module, a dynamic situation map construction module, a candidate path planning module, an optimal path decision module, a lane level guidance generation module and a vehicle control and feedback module. Fusing the vehicle sensing data set of the target vehicle, the roadside monitoring data set of the roadside sensing facility and the traffic data set of the cloud data source to obtain a local traffic situation; after complementing missing information, constructing a dynamic traffic situation map containing a topology layer, a state layer and an event layer; a candidate navigation path is generated, and a target path is determined through geospatial matching evaluation avoidance efficiency; analyzing the lane topology and the real-time occupation condition, generating lane-level guide information to control the steering and the speed of the vehicle, and feeding back execution state data; according to the invention, the navigation planning efficiency can be improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent navigation technology, and in particular to an Internet of Things-based vehicle-to-everything (V2X) intelligent navigation system. Background Technology

[0002] Current vehicle-to-everything (V2X) navigation technologies suffer from insufficient multi-source traffic data fusion capabilities, making it difficult to effectively integrate target vehicle's own perception, roadside monitoring, and cloud-based traffic data. This results in low spatiotemporal registration accuracy and poor resolution of conflicting traffic state information, leading to frequent information gaps or contradictions in the generated local traffic situations. Consequently, these technologies fail to accurately reflect actual traffic conditions, creating potential risks for subsequent traffic analysis and route planning.

[0003] The existing traffic situation maps lack multi-level dynamic correlations and are mostly based on road connectivity. They do not fully integrate traffic flow status and traffic anomaly area information, resulting in insufficient targeting of candidate path generation. The optimal path decision-making is difficult to effectively evaluate the efficiency of avoiding traffic anomalies. Furthermore, lane-level guidance is not dynamically adjusted in conjunction with real-time lane occupancy, and the vehicle control has a low degree of matching with actual traffic demand, resulting in poor overall navigation performance. Therefore, how to improve the efficiency of vehicle-to-everything (V2X) intelligent navigation has become an urgent problem to be solved. Summary of the Invention

[0004] To achieve the above objectives, the present invention provides an Internet of Things (IoT)-based vehicle-to-everything (V2X) intelligent navigation system, characterized in that the system includes a multi-source data fusion module, a traffic situation completion module, a dynamic situation map construction module, a candidate path planning module, an optimal path decision module, a lane-level guidance generation module, and a vehicle control and feedback module, wherein: The multi-source data fusion module is used to fuse the target vehicle's own perception dataset, the roadside monitoring dataset of the roadside perception facility, and the traffic dataset from the cloud data source to obtain the local traffic situation of the target vehicle. The traffic situation completion module is used to complete the missing traffic information in the local traffic situation to obtain the traffic situation data of the target vehicle. The dynamic situation map construction module is used to construct a dynamic traffic situation map of the target vehicle at roadside nodes, using the road connectivity relationship in the traffic situation data as the topology layer, the traffic flow status in the traffic situation data as the state layer, and the traffic anomaly area information in the traffic situation data as the event layer. The candidate path planning module is used to generate candidate navigation paths for the target vehicle based on the road connectivity and traffic conditions in the dynamic traffic situation map. The optimal path decision module is used to perform geospatial matching between the candidate navigation path and the traffic anomaly area in the dynamic traffic situation map, and evaluate the avoidance effectiveness of the candidate navigation path based on the matching result, so as to obtain the target navigation path of the target vehicle. The lane-level guidance generation module is used to parse and reconstruct the lane topology and real-time occupancy status of the target navigation path to obtain the lane-level guidance information of the target vehicle. The vehicle control and feedback module is used to control the steering and speed of the target vehicle with the lane-level guidance information, collect the actual driving trajectory and speed data of the target vehicle as execution status data, and feed the execution status data back to the roadside node.

[0005] In a preferred embodiment, when the multi-source data fusion module performs the fusion of the target vehicle's own perception dataset, the roadside monitoring dataset from the roadside perception facility, and the traffic dataset from the cloud data source to obtain the local traffic situation of the target vehicle, it is specifically used for: Receive the target vehicle's own perception dataset, the roadside monitoring dataset from the roadside perception facility, and the traffic dataset from the cloud data source; The vehicle perception dataset and the roadside monitoring dataset are spatiotemporally registered to obtain the spatiotemporally aligned dataset of the target vehicle. Identify conflicting traffic state information in the spatiotemporally aligned dataset; The conflict traffic state information is resolved to obtain a list of conflict-free local traffic states for the target vehicle. The list of conflict-free local traffic states is integrated with the traffic flow information in the traffic dataset to form the local traffic situation of the target vehicle.

[0006] In a preferred embodiment, when the traffic situation completion module completes the missing traffic information in the local traffic situation to obtain the traffic situation data of the target vehicle, it is specifically used for: A consistency check is performed on the traffic state data in the local traffic situation to obtain anomaly markers for the local traffic situation. The geographical area where the anomaly-marked data points are located is defined as the area to be supplemented for the local traffic situation; When the traffic status data in the local traffic situation has a road segment that is directly adjacent to the area to be completed, the traffic status data is used as boundary condition data. Based on the boundary condition data, state propagation calculation is performed on the traffic state of the area to be completed to obtain the estimated traffic state data of the area to be completed. Within the area to be filled, the estimated traffic state data is mapped to the local traffic situation to obtain the intermediate traffic situation data of the target vehicle; The intermediate traffic situation data is subjected to spatiotemporal consistency normalization to obtain the traffic situation data of the target vehicle.

[0007] In a preferred embodiment, the formula for calculating the estimated traffic density from the estimated traffic state data is as follows: ; In the formula, The location within the area to be completed At any moment The estimated traffic density, For the position The propagation weight coefficient, The upstream boundary traffic density in the boundary condition data. The downstream boundary traffic density in the boundary condition data. The upstream boundary traffic speed in the boundary condition data. The downstream boundary traffic speed in the boundary condition data. For the position The road network topology distance to the upstream boundary For the position The road network topology distance to the downstream boundary.

[0008] In a preferred embodiment, when the dynamic situation map construction module is executed at a roadside node, using the road connectivity relationships in the traffic situation data as the topology layer, the traffic flow status in the traffic situation data as the state layer, and the traffic anomaly area information in the traffic situation data as the event layer, to construct the dynamic traffic situation map of the target vehicle, it is specifically used for: At the roadside nodes, the road endpoints with connectivity in the traffic situation data are taken as connection points, and the road segments located between the connection points are taken as connection segments. A topology layer for the traffic situation data is constructed using the connection points as nodes and the connection segments as edges. By integrating the traffic flow status data of the same connection segment in the traffic situation data, the comprehensive traffic flow parameters of the traffic situation data are obtained; Using the comprehensive traffic flow parameters as the dynamic attributes of the corresponding edges in the topology layer, the state layer of the traffic situation data is obtained; Read the geographical range affected by the traffic anomaly area information in the traffic situation data; The geographical range is mapped to the corresponding graph elements in the topology layer, and the graph elements are marked with event attributes to obtain the event layer of the traffic situation data; The dynamic attributes of the state layer and the event attribute tags of the event layer are respectively sent to the topology layer to obtain the attribute binding graph of the target vehicle; The attribute binding graph is subjected to graph attribute situationalization to obtain the dynamic traffic situational graph of the target vehicle.

[0009] In a preferred embodiment, when the dynamic situation map construction module performs graph attribute situationalization on the attribute binding graph to obtain the dynamic traffic situation map of the target vehicle, it is specifically used for: Identify graph elements in the attribute binding graph that have missing attributes and graph elements that have conflicting event attributes, in order to generate a set of elements to be processed in the attribute binding graph; Based on the time series information of the traffic situation data, the missing attributes in the set of elements to be processed are supplemented to obtain the first processing result of the set of elements to be processed; Based on the spatiotemporal correlation of the original abnormal events in the traffic situation data, the contradictions in the event attributes in the first processing result are resolved to obtain the second processing result of the set of elements to be processed. The second processing result is fused and reconstructed to obtain the dynamic traffic situation map of the target vehicle.

[0010] In a preferred embodiment, when the candidate route planning module generates candidate navigation routes for the target vehicle based on the road connectivity and traffic conditions in the dynamic traffic situation map, it is specifically used for: Based on the dynamic traffic situation map, the initial reachable road segment of the target vehicle is constructed, starting from the current location of the target vehicle and ending at the destination location of the target vehicle. Based on the traffic status data of the initially accessible road segment, the real-time traffic impedance of the road segment edges in the initially accessible road segment is quantified; Based on the real-time traffic impedance, alternative routes are enumerated on the dynamic traffic situation map to obtain the initial path set for the target vehicle. The topological similarity and traffic impedance similarity of the paths in the initial path set are analyzed to obtain the candidate navigation path set for the target vehicle.

[0011] In a preferred embodiment, when the optimal route decision module performs geospatial matching of the candidate navigation path with traffic anomaly areas in the dynamic traffic situation map, and evaluates the avoidance effectiveness of the candidate navigation path based on the matching result to obtain the target navigation path for the target vehicle, it is specifically used for: Spatial topology analysis is performed on the candidate navigation path and the traffic anomaly area in the dynamic traffic situation map to obtain the affected path segment of the candidate navigation path; Based on the type and severity level of the affected path segments, and in combination with the length of the affected path segments, a comprehensive anomaly impact degree for implementing candidate navigation paths is obtained; Based on the comprehensive anomaly impact, the avoidance effectiveness of the candidate navigation path is determined, and the candidate navigation path with the highest avoidance effectiveness is taken as the target navigation path of the target vehicle.

[0012] In a preferred embodiment, when the lane-level guidance generation module performs parsing and reconstruction of the lane topology and real-time occupancy status of the target navigation path to obtain the lane-level guidance information of the target vehicle, it is specifically used for: The lane topology of the target navigation path is analyzed to obtain the lane connection relationship of the target navigation path; Based on the target navigation path, obtain the lane occupancy status data in the dynamic traffic situation map; Based on the lane occupancy status data, identify the impassable lanes in the lane connection relationships to obtain the lane drivability status of the target navigation path; Based on the lane connectivity and lane drivability status, lane drivability links are synthesized for the target navigation path to obtain a sequence of feasible lanes for the target navigation path. Based on the feasible lane sequence and the lane occupancy status data, lane-level guidance information for the target vehicle is generated.

[0013] In a preferred embodiment, when the vehicle control and feedback module executes the control of the target vehicle's steering and speed using the lane-level guidance information, and collects the target vehicle's actual driving trajectory and speed data as execution status data, and feeds the execution status data back to the roadside node, it is specifically used for: The target lane, target speed, and lane change command in the lane-level guidance information are compiled into a control sequence to obtain the vehicle control command set of the target vehicle; The vehicle control command set is transmitted to the target vehicle to drive the target vehicle along the target lane; The steering wheel angle, wheel speed, and lane line recognition results of the target vehicle are monitored to obtain the actual driving trajectory and real-time speed of the target vehicle. The actual driving trajectory is compared with the target lane, and the real-time speed is compared with the target speed to obtain the trajectory offset and speed deviation of the target vehicle. The trajectory offset, the speed deviation, and the vehicle control command set are encapsulated to obtain the execution status data message of the target vehicle; The execution status data message is fed back to the roadside node.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention integrates vehicle-to-vehicle perception, roadside monitoring, and cloud-based traffic data through a multi-source data fusion module. After spatiotemporal registration and conflict resolution, a conflict-free local traffic situation is obtained. Then, a traffic situation completion module completes the missing information and regulates spatiotemporal consistency, ultimately constructing a dynamic traffic situation map containing a topology layer, a state layer, and an event layer. This effectively improves the completeness and accuracy of traffic situation data, providing comprehensive and reliable data source support for navigation decisions.

[0015] 2. This invention generates candidate navigation paths by combining a candidate path planning module with a dynamic traffic situation map. After geospatial matching to evaluate avoidance effectiveness, the target path is determined. Then, lane-level guidance information is generated by parsing lane topology and real-time occupancy. At the same time, the guidance information is used to control vehicle steering and speed and provide feedback on the execution status. This not only improves the rationality of path planning and the accuracy of lane guidance, but also optimizes driving control in real time, significantly improving the efficiency of vehicle-to-everything (V2X) intelligent navigation and the smoothness of vehicle driving. Attached Figure Description

[0016] Figure 1 A system architecture diagram of an Internet of Things-based vehicle-to-everything (V2X) intelligent navigation system is provided as an embodiment of the present invention; 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

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments belong to some, but not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “said” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0019] Depending on the context, the word "if" or "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0020] Furthermore, the timing of the steps in the following method embodiments is merely an example and not a strict limitation.

[0021] In practice, the server-side equipment deployed in an IoT-based vehicle-to-everything (V2X) intelligent navigation system may consist of one or more devices. This IoT-based V2X intelligent navigation system can be implemented as: a business instance, a virtual machine, or hardware devices. For example, this IoT-based V2X intelligent navigation system can be implemented as a business instance deployed on one or more devices in a cloud node. Simply put, this IoT-based V2X intelligent navigation system can be understood as software deployed on a cloud node, used to provide an IoT-based V2X intelligent navigation system to various user terminals. Alternatively, this IoT-based V2X intelligent navigation system can also be implemented as a virtual machine deployed on one or more devices in a cloud node. This virtual machine contains application software for managing various user terminals. Or, this IoT-based V2X intelligent navigation system can also be implemented as a server composed of numerous identical or different types of hardware devices, with one or more hardware devices configured to provide an IoT-based V2X intelligent navigation system to various user terminals.

[0022] In terms of implementation, an IoT-based vehicle-to-everything (V2X) intelligent navigation system and its user terminal are mutually compatible. Specifically, if the IoT-based V2X intelligent navigation system is implemented as an application installed on a cloud service platform, the user terminal acts as a client establishing a communication connection with that application; or if the IoT-based V2X intelligent navigation system is implemented as a website, the user terminal acts as a webpage; or if the IoT-based V2X intelligent navigation system is implemented as a cloud service platform, the user terminal acts as a mini-program within an instant messaging application.

[0023] like Figure 1 The diagram shown is a system architecture diagram of an Internet of Things-based vehicle network intelligent navigation system provided in an embodiment of the present invention.

[0024] The Internet of Things (IoT)-based intelligent navigation system 100 for vehicles can be hosted on a cloud server. In terms of implementation, it can function as one or more service devices, or as an application installed on the cloud (e.g., a mobile service operator's server, server cluster, etc.), or it can be developed as a website. Depending on the functions implemented, the IoT-based intelligent navigation system 100 may include a multi-source data fusion module 101, a traffic situation completion module 102, a dynamic situation map construction module 103, a candidate path planning module 104, an optimal path decision module 105, a lane-level guidance generation module 106, and a vehicle control and feedback module 107. The modules described in this invention can also be referred to as units, which are a series of computer program segments that can be executed by an electronic device's processor and perform a fixed function, stored in the electronic device's memory.

[0025] In this embodiment of the invention, in an IoT-based vehicle-to-everything (V2X) intelligent navigation system, each of the above modules can be implemented independently and can call other modules. Here, "calling" can be understood as one module connecting to multiple modules of another type and providing corresponding services to those connected modules. The IoT-based V2X intelligent navigation system provided by this embodiment of the invention allows for adjustments to the applicability of the system architecture without modifying the program code. This is achieved by adding modules and directly calling them, enabling cluster-based horizontal expansion and flexibly expanding the IoT-based V2X intelligent navigation system. In practical applications, the above modules can be set in the same device or different devices, or they can be set in virtual devices, such as service instances in a cloud server.

[0026] The following describes, with reference to specific embodiments, the various components and specific workflows of an Internet of Things (IoT)-based vehicle-to-everything (V2X) intelligent navigation system: The multi-source data fusion module 101 is used to fuse the target vehicle's own perception dataset, the roadside monitoring dataset of the roadside perception facility, and the traffic dataset of the cloud data source to obtain the local traffic situation of the target vehicle. In this embodiment of the invention, when the multi-source data fusion module performs the fusion of the target vehicle's own perception dataset, the roadside monitoring dataset from the roadside perception facility, and the traffic dataset from the cloud data source to obtain the local traffic situation of the target vehicle, it is specifically used for: Receive the target vehicle's own perception dataset, the roadside monitoring dataset from the roadside perception facility, and the traffic dataset from the cloud data source; The vehicle perception dataset and the roadside monitoring dataset are spatiotemporally registered to obtain the spatiotemporally aligned dataset of the target vehicle. Identify conflicting traffic state information in the spatiotemporally aligned dataset; The conflict traffic state information is resolved to obtain a list of conflict-free local traffic states for the target vehicle. The list of conflict-free local traffic states is integrated with the traffic flow information in the traffic dataset to form the local traffic situation of the target vehicle.

[0027] Stable connections are established with the vehicle's onboard sensing equipment, roadside sensing facilities, and cloud data server via preset communication interfaces. The onboard sensing equipment collects information such as the vehicle's position, speed, distribution of surrounding obstacles, and turn signal status in real time, and organizes it into a vehicle sensing dataset according to a preset format. This dataset is then transmitted to the designated receiving end in real time via a dedicated vehicle-to-everything (V2X) communication protocol. The roadside sensing facilities continuously monitor the number of vehicles, the driving trajectories of each vehicle, and road surface conditions within their coverage area. This information is encoded according to a unified data standard to generate a roadside monitoring dataset, which is then transmitted to the designated receiving end via the communication unit of the roadside node. The cloud data server aggregates historical traffic data, real-time traffic control information, and temporary traffic events from all roads in the area to form a traffic dataset, which is then pushed to the designated receiving end via a 5G wireless network at fixed time intervals. The receiving end receives and temporarily stores these three datasets according to preset data storage rules.

[0028] First, the timestamp and geographic coordinate information of each data record in the vehicle perception dataset and the roadside monitoring dataset are extracted. The timestamp is based on the internationally unified UTC time. Data with different timestamps in the two datasets are time-calibrated. The acquisition time difference of two conflicting data records is calculated, and the timestamp of the later data record is adjusted to the same time point as the earlier data record to ensure that the data in the same time dimension can be compared. The geographic coordinates are based on the national 2000 geodetic coordinate system. The geographic coordinate data in the roadside monitoring dataset collected based on the local coordinate system of the roadside equipment is transformed. Through the preset coordinate transformation rules, the geographic coordinates of the roadside monitoring dataset are made consistent with the geographic coordinates of the vehicle perception dataset. After completing the spatiotemporal registration, the information corresponding to the same time and the same geographic coordinate in the vehicle perception dataset and the roadside monitoring dataset are associated and bound to form the spatiotemporally aligned dataset of the target vehicle.

[0029] Traffic state information in the spatiotemporally aligned dataset is categorized into categories including real-time vehicle location, vehicle speed, road traffic status, and the type and location of surrounding obstacles. For each category, numerical or state comparisons are performed on the same type of information at the same time and geographical coordinates in the vehicle's perception dataset and the roadside monitoring dataset. If, at the same time and geographical coordinates, the vehicle location recorded in the vehicle's perception dataset deviates from the vehicle location recorded in the roadside monitoring dataset by more than 5 meters, or the vehicle speed recorded in the vehicle's perception dataset deviates from the vehicle speed recorded in the roadside monitoring dataset by more than 10 km / h, or the road traffic status descriptions recorded by the two datasets are inconsistent, then the traffic state information is determined to be conflicting traffic state information. All the information determined to be conflicting is filtered out one by one to complete the identification of conflicting traffic state information in the spatiotemporally aligned dataset.

[0030] Data collected by roadside sensing facilities is used as the benchmark for conflict resolution. For each identified conflict traffic state information, the corresponding traffic state information of the same type at the corresponding time and geographical coordinates in the roadside monitoring dataset is searched. The corresponding information in the roadside monitoring dataset is directly used to replace the conflicting information in the vehicle's sensing dataset. If the corresponding information of the same type at the corresponding time and geographical coordinates in the roadside monitoring dataset is missing, reference information of the same type of traffic state in the same area and at a similar time is extracted from the cloud traffic dataset. The reasonable traffic state information is confirmed by combining the historical traffic patterns of the area. The confirmed information is used to replace the conflict information in the vehicle's sensing dataset. After all conflict traffic state information has been resolved, the remaining conflict-free traffic state information in the vehicle's sensing dataset and the roadside monitoring dataset is integrated and classified according to time sequence and geographical coordinate distribution to form a list of conflict-free local traffic states for the target vehicle.

[0031] First, the information in the conflict-free local traffic status list is structured. Each piece of information in the list is populated into the corresponding field according to the preset field format of "timestamp-geographic coordinates-vehicle location-speed-road traffic status-obstacle information," forming a structured local traffic information table. Next, traffic flow information is extracted from the traffic dataset in the cloud data source. This traffic flow information includes the average speed of each road segment in the area, the number of vehicles passing through the road segment per unit time, and the vehicle distribution density within the road segment. Similarly, the traffic flow information is structured according to the field format of "road segment number-average speed-traffic flow-vehicle density," forming a structured traffic flow information table. Then, based on the road segment number and geographic coordinates, the average speed, traffic flow, and vehicle density information of each road segment in the traffic flow information table are matched to the corresponding traffic status records in the local traffic information table. This ensures that each record in the local traffic information table is supplemented with the corresponding regional traffic flow background information. Finally, all the associated and integrated information is organized according to a preset situational presentation format to form the local traffic situation of the target vehicle.

[0032] The beneficial effects are as follows: A clear and executable dataset reception process ensures the complete and accurate acquisition of vehicle-sensed, roadside monitoring, and cloud-based traffic datasets; standardized spatiotemporal registration eliminates data discrepancies in time and space, preventing data ineffectiveness due to spatiotemporal inconsistencies; specific deviation thresholds and state comparison rules accurately identify conflicting traffic state information, and then reliably resolve conflicts using roadside data as a benchmark combined with cloud-based reference information, ensuring data validity; finally, structured integration supplements traffic flow information, generating a complete and accurate local traffic situation, providing high-quality foundational data for subsequent traffic situation completion, dynamic situation map construction, and other follow-up stages, effectively improving the reliability of system data processing and laying a solid data foundation for accurate navigation decisions.

[0033] The traffic situation completion module 102 is used to complete the missing traffic information in the local traffic situation to obtain the traffic situation data of the target vehicle. In this embodiment of the invention, when the traffic situation completion module completes the missing traffic information in the local traffic situation to obtain the traffic situation data of the target vehicle, it is specifically used for: A consistency check is performed on the traffic state data in the local traffic situation to obtain anomaly markers for the local traffic situation. The geographical area where the anomaly-marked data points are located is defined as the area to be supplemented for the local traffic situation; When the traffic status data in the local traffic situation has a road segment that is directly adjacent to the area to be completed, the traffic status data is used as boundary condition data. Based on the boundary condition data, state propagation calculation is performed on the traffic state of the area to be completed to obtain the estimated traffic state data of the area to be completed. Within the area to be filled, the estimated traffic state data is mapped to the local traffic situation to obtain the intermediate traffic situation data of the target vehicle; The intermediate traffic situation data is subjected to spatiotemporal consistency normalization to obtain the traffic situation data of the target vehicle.

[0034] The formula for calculating the estimated traffic density from the estimated traffic state data is as follows: ; In the formula, The location within the area to be completed At any moment The estimated traffic density, For the position The propagation weight coefficient, The upstream boundary traffic density in the boundary condition data. The downstream boundary traffic density in the boundary condition data. The upstream boundary traffic speed in the boundary condition data. The downstream boundary traffic speed in the boundary condition data. For the position The road network topology distance to the upstream boundary For the position The road network topology distance to the downstream boundary.

[0035] First, two core dimensions for traffic status data consistency verification are identified: temporal continuity verification and spatial rationality verification. Temporal continuity verification examines traffic status data at different timestamps for the same geographic coordinate point within a local traffic situation, checking whether the variation range of adjacent timestamp data conforms to actual traffic patterns. For example, the vehicle speed at the same location should not exhibit a large jump without a reasonable cause within adjacent timestamps. If the vehicle speed at a certain coordinate point is 30 km / h at time t1 and suddenly jumps to 70 km / h at time t2 without any special traffic event records, then the data is determined to have temporal continuity anomalies. Spatial rationality verification examines traffic status data at adjacent geographic coordinate points at the same timestamp, checking whether there are logical contradictions in the traffic status of adjacent road segments. For example, if a road segment is recorded as being in a smooth state within the same timestamp, but the next road segment directly connected to it is recorded as being in a congested state without any abnormal events such as construction or accidents, then the data of that adjacent road segment is determined to have spatial rationality anomalies. All data points determined to be abnormal by temporal continuity verification and spatial rationality verification are marked one by one, ultimately obtaining the anomaly markings for the local traffic situation.

[0036] Based on each anomaly-marked data point, the boundary of the area to be completed is determined by combining the actual division range of the road segment. First, the road segment number and the start and end geographic coordinates of the road segment where the anomaly-marked data point is located are obtained. If there are multiple anomaly-marked data points in a single road segment, the entire geographic range of the road segment is directly defined as the area to be completed. If the anomaly-marked data points are located on multiple adjacent road segments, it is first confirmed whether there is a continuous road connection between these road segments. Then, the geographic ranges of these adjacent road segments are merged to form a continuous geographic area. The merged geographic area is the area to be completed for the local traffic situation, ensuring that the area to be completed can completely cover the geographic range where all anomaly-marked data points are located.

[0037] First, we analyze all road segments corresponding to traffic status data in the local traffic situation and determine their connectivity with the road segments in the area to be supplemented. Directly adjacent road segments refer to those that are physically connected to the road segments in the area to be supplemented by an intersection and have no other road segments in between. For example, if the area to be supplemented is road segment A, the end point of road segment B is directly connected to the start point of road segment A through a crossroads, and the start point of road segment C is directly connected to the end point of road segment A through a T-junction, then road segment B and road segment C are directly adjacent road segments in the area to be supplemented. Then, we extract all traffic status data from these directly adjacent road segments, including vehicle speed, traffic density, road traffic conditions, etc., and organize and archive these extracted traffic status data to form boundary condition data.

[0038] The upstream and downstream boundary data of the area to be completed are distinguished from the boundary condition data. The upstream boundary data refers to the traffic status data of the road segments adjacent to the starting point of the area to be completed, and the downstream boundary data refers to the traffic status data of the road segments adjacent to the ending point of the area to be completed. Among them, the traffic density of the upstream boundary is... The traffic density data for this road segment is extracted directly from the upstream boundary data, while the downstream boundary traffic density... The traffic density data for this road segment is directly extracted from the downstream boundary data, while the traffic speed data is from the upstream boundary data. The traffic speed data for this road segment is extracted directly from the upstream boundary data, and the downstream boundary traffic speed... The traffic speed data for this road segment was directly extracted from the downstream boundary data; the location within the area to be supplemented is yet to be determined. Road network topology distance to upstream boundary From location along the road centerline using a distance measuring tool The path length and location measured from the starting point to the upstream boundary endpoint Road network topology distance to downstream boundary From location along the road centerline using a distance measuring tool The path length and location measured from the starting point to the downstream boundary endpoint. Propagation weight coefficient It is based on and Confirm, by position The closer to the upstream boundary, the better. The smaller the value and the closer it is to the downstream boundary, the better. Larger values ​​are assigned a fixed value to the position within the area to be filled. At any moment Estimated traffic density Calculated in the following manner: First, based on... and Calculate the upstream boundary traffic status and transfer it to the location. Time, using Subtracting this time yields the effective time corresponding to the traffic density at the upstream boundary, and this time is determined. At the same time, according to and Calculate the downstream boundary traffic status and transfer it to the location. Time, using Subtracting this time yields the effective time corresponding to the downstream boundary traffic density and determines it. Then subtract 1 Later and Multiply, use and Multiplication, adding the two products together, yields... Based on this, the traffic conditions of the initial part of the area to be supplemented are gradually extrapolated according to the natural flow of traffic from upstream to downstream. For example, if the traffic density of the upstream boundary section is 20 vehicles per kilometer and the speed is 40 kilometers per hour, and the road type of the initial part of the area to be supplemented is the same as that of the upstream boundary section, then the traffic density and speed of the initial part of the area to be supplemented are initially set to be consistent with the upstream boundary data. Then, the extrapolation process is adjusted in combination with the downstream boundary data. If the traffic density of the downstream boundary section is 22 vehicles per kilometer and the speed is 38 kilometers per hour, when extrapolating from the initial part of the area to be supplemented downstream, the traffic density is gradually adjusted from 20 vehicles per kilometer to 22 vehicles per kilometer, and the speed is adjusted from 40 kilometers per hour to 38 kilometers per hour, to ensure that the extrapolation process conforms to the gradual change of traffic flow. After the traffic condition extrapolation of the entire area to be supplemented is completed, the extrapolated traffic condition data of the area to be supplemented is compiled.

[0039] First, the geographic coordinates within the area to be supplemented are divided into grids. The area to be supplemented is divided into multiple continuous grid units at fixed geographic coordinate intervals, with each grid unit corresponding to a unique geographic coordinate range. Then, the estimated traffic status data is split into grid units to ensure that each grid unit has corresponding estimated traffic status information. Next, the original data location corresponding to each grid unit in the local traffic situation is found, and the split estimated traffic status data is filled into the corresponding original data location one by one, covering the original abnormal or missing data. After filling, the normal data that has not been adjusted in the local traffic situation and the filled estimated data in the area to be supplemented are integrated to form the intermediate traffic situation data of the target vehicle.

[0040] First, temporal consistency normalization is performed by extracting time-series data of all identical geographic coordinates from the intermediate traffic situation data, arranging them in chronological order of timestamps, and checking the variation range of adjacent timestamp data. If the speed change of a coordinate point between adjacent timestamps exceeds 15 km / h, and there is no traffic event to support this change, then the speed data with the larger variation range is adjusted to a reasonable value with a speed difference of less than 15 km / h from the previous timestamp. Next, spatial consistency normalization is performed by extracting data of all adjacent geographic coordinates at the same timestamp from the intermediate traffic situation data, and checking the difference in traffic density between adjacent coordinates. If the difference in traffic density between adjacent coordinates at the same timestamp exceeds 5 vehicles per kilometer, and there are no differences in physical conditions such as changes in road width or entrances / exits, then the data of the coordinate point with higher density is adjusted to a reasonable value with a density difference of less than 5 vehicles per kilometer from the adjacent coordinates. After completing temporal and spatial consistency normalization, the traffic situation data of the target vehicle is obtained.

[0041] The beneficial effects are as follows: By clearly verifying temporal continuity and spatial rationality, abnormal data in local traffic situations can be accurately identified and anomaly markers can be formed, providing a clear objective for subsequent data completion work; the areas to be completed are delineated according to road segments, ensuring the accuracy of the completion scope and its alignment with the actual road layout; traffic status data from directly adjacent road segments are used as boundary condition data, while clearly defining the source and calculation logic of each parameter for calculating traffic density, providing a reliable and practical basis for traffic status estimation in the areas to be completed; through state propagation calculations combining upstream and downstream boundary data with traffic flow patterns, estimated traffic status data that conforms to actual traffic scenarios is obtained; the estimated data is mapped to local traffic situations to form intermediate data, and then spatiotemporal consistency regularization is performed to eliminate data contradictions, ultimately yielding complete, accurate, and logically consistent traffic situation data. This provides high-quality data support for the construction of dynamic traffic situation maps and subsequent navigation path planning, effectively ensuring the accuracy and reliability of the system's navigation decisions.

[0042] The dynamic situation map construction module 103 is used to construct a dynamic traffic situation map of the target vehicle at a roadside node, using the road connectivity relationship in the traffic situation data as the topology layer, the traffic flow status in the traffic situation data as the state layer, and the traffic anomaly area information in the traffic situation data as the event layer. In this embodiment of the invention, when the dynamic situation map construction module is executed at a roadside node, using the road connectivity relationships in the traffic situation data as the topology layer, the traffic flow status in the traffic situation data as the state layer, and the traffic anomaly area information in the traffic situation data as the event layer, to construct the dynamic traffic situation map of the target vehicle, it is specifically used for: At the roadside nodes, the road endpoints with connectivity in the traffic situation data are taken as connection points, and the road segments located between the connection points are taken as connection segments. A topology layer for the traffic situation data is constructed using the connection points as nodes and the connection segments as edges. By integrating the traffic flow status data of the same connection segment in the traffic situation data, the comprehensive traffic flow parameters of the traffic situation data are obtained; Using the comprehensive traffic flow parameters as the dynamic attributes of the corresponding edges in the topology layer, the state layer of the traffic situation data is obtained; Read the geographical range affected by the traffic anomaly area information in the traffic situation data; The geographical range is mapped to the corresponding graph elements in the topology layer, and the graph elements are marked with event attributes to obtain the event layer of the traffic situation data; The dynamic attributes of the state layer and the event attribute tags of the event layer are respectively sent to the topology layer to obtain the attribute binding graph of the target vehicle; The attribute binding graph is subjected to graph attribute situationalization to obtain the dynamic traffic situational graph of the target vehicle.

[0043] When the dynamic situation map construction module performs graph attribute situationalization on the attribute binding graph to obtain the dynamic traffic situation map of the target vehicle, it is specifically used for: Identify graph elements in the attribute binding graph that have missing attributes and graph elements that have conflicting event attributes, in order to generate a set of elements to be processed in the attribute binding graph; Based on the time series information of the traffic situation data, the missing attributes in the set of elements to be processed are supplemented to obtain the first processing result of the set of elements to be processed; Based on the spatiotemporal correlation of the original abnormal events in the traffic situation data, the contradictions in the event attributes in the first processing result are resolved to obtain the second processing result of the set of elements to be processed. The second processing result is fused and reconstructed to obtain the dynamic traffic situation map of the target vehicle.

[0044] At the roadside nodes, the endpoint information of all roads is first extracted from the traffic situation data, including the unique identifier and geographic coordinates of each endpoint. Then, based on the road connectivity recorded in the traffic situation data, it is determined whether there is a physical connection between the endpoints of different roads. For example, if two endpoints belong to adjacent roads and correspond to the same location at the actual intersection, these two endpoints are identified as connection points. Subsequently, using the connection points as boundaries, the road sections located between the two connection points that are continuous and have no other connection points inserted are identified as connection segments, ensuring that the start and end points of each connection segment are connection points.

[0045] Defined connection points are used as nodes in the graphical structure. A unique node number is assigned to each connection point, and its corresponding geographic coordinate information is recorded. Defined connection segments are used as edges in the graphical structure. A unique edge number is assigned to each connection segment, and the node numbers of its two corresponding connection points are associated to clarify the connection relationship between the edge and the node. According to the association relationship between the node and the edge, all nodes and edges are organized into a graphical structure that conforms to the actual road connectivity. This graphical structure is the topology layer of the traffic situation data.

[0046] Traffic flow status data for the same connecting segment within a preset time window is extracted from traffic situation data. This data includes vehicle speed, number of vehicles passing through per unit time, and vehicle density within the segment. For the extracted vehicle speed data, the arithmetic mean of all speed values ​​within the time window is calculated to obtain the average speed. For the number of vehicles passing through per unit time, the total number within the time window is accumulated and divided by the time length to obtain the average traffic flow. For the vehicle density, the arithmetic mean of the density values ​​at each time point within the time window is calculated to obtain the average vehicle density. The average speed, average traffic flow, and average vehicle density are combined to form the comprehensive traffic flow parameters for the connecting segment. This method is used to obtain the comprehensive traffic flow parameters for all connecting segments.

[0047] The connection segment corresponding to each comprehensive traffic flow parameter is found in the topology layer. The association between the comprehensive traffic flow parameter and the edge is established by the unique number of the edge. The average driving speed, average traffic flow, and average vehicle density in the comprehensive traffic flow parameter are respectively used as dynamic attribute items of the corresponding edge and stored in the attribute field of the edge to ensure that each edge corresponds to a unique comprehensive traffic flow parameter. After all edges have completed the dynamic attribute association, the resulting graph structure with edge dynamic attributes is the state layer of traffic situation data.

[0048] Traffic anomaly area information is read from traffic situation data. This information includes the type of anomaly event and the boundary geographic coordinates of the affected area. The boundary geographic coordinates are recorded in the form of latitude and longitude and usually include the coordinates of the four vertices of the anomaly area. The longitude and latitude data of each vertex are extracted in the order of the vertex coordinates to determine the geographic range of the anomaly area.

[0049] The latitude and longitude coordinates of the identified abnormal area are compared with the geographic coordinates of all nodes and edges in the topology layer to determine whether the geographic coordinates of the nodes fall within the abnormal area, and whether the starting and ending coordinates of the edges and the coordinates of the middle road segments of the connecting segments overlap with the abnormal area. Nodes and edges that fall within or overlap with the abnormal area are identified as corresponding graph elements. Event attribute tags are added to these graph elements, including the abnormal event type, abnormal start time, and abnormal impact level. The event attribute tag of each graph element corresponds one-to-one with the information of the corresponding abnormal area. After the tagging is completed, the event layer of traffic situation data is formed.

[0050] The dynamic attributes of each edge are extracted from the state layer, and these dynamic attributes are sent to the corresponding edges in the topology layer using the edge's unique number, so that the edges in the topology layer carry dynamic attributes. The event attribute markers of each graph element are extracted from the event layer, and these event attribute markers are sent to the corresponding nodes or edges in the topology layer using the node or edge's unique number, so that the nodes and edges in the topology layer carry event attributes. At this time, all nodes and edges in the topology layer are simultaneously bound to dynamic attributes and event attributes, and the resulting graph structure is the attribute binding graph of the target vehicle.

[0051] Examine the attributes of each node and edge in the attribute-bound graph one by one to see if there are any missing dynamic attributes or event attributes, and filter out the graph elements with missing attributes; at the same time, check the event attributes of each graph element to see if there are any contradictory markers, and filter out the graph elements with conflicting event attributes; summarize all graph elements with missing attributes and conflicting event attributes to form the set of unprocessed elements in the attribute-bound graph.

[0052] The time series information of each graph element in the set of elements to be processed is extracted from the traffic situation data. The time series information contains the complete attribute data of the graph element at continuous historical time points. For graph elements with missing attributes, the complete attribute data closest to the current time is found in the time series information. For example, if an edge is currently missing the average driving speed attribute, and its average driving speed in the previous minute is 45 kilometers per hour and the attribute is complete, then the average driving speed of 45 kilometers per hour is used as supplementary data to fill the currently missing attribute field. After supplementing all graph elements with missing attributes, the first processing result of the set of elements to be processed is obtained.

[0053] The spatiotemporal correlation information of original abnormal events is extracted from traffic situation data. The spatiotemporal correlation information includes the occurrence time, end time, and corresponding geographical impact range of each abnormal event. For graph elements with conflicting event attributes in the first processing result, the occurrence time of the original abnormal events corresponding to the conflicting event attributes is compared, and the abnormal event attribute with the later occurrence time is retained. For example, if a node is marked with both "road construction" and "traffic accident", the event attribute of "traffic accident" is retained. If the conflicting events occur at the same time, the actual impact event is determined by combining the geographical impact range. For example, if both events occur at 14:00, but the geographical impact range of "traffic accident" completely covers the node, the event attribute of "traffic accident" is retained. After resolving all conflicting event attributes, the second processing result of the set of elements to be processed is obtained.

[0054] Each graph element and its complete, conflict-free attributes in the second processing result are re-associated with the corresponding nodes and edges in the topology layer to ensure that each node and edge carries accurate attribute information. According to the actual geographical coordinates and connectivity of the road, all nodes and edges are spatially calibrated to make the graphical structure consistent with the actual road layout. At the same time, the attribute information of all nodes and edges is integrated to form a graphical data structure that can intuitively display the road topology, real-time traffic flow status, and traffic anomalies. This data structure is the dynamic traffic situation map of the target vehicle.

[0055] The beneficial effects are as follows: by clearly defining the connection points and segments and constructing a hierarchical process of topology, state, and event layers, the basic structure of the dynamic traffic situation map is ensured to be accurate and closely matches the actual road conditions; by supplementing missing attributes with time-series information and resolving conflicts through the spatiotemporal correlation of original abnormal events, the integrity and consistency of graph element attributes are guaranteed; the final dynamic traffic situation map can clearly present road connectivity, real-time traffic flow status, and traffic anomaly information, providing comprehensive and accurate traffic situation data for subsequent candidate path planning and optimal path decision-making, effectively improving the navigation system's ability to perceive and judge the traffic environment.

[0056] The candidate path planning module 104 is used to generate candidate navigation paths for the target vehicle based on the road connectivity and traffic conditions in the dynamic traffic situation map. In this embodiment of the invention, when the candidate path planning module generates candidate navigation paths for the target vehicle based on the road connectivity and traffic conditions in the dynamic traffic situation map, it is specifically used for: Based on the dynamic traffic situation map, the initial reachable road segment of the target vehicle is constructed, starting from the current location of the target vehicle and ending at the destination location of the target vehicle. Based on the traffic status data of the initially accessible road segment, the real-time traffic impedance of the road segment edges in the initially accessible road segment is quantified; Based on the real-time traffic impedance, alternative routes are enumerated on the dynamic traffic situation map to obtain the initial path set for the target vehicle. The topological similarity and traffic impedance similarity of the paths in the initial path set are analyzed to obtain the candidate navigation path set for the target vehicle.

[0057] Extract the geographic coordinates and connectivity of all connection points and segments from the topology layer of the dynamic traffic situation map. Determine the geographic coordinates corresponding to the current location of the target vehicle, find the connection point where the coordinates are located, and set it as the starting connection point. Simultaneously, determine the geographic coordinates corresponding to the destination location of the target vehicle, find the connection point where the coordinates are located, and set it as the ending connection point. Starting from the starting connection point, search for all connection segments directly connected to the connection point through road segment edges, and add these connection segments to the temporary road segment list. Then, starting from the other end connection point of each connection segment in the temporary road segment list, continue to search for connection segments directly connected to these connection points that are not in the temporary road segment list, and repeat this operation until the temporary road segment list contains all connection segments that can extend to the ending connection point. Organize the temporary road segment list in the logical order of "starting connection point - road segment edge - intermediate connection point - road segment edge - ending connection point" to form an ordered set containing road segment edge numbers and information of the two end connection points, thus obtaining the initial reachable road segment for the target vehicle.

[0058] Traffic state data for each road segment edge within the initially accessible road segment is retrieved from the state layer of the dynamic traffic situation map. This data includes the average driving speed, average traffic flow, and average vehicle density of the road segment edge. A unified real-time traffic impedance quantification rule is established: for the average driving speed dimension, an impedance value of 2 corresponds to an average driving speed of 50 km / h, increasing by 1 for every 5 km / h decrease; for the average traffic flow dimension, an impedance value of 2 corresponds to an average traffic flow of 30 vehicles / hour, increasing by 1 for every 15 vehicles / hour increase; and for the average vehicle density dimension, an impedance value of 2 corresponds to an average vehicle density of 15 vehicles / km, increasing by 1 for every 5 vehicles / km increase. For each road segment edge, the average driving speed, average traffic flow, and average vehicle density are converted into corresponding component impedance values ​​according to the above rules. These three component impedance values ​​are then added together to obtain the real-time traffic impedance of that road segment edge, thus completing the quantification of the real-time traffic impedance for all road segment edges within the initially accessible road segment.

[0059] Using the topology of the dynamic traffic situation map as a framework, the starting connection point is taken as the starting position for path enumeration. The road segment edge connected to the starting connection point with the lowest real-time traffic impedance is selected first, and the road segment edge is extended to its other connection point. At the connection point, the road segment edge connected to the point, not included in the current path, and with the lowest real-time traffic impedance is selected again. This operation is repeated until the end connection point is reached. This complete route formed by connecting multiple road segment edges in sequence is recorded as an initial path. In the above manner, the selection order of the starting road segment edge and subsequent road segment edges is changed in sequence. All routes that can reach the end connection point from the starting connection point without duplicate road segment edges are enumerated. All recorded initial paths are summarized to form the initial path set of the target vehicle.

[0060] Extract the topology information and total impedance of each path in the initial path set. The topology information includes the numbers of all connection points, connection segments, and their connection order. The total impedance is the sum of the real-time impedances of all road segments within the path. For any two paths, compare their topology information. If the two paths contain identical connection point numbers, identical connection segment numbers, and identical connection order, they are considered to have similar topologies. Calculate the difference in total impedance between the two paths. If the difference is less than or equal to 3, they are considered to have similar impedances. Group the paths in the initial path set that have similar topologies and impedances into the same group. Within each group, compare the total impedance of all paths and retain the path with the lowest total impedance. Organize the paths retained in all groups into an ordered list to form the candidate navigation path set for the target vehicle.

[0061] The beneficial effects are as follows: by combining the topological relationships of dynamic traffic situation maps to determine initial reachable road segments, it ensures coverage of all potential paths from the origin to the destination, avoiding the omission of effective routes; by using fixed rules to quantify traffic state data into real-time traffic impedance, it provides a unified standard for measuring the difficulty of road segments and provides an objective basis for path comparison; by enumerating the initial path set based on impedance, it ensures the comprehensiveness of path selection and avoids the loss of the optimal path due to incomplete enumeration; and by simplifying paths through dual similarity analysis of topology and impedance, it removes redundant and inefficient paths, ultimately obtaining a concise and efficient candidate navigation path set, providing an accurate and high-quality selection basis for subsequent optimal path decisions, and improving the reliability and efficiency of navigation path planning.

[0062] The optimal path decision module 105 is used to perform geospatial matching between the candidate navigation path and the traffic anomaly area in the dynamic traffic situation map, and evaluate the avoidance effectiveness of the candidate navigation path based on the matching result, so as to obtain the target navigation path of the target vehicle. In this embodiment of the invention, when the optimal route decision module performs geospatial matching of the candidate navigation path with the traffic anomaly area in the dynamic traffic situation map, and evaluates the avoidance effectiveness of the candidate navigation path based on the matching result to obtain the target navigation path for the target vehicle, it is specifically used for: Spatial topology analysis is performed on the candidate navigation path and the traffic anomaly area in the dynamic traffic situation map to obtain the affected path segment of the candidate navigation path; Based on the type and severity level of the affected path segments, and in combination with the length of the affected path segments, a comprehensive anomaly impact degree for implementing candidate navigation paths is obtained; Based on the comprehensive anomaly impact, the avoidance effectiveness of the candidate navigation path is determined, and the candidate navigation path with the highest avoidance effectiveness is taken as the target navigation path of the target vehicle.

[0063] First, extract all road segment edges contained in each candidate navigation path and record the start and end geographic coordinates of each road segment edge. Simultaneously, extract the geographic range of all traffic anomaly areas from the event layer of the dynamic traffic situation map. This range is determined by the latitude and longitude coordinates of the four vertices of the anomaly area boundary. Spatially compare the start and end geographic coordinates of the road segment edges of each candidate navigation path with the boundary vertex coordinates of the traffic anomaly area to determine whether the longitude range of the road segment edge overlaps with the longitude range of the anomaly area, and whether the latitude range of the road segment edge overlaps with the latitude range of the anomaly area. If the latitude and longitude range of a certain road segment edge overlaps with the latitude and longitude range of any traffic anomaly area, then mark the road segment edge as an affected road segment edge. Summarize all the road segment edges marked as affected in each candidate navigation path to form the affected path segments of the candidate navigation path.

[0064] First, determine the corresponding scoring rules for the types and severity levels of affected path segments. Traffic anomalies are divided into three categories: traffic accidents, road construction, and temporary traffic control. Traffic accidents correspond to a high severity level and are assigned 5 points; road construction corresponds to a medium severity level and is assigned 3 points; and temporary traffic control corresponds to a low severity level and is assigned 2 points. Next, determine the scoring rules for the length of affected path segments, using 100 meters as the unit, assigning 1 point for every 100 meters, with any length less than 100 meters counted as 100 meters. For each candidate navigation path's affected path segment, first determine the base score based on its corresponding anomaly type and severity level, then measure the actual length of the affected path segment and calculate the length score according to the length scoring rules. Add the base score and the length score to obtain the comprehensive anomaly impact of the candidate navigation path, ensuring that each candidate navigation path corresponds to a unique comprehensive anomaly impact.

[0065] First, establish the rules for judging avoidance effectiveness. The comprehensive anomaly impact degree and avoidance effectiveness are inversely related. That is, the smaller the comprehensive anomaly impact degree value, the less the candidate navigation path is affected by traffic anomalies, and the higher the avoidance effectiveness. Sort all candidate navigation paths according to their comprehensive anomaly impact degree values ​​from smallest to largest. The candidate navigation path ranked first after sorting is the path with the smallest comprehensive anomaly impact degree and the highest avoidance effectiveness. Directly determine the candidate navigation path with the highest avoidance effectiveness as the target navigation path of the target vehicle to ensure that the target navigation path can avoid the impact of traffic anomaly areas to the greatest extent.

[0066] The beneficial effects are as follows: through precise spatial topology comparison, it can accurately identify affected path segments in candidate navigation paths, providing clear targets for subsequent evaluation; by calculating the comprehensive anomaly impact degree through fixed type level assignment and length assignment rules, it provides an objective and unified measurement standard for the anomaly impact degree of different candidate paths, avoiding subjective bias; based on the comprehensive anomaly impact degree, it reversely determines the avoidance effectiveness and selects the optimal path, ensuring that the target navigation path avoids traffic anomaly interference to the greatest extent, effectively improving the traffic efficiency and reliability of navigation paths, and ensuring smooth vehicle driving.

[0067] The lane-level guidance generation module 106 is used to analyze and reconstruct the lane topology and real-time occupancy status of the target navigation path to obtain the lane-level guidance information of the target vehicle. In this embodiment of the invention, when the lane-level guidance generation module performs parsing and reconstruction of the lane topology and real-time occupancy status of the target navigation path to obtain the lane-level guidance information of the target vehicle, it is specifically used for: The lane topology of the target navigation path is analyzed to obtain the lane connection relationship of the target navigation path; Based on the target navigation path, obtain the lane occupancy status data in the dynamic traffic situation map; Based on the lane occupancy status data, identify the impassable lanes in the lane connection relationships to obtain the lane drivability status of the target navigation path; Based on the lane connectivity and lane drivability status, lane drivability links are synthesized for the target navigation path to obtain a sequence of feasible lanes for the target navigation path. Based on the feasible lane sequence and the lane occupancy status data, lane-level guidance information for the target vehicle is generated.

[0068] Extract all continuous road segments included in the target navigation path. For each road segment, obtain the number of lanes, the unique number of each lane, and the lane function attributes through the topology layer of the dynamic traffic situation map. Then, for the connection position of adjacent road segments, match the turning correspondence between each lane of the previous road segment and each lane of the next road segment. For example, the straight lane of the previous road segment corresponds to the straight lane of the next road segment, and the left-turn lane of the previous road segment corresponds to the left-turn lane of the next road segment. Organize the number of lanes and the turning correspondence between adjacent road segments into a structured table to obtain the lane connection relationship of the target navigation path.

[0069] Based on the road segment range of the target navigation path, real-time occupancy data for each lane within that range is retrieved from the state layer of the dynamic traffic situation map. Specifically, this includes the vehicle presence status, vehicle dwell time, and lane space occupancy ratio of each lane at the current moment. These data are recorded sequentially according to lane number to ensure that the occupancy information of each lane is completely extracted, thus obtaining lane occupancy status data in the dynamic traffic situation map.

[0070] The criteria for determining whether a lane is impassable are set as follows: when the space occupancy rate of a lane in the lane occupancy status data exceeds 90%, or when a vehicle remains stationary for more than 5 minutes without any signs of movement, the lane is determined to be impassable. Each lane in the lane connection relationship of the target navigation path is checked against the lane occupancy status data to see if it meets the impassable criteria. Lanes that meet the criteria are marked as "impassable", and lanes that do not meet the criteria are marked as "passable". All lane passable marks are associated with lane numbers and road segments to obtain the lane passability status of the target navigation path.

[0071] Starting from the starting segment of the target navigation path, select one of the "passable" lanes in that segment as the initial lane. Based on the lane connection relationship, find the "passable" lane corresponding to the initial lane in the next adjacent segment and associate the two lanes according to the driving order. Continue to use the associated lane in the next segment as the starting point, find the "passable" lane corresponding to it in the next segment after that and associate it. Repeat this process until all segments of the target navigation path are covered. Organize all the sequentially associated passable lanes into the format of "segment 1 lane X → segment 2 lane Y → segment 3 lane Z" to obtain the sequence of passable lanes for the target navigation path.

[0072] Based on the lane occupancy status data of each lane in the feasible lane sequence, the following speed guidance suggestions are set: if the lane space occupancy rate is less than 30%, the recommended driving speed is 80%-100% of the speed limit for that section; if the occupancy rate is between 30% and 60%, the recommended driving speed is 50%-80% of the speed limit for that section; if the occupancy rate is between 60% and 90%, the recommended driving speed is 30%-50% of the speed limit for that section. At the same time, based on the lane switching sequence of the feasible lane sequence, the timing of lane change prompts is determined. For example, if the vehicle is currently traveling in lane 2 of section 1 and needs to switch to lane 3 in the next section, a lane change prompt is generated 500 meters away from the intersection connecting section 1 and section 2. The speed suggestions, lane change prompt timings, and lane numbers for each lane are integrated to form real-time guidance content containing "current lane - suggested speed - next action (go straight / change lanes to a certain lane)", thus obtaining lane-level guidance information for the target vehicle.

[0073] The beneficial effects are as follows: by analyzing the lane topology of the target navigation path in detail, it ensures that the lane connection relationship accurately reflects the actual lane turning situation; it accurately obtains lane occupancy status data, providing an objective basis for identifying impassable lanes; it synthesizes a sequence of feasible lanes based on lane connection relationship and drivability status, ensuring that the sequence conforms to the driving direction and has no obstruction to passage; and it generates lane-level guidance information by combining feasible sequences and occupancy data, clarifying speed suggestions and lane change timing, thereby achieving precise navigation of vehicles at the lane level, effectively improving driving safety and traffic efficiency, and avoiding congestion or accidents caused by lane selection errors.

[0074] The vehicle control and feedback module 107 is used to control the steering and speed of the target vehicle with the lane-level guidance information, collect the actual driving trajectory and speed data of the target vehicle as execution status data, and feed the execution status data back to the roadside node.

[0075] In this embodiment of the invention, when the vehicle control and feedback module executes the control of the target vehicle's steering and speed using the lane-level guidance information, and collects the target vehicle's actual driving trajectory and speed data as execution status data, and feeds the execution status data back to the roadside node, it is specifically used for: The target lane, target speed, and lane change command in the lane-level guidance information are compiled into a control sequence to obtain the vehicle control command set of the target vehicle; The vehicle control command set is transmitted to the target vehicle to drive the target vehicle along the target lane; The steering wheel angle, wheel speed, and lane line recognition results of the target vehicle are monitored to obtain the actual driving trajectory and real-time speed of the target vehicle. The actual driving trajectory is compared with the target lane, and the real-time speed is compared with the target speed to obtain the trajectory offset and speed deviation of the target vehicle. The trajectory offset, the speed deviation, and the vehicle control command set are encapsulated to obtain the execution status data message of the target vehicle; The execution status data message is fed back to the roadside node.

[0076] The specific lane number, target speed, and lane change command requirements are extracted from the lane-level guidance information. The target lane number is binary encoded, converting lane 3 into a corresponding 8-bit binary code. For the target speed, based on the signal format recognizable by the vehicle controller, 50 km / h is converted into the corresponding pulse width modulation signal parameter, ensuring that this parameter can directly control the corresponding speed output of the vehicle's powertrain. For the lane change command, the required steering wheel angle and steering duration are calculated based on the lateral distance between the target lane and the current lane. The steering angle is converted into a steering motor control signal, and the duration is converted into a signal output duration. The encoded target lane signal, the converted target speed signal, and the steering and duration signals corresponding to the lane change command are arranged in chronological order of vehicle travel time to form a continuous control signal sequence, which is the vehicle control command set for the target vehicle.

[0077] The vehicle control command set is sent to the on-board controller of the target vehicle in the form of a real-time data stream via a dedicated vehicle-to-everything (V2X) communication module. After receiving the command set, the on-board controller first decodes the command signals to restore the target lane number, target speed, and lane change command parameters. Then, the on-board controller sends a steering control signal to the vehicle's steering system, driving the steering motor to adjust the steering wheel to the required angle for lane change, ensuring that the vehicle deviates from the target lane. At the same time, it sends speed control signals to the vehicle's power system and braking system, adjusting the throttle opening or brake pedal travel to stabilize the vehicle's speed at the target speed. During the lane change process, the on-board controller continuously receives timing signals from the command set and fine-tunes the steering angle and speed in real time until the vehicle fully enters the target lane and maintains stable driving, thus driving the target vehicle to travel along the target lane.

[0078] An angle sensor is installed on the steering wheel of the target vehicle to collect the steering wheel rotation angle data in real time, and the angle data is converted into an electrical signal and transmitted to the on-board data acquisition unit. Wheel speed sensors are installed on each of the four wheels of the vehicle to collect the number of rotations and rotation frequency of each wheel in real time. The instantaneous speed of the vehicle is calculated by multiplying the wheel circumference and rotation frequency. The instantaneous speeds of the four wheels are summed and averaged to obtain the real-time speed of the target vehicle. At the same time, a high-definition camera and a lidar are installed at the front of the vehicle to identify the position coordinates of the road lane lines. Combined with the steering wheel angle data and real-time speed data, the real-time position coordinates of the vehicle in the road coordinate system are calculated by the vehicle kinematic model. The continuous position coordinates are connected in time sequence to form the actual driving trajectory of the target vehicle, and finally the actual driving trajectory and real-time speed of the target vehicle are obtained.

[0079] The standard position coordinate range and target speed value of the target lane are retrieved from the vehicle data acquisition unit. The position coordinates of the actual driving trajectory are compared with the standard coordinate range of the target lane point by point. The difference in lateral distance between the vehicle position coordinates and the center line coordinates of the lane at each time point is calculated. This difference is the trajectory offset of the target vehicle. The real-time speed value and the target speed value are subtracted point by point. If the real-time speed is greater than the target speed, the positive number is the positive speed deviation. If the real-time speed is less than the target speed, the negative number is the negative speed deviation. Finally, the trajectory offset and speed deviation of the target vehicle are obtained.

[0080] The trajectory offset, speed deviation, and vehicle control command set are encapsulated according to a preset data message format. The message format consists of three parts: header, body, and checksum. The header contains the vehicle identification code and the message generation timestamp. The body is filled with specific data in the order of "vehicle control command set number - trajectory offset value - speed deviation value". The checksum uses a cyclic redundancy check algorithm to calculate the checksum of the body data to ensure that the data is not damaged during transmission. The header, body, and checksum are combined in sequence to form a complete data packet, which is the execution status data message of the target vehicle.

[0081] The target vehicle's onboard communication unit sends execution status data messages to the roadside node's communication receiving unit. Upon receiving the message, the roadside node's communication receiving unit first verifies the CRC code at the check tail to confirm that the message data is error-free or missing. After successful verification, the message is stored in the roadside node's local database, and the trajectory offset and speed deviation data in the message are synchronized to the dynamic traffic situation map update module for subsequent traffic situation correction and optimization, thus completing the operation of feeding back the execution status data message to the roadside node.

[0082] The beneficial effects are as follows: by accurately compiling control sequences from lane-level guidance information, the vehicle control commands can be accurately identified and executed by the onboard system; the command set is reliably transmitted to the vehicle and driven, ensuring that the vehicle strictly follows the target lane and speed; by acquiring actual driving data through multi-device collaborative monitoring and comparing it with target parameters to obtain deviation information, real-time control of the vehicle's driving status is achieved; the deviation information and command set are encapsulated into standard messages and fed back to roadside nodes, providing data support for roadside nodes to update traffic conditions and optimize subsequent navigation commands, forming a closed-loop control of "command-execution-monitoring-feedback", effectively improving the accuracy and safety of vehicle driving, and ensuring the stable and efficient operation of the vehicle-to-everything (V2X) intelligent navigation system.

[0083] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0084] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0085] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An Internet of Things based intelligent navigation system for vehicle-to-vehicle communication, characterized in that, The system comprises a multi-source data fusion module, a traffic situation completion module, a dynamic situation map construction module, a candidate path planning module, an optimal path decision module, a lane-level guidance generation module, and a vehicle control and feedback module, wherein: The multi-source data fusion module is configured to fuse a local traffic situation of a target vehicle by fusing a vehicle perception dataset of the target vehicle, a roadside monitoring dataset of a roadside sensing facility, and a traffic dataset from a cloud data source. The traffic situation completion module is configured to complete missing traffic information in the local traffic situation to obtain traffic situation data of the target vehicle. The dynamic situation map construction module is configured to construct a dynamic traffic situation map of the target vehicle at a roadside node by taking a road connectivity relationship in the traffic situation data as a topology layer, taking a traffic flow state in the traffic situation data as a state layer, and taking traffic anomaly region information in the traffic situation data as an event layer. The candidate path planning module is configured to generate a candidate navigation path of the target vehicle according to the road connectivity relationship and the traffic state in the dynamic traffic situation map. The optimal path decision module is configured to perform a geospatial matching between the candidate navigation path and a traffic anomaly region in the dynamic traffic situation map, and evaluate an evasion performance of the candidate navigation path according to a matching result to obtain a target navigation path of the target vehicle. The lane-level guidance generation module is configured to analyze and reconstruct a lane topology and a real-time occupancy of the target navigation path to obtain lane-level guidance information of the target vehicle. The vehicle control and feedback module is configured to control steering and speed of the target vehicle based on the lane-level guidance information, and collect actual driving trajectory and speed data of the target vehicle as execution state data, and feed back the execution state data to the roadside node.

2. The Internet of Things based intelligent navigation system for vehicular Internet of Things of claim 1, wherein, When the multi-source data fusion module is executed to fuse a local traffic situation of a target vehicle by fusing a vehicle perception dataset of the target vehicle, a roadside monitoring dataset of a roadside sensing facility, and a traffic dataset from a cloud data source, the multi-source data fusion module is specifically configured to: receive a vehicle perception dataset of a target vehicle, a roadside monitoring dataset of a roadside sensing facility, and a traffic dataset from a cloud data source; perform spatio-temporal registration on the vehicle perception dataset and the roadside monitoring dataset to obtain a spatio-temporal aligned dataset of the target vehicle; identify conflict traffic state information in the spatio-temporal aligned dataset; perform conflict resolution on the conflict traffic state information to obtain a conflict-free local traffic state list of the target vehicle; integrate the conflict-free local traffic state list and traffic flow information in the traffic dataset into a local traffic situation of the target vehicle.

3. The Internet of Things based intelligent navigation system for vehicular Internet of Things of claim 1, wherein, When the traffic situation completion module is executed to complete missing traffic information in the local traffic situation to obtain traffic situation data of the target vehicle, the traffic situation completion module is specifically configured to: perform consistency verification on traffic state data in the local traffic situation to obtain an anomaly label of the local traffic situation; delineate a geographical range where a data point of the anomaly label is located as a to-be-completed region of the local traffic situation; When the traffic state data in the local traffic situation is directly adjacent to a road segment in the to-be-completed region, the traffic state data is taken as the boundary condition data; Based on the boundary condition data, state propagation calculation is performed on the traffic state of the to-be-completed region to obtain calculated traffic state data of the to-be-completed region; The calculated traffic state data is mapped to the local traffic situation to obtain intermediate traffic situation data of the target vehicle in the to-be-completed region; The intermediate traffic situation data is subjected to spatiotemporal consistency regularization to obtain traffic situation data of the target vehicle.

4. The Internet of Things based intelligent navigation system for vehicular Internet of Things of claim 3, wherein, The calculation formula of the calculated traffic density in the calculated traffic state data is as follows: ； wherein, is the position within the region to be completed is the estimated traffic density at time , is the position is the propagation weight coefficient for the position is the upstream boundary traffic density in the boundary condition data, is the downstream boundary traffic density in the boundary condition data, is the upstream boundary traffic speed in the boundary condition data, is the downstream boundary traffic speed in the boundary condition data, is the position is the road network topological distance from the position to the upstream boundary, is the road network topological distance from the position to the downstream boundary.

5. The Internet of Things based intelligent navigation system for vehicular Internet of Things of claim 1, wherein, When the dynamic situation map construction module executes the construction of the dynamic traffic situation map of the target vehicle based on the road connectivity relationship in the traffic situation data as the topology layer, the traffic flow state in the traffic situation data as the state layer, and the traffic anomaly region information in the traffic situation data as the event layer, the dynamic situation map construction module is specifically configured to: In the roadside node, the road endpoints having a connection relationship in the traffic situation data are taken as connection points, and the road segments located between the connection points are taken as connection segments; The connection points are taken as nodes, and the connection segments are taken as edges to construct the topology layer of the traffic situation data; The traffic flow state data of the same connection segment in the traffic situation data is integrated to obtain the comprehensive traffic flow parameter of the traffic situation data; The comprehensive traffic flow parameter is taken as the dynamic attribute of the corresponding edge in the topology layer to obtain the state layer of the traffic situation data; The geographical range affected by the traffic anomaly region information in the traffic situation data is read; The geographical range is mapped to the corresponding graph element in the topology layer, and the graph element is subjected to event attribute marking to obtain the event layer of the traffic situation data; The dynamic attribute of the state layer and the event attribute marking of the event layer are respectively delivered to the topology layer to obtain the attribute binding graph of the target vehicle; The attribute binding graph is subjected to graph attribute situation to obtain the dynamic traffic situation map of the target vehicle.

6. The Internet of Things based intelligent navigation system for vehicular Internet of Things of claim 5, wherein, When the dynamic situation map construction module executes the graph attribute situation of the attribute binding graph to obtain the dynamic traffic situation map of the target vehicle, the dynamic situation map construction module is specifically configured to: Identify the graph elements with missing attributes and the graph elements with conflicting event attributes in the attribute binding graph to generate a to-be-processed element set of the attribute binding graph; According to the time sequence information of the traffic situation data, the missing attributes in the to-be-processed element set are supplemented to obtain a first processing result of the to-be-processed element set; According to the spatiotemporal correlation of the original abnormal events in the traffic situation data, the event attribute contradictions in the first processing result are eliminated to obtain a second processing result of the to-be-processed element set; The second processing result is subjected to fusion reconstruction to obtain the dynamic traffic situation map of the target vehicle.

7. The Internet of Things based intelligent navigation system for Internet of Vehicles as claimed in claim 1 wherein, When the candidate path planning module executes the generation of the candidate navigation path of the target vehicle according to the road connectivity relationship and the traffic state in the dynamic traffic situation map, the candidate path planning module is specifically configured to: constructing an initial reachable road segment of the target vehicle according to the dynamic traffic situation map, taking the current position of the target vehicle as a starting point and the destination position of the target vehicle as an ending point; quantifying real-time traffic impedance of a road segment edge in the initial reachable road segment according to traffic state data of the initial reachable road segment; performing alternative path enumeration on the dynamic traffic situation map based on the real-time traffic impedance, to obtain an initial path set of the target vehicle; analyzing topological structure similarity of paths and traffic impedance similarity of paths in the initial path set, to obtain a candidate navigation path set of the target vehicle.

8. The Internet of Things based intelligent navigation system for connected cars of claim 1, wherein, The optimal path decision module, when performing geographical spatial matching of the candidate navigation path and a traffic anomaly region in the dynamic traffic situation map, and evaluating evading efficiency of the candidate navigation path according to a matching result, to obtain a target navigation path of the target vehicle, is specifically configured to: performing spatial topological analysis of the candidate navigation path and the traffic anomaly region in the dynamic traffic situation map, to obtain an affected path segment of the candidate navigation path; obtaining a comprehensive anomaly influence degree of the candidate navigation path based on a type and a severity level of the affected path segment, and in combination with a length of the affected path segment; determining evading efficiency of the candidate navigation path according to the comprehensive anomaly influence degree, and taking a candidate navigation path with the highest evading efficiency as the target navigation path of the target vehicle.

9. The Internet of Things based intelligent navigation system for connected cars of claim 1, wherein, The lane-level guidance generation module, when performing analysis and reconstruction of a lane topology and a real-time occupancy situation of the target navigation path, to obtain lane-level guidance information of the target vehicle, is specifically configured to: analyzing a lane topology structure of the target navigation path, to obtain a lane connection relationship of the target navigation path; obtaining lane occupancy state data in the dynamic traffic situation map according to the target navigation path; identifying a lane impassable in the lane connection relationship according to the lane occupancy state data, to obtain a lane passability state of the target navigation path; performing lane passing link synthesis on the target navigation path based on the lane connection relationship and the lane passability state, to obtain a feasible lane sequence of the target navigation path; generating lane-level guidance information of the target vehicle according to the feasible lane sequence and the lane occupancy state data.

10. The Internet of Things based intelligent navigation system for vehicular Internet of Things of claim 1, wherein, The vehicle control and feedback module, when performing control of steering and speed of the target vehicle according to the lane-level guidance information, and collecting actual driving trajectory and speed data of the target vehicle as execution state data, and feeding back the execution state data to the road side node, is specifically configured to: performing control sequence compilation on a target lane, a target speed and a lane changing instruction in the lane-level guidance information, to obtain a vehicle control instruction set of the target vehicle; delivering the vehicle control instruction set to the target vehicle, to drive the target vehicle to travel along the target lane; monitoring a steering wheel angle, a wheel speed and a lane line recognition result of the target vehicle, to obtain actual driving trajectory and real-time speed of the target vehicle; The actual driving track is compared with the target lane, the real-time speed and the target speed to obtain a track offset amount and a speed deviation amount of the target vehicle; The track offset amount, the speed deviation amount and the vehicle control instruction set are packaged to obtain an execution state data packet of the target vehicle; The execution state data packet is fed back to the road side node.

Images