Navigation Method, System, and Medium Based on Vehicle-Road-Cloud and Digital Twin Technologies

Abstract

The present invention discloses a navigation method, system and medium based on vehicle-road cloud and digital twin technology; relates to the field of navigation technology; based on traditional navigation technology, the method is improved. This solution obtains weather warning data around the main vehicle and lane information of the main vehicle based on vehicle-road cloud technology, and the third-party service platform completes the construction of traffic warning information and dynamic scenes around the main vehicle. Through the digital twin technology, the twin presentation and reminder on the navigation platform can realize the display of warning information, road lanes, and the lane where the main vehicle is located. At the same time, the navigation data of the navigation platform can also be displayed at the same time, so that the user can more intuitively understand the scene conditions around the main vehicle and the warning information; through the collaborative optimization of multiple platforms, the division of labor and cooperation are improved to improve computing efficiency and accuracy.

Description

Technical Field

[0001] The present invention relates to the field of navigation technology, and in particular to a navigation method, system and medium based on vehicle-road-cloud and digital twin technology. Background Art

[0002] Existing navigation platforms typically present route information using a single-lane or multi-lane flat display format when providing driving navigation services. While this traditional navigation interface can provide users with basic road directions, turn instructions, and traffic conditions, it still suffers from the following significant drawbacks in real-world driving scenarios:

[0003] 1. Lane-level positioning ambiguity: Users cannot intuitively determine the specific lane (such as the left lane, middle lane, or right lane) of the current vehicle through the navigation interface. This can easily lead to misjudgment, especially on complex multi-lane roads (such as overpass forks or highway ramps), increasing the risk of incorrect lane changes or missing intersections.

[0004] 2. Lack of dynamic surrounding information: Traditional navigation systems only provide static lane markings or simple traffic signs, but lack visual feedback for real-time lane-level dynamic events (such as unusual parking, collisions, illegal lane-cutting, or unusual overtaking). Drivers must rely entirely on their eyes, which poses a significant safety risk in inclement weather or when visibility is obstructed.

[0005] 3. Insufficient interactivity and early warning: Existing technologies fail to deeply integrate lane-level environmental data (such as speed differences between adjacent vehicles and abnormal behavior) with the navigation interface, making it difficult for the system to provide users with proactive early warnings (such as prompting them to avoid faulty vehicles in adjacent lanes), reducing driving safety and the value of navigation assistance.

[0006] Therefore, there is an urgent need for a navigation solution that can achieve high-precision lane-level positioning, real-time dynamic environment perception and intelligent warning to solve the problems of single information presentation and poor interactivity in traditional technologies. Summary of the Invention

[0007] The technical problem to be solved by the present invention is: the problem of single information presentation and poor interactivity in traditional driving navigation technology. The purpose of the present invention is to provide a navigation method, system and medium based on vehicle-road cloud and digital twin technology. This solution provides a navigation method, system and medium based on vehicle-road cloud and digital twin technology. On the basis of traditional navigation technology, the method is improved. Based on the vehicle-road cloud technology, the scene data around the main vehicle (including the lane information of the main vehicle, etc.) and the blind spot warning, abnormal parking warning and abnormal weather warning data around the main vehicle are obtained. Finally, the digital twin technology is used to present and remind on the navigation platform, which can realize multi-lane display of the road, and the lane position of the current vehicle and surrounding vehicles can be displayed on the navigation platform together with the navigation data in real time, so that users can understand the scene conditions around the main vehicle and warning information more intuitively.

[0008] The present invention is achieved through the following technical solutions:

[0009] This solution provides a navigation method based on vehicle-road-cloud and digital twin technology, implemented on a third-party service platform, a navigation platform, and a vehicle-road-cloud platform. The method includes:

[0010] The third-party service platform requests the navigation platform to start navigation;

[0011] After the navigation platform successfully starts navigation, the navigation data is displayed in real time, and at the same time, the third-party service platform requests to start the vehicle-road cloud platform;

[0012] The vehicle-infrastructure cloud platform collects collaborative data and performs primary calculations, which it then sends to a third-party service platform. These primary calculations include determining weather warning information and the vehicle's lane, and performing denoising on the collaborative data.

[0013] The third-party service platform constructs a dynamic scene around the main vehicle based on the primary calculation results, and displays the dynamic scene around the main vehicle in real time on the navigation platform based on digital twin technology; the dynamic scene around the main vehicle includes: weather warning information, environmental information around the main vehicle and traffic warning information.

[0014] Working principle of this solution: This solution provides a navigation method, system and medium based on vehicle-road cloud and digital twin technology. It improves the method based on traditional navigation technology. This solution obtains scene data around the main vehicle (including lane information of the main vehicle, etc.) and blind spot warning, abnormal parking warning, and abnormal weather warning data around the main vehicle based on vehicle-road cloud technology. Finally, it presents and reminds on the navigation platform through digital twin technology, which can realize multi-lane display of the road and can display the position of the current vehicle and surrounding vehicles in real time on the lane; at the same time, the navigation data of the navigation platform can also be displayed simultaneously, so that users can have a more intuitive understanding of the scene conditions around the main vehicle and warning information. On the other hand, in order to display the positions of the current vehicle and surrounding vehicles on the lane in real time, there are strict requirements on the time required for parsing, calculating, and transmitting the vehicle-road cloud data, as well as the time required to construct and display the dynamic scene around the main vehicle. If the time is too long, the warning information and even the entire dynamic scene around the main vehicle will easily become invalid. Therefore, in order to ensure the timeliness of the dynamic scene around the main vehicle, this solution uses the vehicle-road cloud platform to perform simple primary calculations on the collaborative data to obtain primary calculation results, and a third-party service platform constructs the dynamic scene around the main vehicle based on the primary calculation results. Different platforms collaborate on calculations, and multiple platforms collaborate to optimize the division of labor. Instead of piling all the calculation work on a single platform, this effectively improves the computing efficiency and ensures the timeliness and effectiveness of the dynamic scene around the main vehicle. In addition, during the primary calculation process, the vehicle-road cloud platform will denoise the collaborative data according to the needs of the environmental information around the main vehicle and the construction of traffic warning information, to avoid the transmission and calculation of irrelevant data affecting the generation efficiency of the dynamic scene around the main vehicle.

[0015] This solution uses digital twin technology to uniformly map weather warnings (such as fog visibility models), traffic warnings (illegal parking / abnormal overtaking around the scene), and the lane information of the main vehicle to the navigation platform, realizing spatial and hierarchical information expression that traditional navigation (2D arrows + voice) cannot provide.

[0016] A further optimization solution is that the vehicle-road cloud platform collects collaborative data and performs primary calculations; including the following methods:

[0017] Collect and copy the collaborative data to obtain two sets of collaborative data, and perform the following parallel operations on the two sets of collaborative data:

[0018] For the first set of collaborative data: deleting invalid data and multi-source basic positioning data to obtain denoised collaborative data; the multi-source basic positioning data includes: roadside basic positioning data, main vehicle side basic positioning data and GPS side basic positioning data;

[0019] For the second set of collaborative data:

[0020] Extract multi-source basic positioning data and weather data respectively;

[0021] Perform spatiotemporal alignment and data synchronization on the multi-source basic positioning data to obtain synchronized multi-source basic positioning data; input the synchronized multi-source basic positioning data into the constructed multi-source lane positioning model to obtain the lane information of the main vehicle;

[0022] For weather data: perform warning judgment to obtain weather warning information;

[0023] The primary calculation results are denoised collaborative data, synchronized multi-source basic positioning data, the lane information of the main vehicle and weather warning information.

[0024] A further optimization scheme is to input the synchronized multi-source basic positioning data into the constructed multi-source lane positioning model to obtain the lane information of the host vehicle; including the following method:

[0025] Dynamically determine the participating calculation data based on the synchronized multi-source basic positioning data; the participating calculation data includes the roadside basic positioning data and the GPS side basic positioning data, or the main vehicle side basic positioning data and the GPS side basic positioning data;

[0026] Lane positioning fusion is performed based on the participating calculation data to obtain the main vehicle lane positioning information.

[0027] A further optimization scheme is to dynamically determine the data involved in the calculation based on the multi-source basic positioning data after spatiotemporal alignment and data synchronization; including the following method:

[0028] Obtain the occlusion area parameters of the main vehicle in the current traffic environment based on any multi-source basic positioning data;

[0029] A threshold value of the occlusion area parameter is preset. When the occlusion area parameter exceeds the threshold value, the basic positioning data on the main vehicle side and the basic positioning data on the GPS side are used as the data involved in the calculation; otherwise, the basic positioning data on the road side and the basic positioning data on the GPS side are used as the data involved in the calculation.

[0030] A further optimization scheme is to perform lane positioning fusion based on the participating calculation data to obtain the host vehicle lane positioning information; including the following method:

[0031] Preprocess the basic positioning data on the GPS side based on the Kalman filter algorithm;

[0032] Considering the motion characteristics of the main vehicle, a hidden Markov model based on lidar data and GPS data is established;

[0033] Based on the hidden Markov fusion model, the radar positioning data is fused with the pre-processed GPS side basic positioning data to obtain the main vehicle lane positioning information. The radar positioning data includes: the main vehicle side basic positioning data or the road side basic positioning data.

[0034] A further optimization scheme is that the method for constructing the hidden Markov fusion model includes:

[0035] The lane is used as the hidden state, and radar positioning data and GPS-based positioning data are used as the observed state. The timestamp of the observed state is used as the initial state probability distribution vector, and state transitions are performed based on the distance relationship between adjacent frames. The relative position, global position, and driving parameter similarity of the main vehicle are used as reference factors in the observation probability matrix to construct a hidden Markov fusion model. The observed state includes the main vehicle's speed and direction, while the hidden state includes the main vehicle's lane position and driving direction.

[0036] Dynamically optimize the state transition probability in the state transition matrix according to the confidence W:

[0037] ;

[0038] Among them, N1 represents the number of point clouds in the current frame; N2 represents the threshold of the number of valid inspection point clouds; K represents the point cloud number coefficient; and e represents the natural base.

[0039] A further optimization solution is that the third-party service platform constructs a dynamic scene around the main vehicle based on the primary calculation results, and displays the dynamic scene around the main vehicle in real time on the navigation platform based on digital twin technology; including the following methods:

[0040] Identify traffic objects around the main vehicle based on synchronized multi-source basic positioning data;

[0041] Obtain a map of the target road section and render a static scene based on Unity or Unreal Engine software; the static scene includes lane lines, shoulders, and green belts;

[0042] Constructing a dynamic scene in a static scene based on the lane information of the main vehicle, the traffic objects around the main vehicle, and the denoised collaborative data; the dynamic scene includes: a 3D model of the main vehicle and a model of the traffic objects around the main vehicle;

[0043] Weather warning information and traffic warning information are displayed separately in dynamic scenes.

[0044] A further optimization solution is that the weather warning information and traffic warning information are also displayed in real-time twin form in text or voice form.

[0045] A further optimization solution is that the dynamic scenes around the main vehicle are displayed in twin modes in full-screen mode or split-screen mode based on the navigation data of the navigation platform.

[0046] This solution also provides a navigation system based on vehicle-road cloud and digital twin technology, which is used to implement the above-mentioned navigation method based on vehicle-road cloud and digital twin technology; the system includes: a third-party service platform, a navigation platform and a vehicle-road cloud platform;

[0047] The third-party service platform is used to request the navigation platform to start navigation; it is also used to request to start the vehicle-road cloud platform after the navigation platform successfully starts navigation;

[0048] The navigation platform is used to display navigation data in real time;

[0049] The vehicle-road cloud platform is used to collect collaborative data and perform primary calculations, and also to send the results of these calculations to a third-party service platform. These primary calculations include determining weather warning information and the host vehicle's lane information, and denoising the collaborative data.

[0050] The third-party service platform is also used to construct a dynamic scene around the main vehicle based on the primary calculation results, and based on the digital twin technology, the dynamic scene around the main vehicle is displayed in real time on the navigation platform; the dynamic scene around the main vehicle includes: weather warning information, environmental information around the main vehicle and traffic warning information.

[0051] This solution also provides a computer-readable medium on which a computer program is stored. The computer program is executed by a processor to implement the above-mentioned navigation method based on vehicle-road cloud and digital twin technology.

[0052] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0053] 1. The present invention provides a navigation method, system, and medium based on vehicle-road cloud and digital twin technologies. Based on traditional navigation technology, this solution makes methodological improvements. Using vehicle-road cloud technology, this solution acquires scene data around the main vehicle (including lane information of the main vehicle, etc.), as well as blind spot warnings, abnormal parking warnings, and abnormal weather warnings around the main vehicle. Finally, digital twin technology is used to present and remind users on the navigation platform, enabling multi-lane display of the road and real-time display of the positions of the current vehicle and surrounding vehicles on the lanes. Simultaneously, navigation data from the navigation platform can also be displayed simultaneously, allowing users to more intuitively understand the scene conditions around the main vehicle and warning information.

[0054] 2. The navigation method, system and medium based on vehicle-road cloud and digital twin technology provided by the present invention; in order to ensure the timeliness of the dynamic scene around the main vehicle, the vehicle-road cloud platform performs simple primary calculations on the collaborative data to obtain primary calculation results, and the third-party service platform constructs the dynamic scene around the main vehicle based on the primary calculation results. Different platforms collaborate on calculations, and multiple platforms collaborate to optimize the division of labor; instead of piling all calculation work on a certain platform, the calculation efficiency is effectively improved, and the timeliness and effectiveness of the dynamic scene around the main vehicle are guaranteed. In addition, during the primary calculation process, the vehicle-road cloud platform will denoise the collaborative data according to the needs of the environmental information around the main vehicle and the traffic warning information, so as to avoid the transmission and calculation of irrelevant data affecting the generation efficiency of the dynamic scene around the main vehicle;

[0055] 3. The present invention provides a navigation method, system, and medium based on vehicle-road-cloud and digital twin technology. Through digital twin technology, weather warnings (such as foggy visibility models), traffic warnings (illegal parking / abnormal overtaking around the scene, etc.), and the lane information of the main vehicle are uniformly mapped to the navigation platform, realizing spatial and hierarchical information expression that traditional navigation (2D arrows + voice) cannot provide. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the following briefly introduces the drawings required for use in the examples. It should be understood that the following drawings only illustrate certain embodiments of the present invention and should not be considered as limiting the scope. A person of ordinary skill in the art can also derive other relevant drawings based on these drawings without inventive effort. In the drawings:

[0057] Figure 1 A flowchart of the navigation method based on vehicle-road-cloud and digital twin technology;

[0058] Figure 2 Schematic diagram of navigation data flow based on vehicle-road cloud and digital twin technology. DETAILED DESCRIPTION

[0059] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with examples and drawings. The exemplary embodiments of the present invention and their descriptions are only used to explain the present invention and are not intended to limit the present invention.

[0060] Although the traditional navigation interface can provide users with basic road directions, turn prompts and traffic conditions, it has problems such as single information presentation and poor interactivity. In view of this, this solution provides the following embodiments to solve the above technical problems.

[0061] Example 1: This example provides a navigation method based on vehicle-road-cloud and digital twin technology, which is implemented based on a third-party service platform, a navigation platform, and a vehicle-road-cloud platform; Figure 1 and Figure 2 As shown, the method includes:

[0062] Step 1: The third-party service platform requests the navigation platform to start navigation;

[0063] Step 2: After successfully starting navigation on the navigation platform, the navigation data is displayed in real time. At the same time, the third-party service platform requests to start the vehicle-road cloud platform.

[0064] Step 3: The vehicle-infrastructure cloud platform collects the collaborative data and performs preliminary calculations. The platform then sends the results of these calculations to a third-party service platform. These preliminary calculations include determining weather warning information and the vehicle's lane, and performing denoising on the collaborative data.

[0065] The vehicle-road cloud platform collects collaborative data and performs primary calculations; including methods:

[0066] S31, collect and copy the collaborative data to obtain two sets of collaborative data, and perform the following parallel operations on the two sets of collaborative data:

[0067] For the first set of collaborative data, invalid data and multi-source basic positioning data are deleted to obtain denoised collaborative data. The multi-source basic positioning data includes roadside basic positioning data, vehicle-side basic positioning data, and GPS-side basic positioning data. In this step, the GPS-side basic positioning data can be high-precision GPS positioning data on the vehicle or GPS on a smartphone. As the penetration rate of smartphones equipped with GPS functionality is currently as high as 93.3%, this solution integrates the smartphone, roadside, and vehicle-side data to achieve effective lane-level recognition.

[0068] For the second set of collaborative data: extract the multi-source basic positioning data and weather data respectively; perform spatiotemporal alignment and data synchronization on the multi-source basic positioning data to obtain synchronized multi-source basic positioning data; input the synchronized multi-source basic positioning data into the constructed multi-source lane positioning model to obtain the lane information of the main vehicle; in this step, based on the precision time protocol and the unified device clocks on the roadside and main vehicle sides, the time deviation is controlled within ±10ms, and the interpolation method is used to compensate for the delay of asynchronous data; establish a global coordinate system with the roadside or main vehicle side as the reference, and perform coordinate transformation on the data on the other sides.

[0069] The method of inputting the synchronous multi-source basic positioning data into the constructed multi-source lane positioning model to obtain the lane information of the host vehicle includes:

[0070] S311, dynamically determine participating calculation data based on synchronized multi-source basic positioning data; the participating calculation data includes roadside basic positioning data and GPS side basic positioning data, or host vehicle side basic positioning data and GPS side basic positioning data; this step specifically includes the following method:

[0071] S3111, obtaining the occlusion area parameters of the main vehicle in the current traffic environment based on any multi-source basic positioning data;

[0072] S3112, preset the occlusion area parameter threshold. When the occlusion area parameter exceeds the occlusion area parameter threshold, the main vehicle side basic positioning data and the GPS side basic positioning data are used as the participating calculation data; otherwise, the road side basic positioning data and the GPS side basic positioning data are used as the participating calculation data. Specifically, the occlusion area parameter can be a parameter that can characterize the occlusion area, such as area, volume, point cloud density, etc. For example, for the road side basic positioning data, Calculate the point cloud density change rate of the main vehicle in the current traffic environment. If the point cloud density is lower than the occlusion area parameter threshold (such as <50 points / square meter) for more than 3 frames, it is determined to be occlusion, and the basic positioning data of the main vehicle side and the basic positioning data of the GPS side are used as participating calculation data; for the basic positioning data of the main vehicle side; use the on-board camera + YOLOv8 to detect the bounding box size and position of the obstacles in front (such as trucks, buses); if the obstacle covers the central area of the image (such as the area ratio> 30% of the occlusion area parameter threshold), it is determined to be occlusion, and the basic positioning data of the main vehicle side and the basic positioning data of the GPS side are used as participating calculation data; for the basic positioning data of the GPS side, real-time monitoring Control the GPS signal-to-noise ratio (SNR), number of visible satellites, and horizontal dilution of precision (HDOP): If the SNR is less than the signal-to-noise ratio threshold of 20dB, the number of visible satellites is ≤ the number threshold of 4, and the horizontal dilution of precision (HDOP) is greater than the horizontal positioning accuracy threshold of 3.0, it is determined to be unobstructed, and the basic positioning data on the main vehicle side and the basic positioning data on the GPS side are used as the data for calculation. In addition, the real-time GPS positioning results can be matched with the obstruction marked areas in the high-precision map (such as tunnel entrances and underpasses). If the main vehicle position enters the pre-defined blind spot polygon on the map, the basic positioning data on the main vehicle side and the GPS side are triggered in advance as the data for calculation.

[0073] S312: Perform lane positioning fusion based on the participating calculation data to obtain the host vehicle lane positioning information. This step specifically includes the following methods:

[0074] S3121, preprocessing the basic positioning data on the GPS side based on the Kalman filter algorithm;

[0075] S3122: Considering the motion characteristics of the host vehicle, a hidden Markov model based on lidar data and GPS data is established;

[0076] S3123: Based on a hidden Markov fusion model, the radar positioning data is fused with the pre-processed GPS side basic positioning data to obtain the main vehicle lane positioning information, wherein the radar positioning data includes: the main vehicle side basic positioning data or the road side basic positioning data.

[0077] The method for constructing the hidden Markov fusion model includes:

[0078] The lane is used as the hidden state, and radar positioning data and GPS basic positioning data are used as the observed state. The timestamp of the observed state is used as the initial state probability distribution vector, and state transitions are performed based on the distance relationship between adjacent frames. The relative position of the main vehicle, the global position, and the similarity of driving parameters are used as reference factors in the observation probability matrix to construct a hidden Markov fusion model.

[0079] The observed state includes the driving speed and direction of the main vehicle; the hidden state includes the lane position and driving direction of the main vehicle;

[0080] The initial state probability distribution vector is: , where I represents the number of all laser point cloud data with the same timestamp as the observation state; the observation probability matrix is calculated based on the following formula:

[0081] ;

[0082] Among them, b i represents the observation probability under hidden state i; p1 represents the likelihood probability of the lateral deviation between the center of the main vehicle and the lane centerline under hidden state i; β1 represents the weight of p1; p2 represents the likelihood probability of the lateral position of the main vehicle when the GPS coordinates of the main vehicle are projected into the lane coordinate system under hidden state i; β2 represents the weight of p2; n represents the total number of hidden states; H i represents the speed similarity parameter between the radar positioning data and the GPS side basic positioning data in the hidden state i; S i Indicates the driving direction parameters of the radar positioning data and the basic positioning data on the GPS side in the hidden state i;

[0083] ;

[0084] ;

[0085] Where α represents the radar ranging error; d represents the lateral deviation between the center of the vehicle and the center line of the lane; L i represents the lateral coordinate of the lane centerline; γ represents the GPS horizontal positioning error; (x i ,y i ) represents the coordinates of the lane center reference point; (x, y) represents the GPS positioning coordinates of the main vehicle;

[0086] The radar positioning data and the basic GPS positioning data are derived from the same target. Therefore, the speeds of the LiDAR and GPS trajectories at the same time should be similar, so a corresponding speed similarity parameter is constructed. Similarly, the two data sets for the same target should contain consistent information. Therefore, the target's driving direction should be consistent along the road section, so a driving direction parameter is constructed.

[0087] ;

[0088] ;

[0089] Among them, v1 represents the driving speed of the main vehicle in the radar positioning data; v2 represents the driving speed of the main vehicle in the basic positioning data on the GPS side; F1 represents the driving direction of the main vehicle in the radar positioning data; F2 represents the driving direction of the main vehicle in the basic positioning data on the GPS side.

[0090] S3124, dynamically optimize the state transition probability in the state transition matrix according to the confidence W:

[0091] ;

[0092] Among them, N1 represents the number of point clouds in the current frame; N2 represents the threshold of the number of valid inspection point clouds; K represents the point cloud number coefficient; and e represents the natural base.

[0093] For weather data: perform early warning judgments to obtain weather warning information; specifically, pre-install LSTM prediction network models or random forest models to predict abnormal weather phenomena (such as the generation and dissipation time of abnormal weather such as fog) based on weather data obtained from the Meteorological Bureau API, roadside sensors (such as visibility detectors and temperature sensors), on-board sensors (such as cameras detecting rain and snow), and even user crowdsourced data. This allows the weather data to be integrated with the existing navigation system to improve driving safety.

[0094] Although collaborative data denoising, the calculation of the vehicle's lane information, and weather warning information are performed in parallel, collaborative data denoising is simple and therefore completed first, followed by weather warning information, and finally the vehicle's lane information. To ensure the effectiveness of the dynamic scene around the vehicle, the vehicle-road cloud platform can send the primary calculation results to the third-party service platform in the order of completion of collaborative data denoising, the calculation of the vehicle's lane information, and weather warning information. If collaborative data denoising is completed first, the denoised collaborative data will be sent first, and denoised collaborative data will also be prioritized when constructing the dynamic scene around the vehicle. In actual application, multi-source positioning mainly includes: the high-precision positioning source of the vehicle, roadside perception sources, and on-board perception sources. A lightweight multi-source lane positioning model is constructed and configured on the vehicle-road cloud platform, and the corresponding primary calculations are completed by the vehicle-road cloud platform.

[0095] S32 uses denoised collaborative data, synchronized multi-source basic positioning data, the lane information of the main vehicle, and weather warning information as primary calculation results.

[0096] Step 4: The third-party service platform constructs a dynamic scene around the main vehicle based on the primary calculation results, and uses digital twin technology to display the dynamic scene around the main vehicle in real time on the navigation platform. The dynamic scene around the main vehicle includes: weather warning information, environmental information around the main vehicle, and traffic warning information. This step specifically includes the following methods:

[0097] S41, identifying traffic objects around the main vehicle based on synchronized multi-source basic positioning data; this step specifically includes: clustering radar point clouds based on the DBSCAN algorithm to track different traffic objects (this can be achieved by relying on vehicle motion models and pedestrian motion models), independently running a Kalman filter on each traffic object, and fusing roadside radar data with vehicle-mounted radar data to identify different traffic objects;

[0098] S42, obtaining a map of the target road section and rendering a static scene based on Unity software or Unreal Engine software; the static scene includes: lane lines, road shoulders, and green belts;

[0099] S43, constructing a dynamic scene in the static scene based on the lane information of the host vehicle, the traffic objects around the host vehicle, and the denoised collaborative data; the dynamic scene includes: a 3D model of the host vehicle and a model of the traffic objects around the host vehicle;

[0100] S44: Differentiate and display the weather warning information and the traffic warning information in the dynamic scene.

[0101] Weather and traffic warning information is also displayed in real-time, in text or voice format. The vehicle's surrounding dynamic scene is displayed in full or split-screen mode, based on the navigation data from the navigation platform. In split-screen mode, the vehicle's surrounding dynamic scene and navigation data are displayed separately. In full-screen mode, the vehicle's surrounding dynamic scene is overlaid on the navigation data. Regardless of split-screen or full-screen mode, text-based weather and traffic warning information are displayed within the vehicle's surrounding dynamic scene.

[0102] This embodiment provides a navigation method based on vehicle-road cloud and digital twin technology. Based on the vehicle-road cloud technology, scene data around the main vehicle (including lane information of the main vehicle, etc.) as well as blind spot warning, abnormal parking warning, and abnormal weather warning data around the main vehicle are obtained, and finally presented and reminded on the navigation platform through digital twin technology. It can realize multi-lane display of the road and can display the position of the current vehicle and surrounding vehicles in real time on the lane; at the same time, the navigation data of the navigation platform can also be displayed simultaneously, so that users can more intuitively understand the scene conditions around the main vehicle and warning information. On the other hand, in order to display the positions of the current vehicle and surrounding vehicles on the lane in real time, there are strict requirements on the time required for parsing, calculating, and transmitting the vehicle-road cloud data, as well as the time required for constructing and displaying the dynamic scene around the main vehicle. If the time is too long, the warning information and even the entire dynamic scene around the main vehicle will easily become invalid. Therefore, in order to ensure the timeliness of the dynamic scene around the main vehicle, this solution uses the vehicle-road cloud platform to perform simple primary calculations on the collaborative data to obtain primary calculation results, and a third-party service platform constructs the dynamic scene around the main vehicle based on the primary calculation results. Different platforms collaborate on calculations, and multiple platforms collaborate to optimize the division of labor; instead of piling all the calculation work on a certain platform, the calculation efficiency is effectively improved, and the timeliness and effectiveness of the dynamic scene around the main vehicle are guaranteed. In addition, during the primary calculation process, the vehicle-road cloud platform will denoise the collaborative data according to the needs of the construction of the environmental information around the main vehicle and the traffic warning information, to avoid the transmission and calculation of irrelevant data affecting the generation efficiency of the dynamic scene around the main vehicle.

[0103] This embodiment uses digital twin technology to uniformly map weather warnings (such as fog visibility models), traffic warnings (illegal parking / abnormal overtaking around the scene, etc.), and the lane information of the main vehicle to the navigation platform, realizing spatial and hierarchical information expression that traditional navigation (2D arrows + voice) cannot provide.

[0104] Example 2: This example provides a navigation system based on vehicle-road cloud and digital twin technology, which is used to implement the navigation method based on vehicle-road cloud and digital twin technology described in Example 1; the system includes: a third-party service platform, a navigation platform, and a vehicle-road cloud platform;

[0105] The third-party service platform is used to request the navigation platform to start navigation; it is also used to request to start the vehicle-road cloud platform after the navigation platform successfully starts navigation;

[0106] The navigation platform is used to display navigation data in real time;

[0107] The vehicle-road cloud platform is used to collect collaborative data and perform primary calculations, and also to send the results of these calculations to a third-party service platform. These primary calculations include determining weather warning information and the host vehicle's lane information, and denoising the collaborative data.

[0108] The third-party service platform is also used to construct dynamic scenes around the main vehicle based on the primary calculation results, and to display the dynamic scenes around the main vehicle in real time on the navigation platform based on digital twin technology; the dynamic scenes around the main vehicle include: weather warning information, environmental information around the main vehicle and traffic warning information.

[0109] Embodiment 3: This embodiment provides a computer-readable medium having a computer program stored thereon. The computer program is executed by a processor to implement the navigation method based on vehicle-road cloud and digital twin technology as described in Embodiment 1. Specifically, the following steps are performed:

[0110] Step 1: The third-party service platform requests the navigation platform to start navigation;

[0111] Step 2: After successfully starting navigation on the navigation platform, the navigation data is displayed in real time. At the same time, the third-party service platform requests to start the vehicle-road cloud platform.

[0112] Step 3: The vehicle-infrastructure cloud platform collects the collaborative data and performs preliminary calculations. The platform then sends the results of these calculations to a third-party service platform. These preliminary calculations include determining weather warning information and the vehicle's lane, and performing denoising on the collaborative data.

[0113] Step 4: The third-party service platform constructs a dynamic scene around the main vehicle based on the primary calculation results, and displays the dynamic scene around the main vehicle in real time on the navigation platform based on digital twin technology; the dynamic scene around the main vehicle includes: weather warning information, environmental information around the main vehicle and traffic warning information.

[0114] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. The navigation method based on vehicle-road-cloud and digital twin technology is characterized by: Implemented based on a third-party service platform, a navigation platform, and a vehicle-road cloud platform; the method includes: The third-party service platform requests the navigation platform to start navigation; After the navigation platform successfully starts navigation, the navigation data is displayed in real time, and at the same time, the third-party service platform requests to start the vehicle-road cloud platform; The vehicle-infrastructure cloud platform collects collaborative data and performs primary calculations, which it then sends to a third-party service platform. These primary calculations include determining weather warning information and the vehicle's lane, and performing denoising on the collaborative data. The vehicle-road cloud platform collects collaborative data and performs primary operations; including methods: S31, collect and copy the collaborative data to obtain two sets of collaborative data, and perform the following parallel operations on the two sets of collaborative data: For the first set of collaborative data: deleting invalid data and multi-source basic positioning data to obtain denoised collaborative data; the multi-source basic positioning data includes: roadside basic positioning data, main vehicle side basic positioning data and GPS side basic positioning data; For the second set of collaborative data: extract the multi-source basic positioning data and weather data respectively; perform spatiotemporal alignment and data synchronization on the multi-source basic positioning data to obtain synchronized multi-source basic positioning data; input the synchronized multi-source basic positioning data into the constructed multi-source lane positioning model to obtain the lane information of the host vehicle; The method of inputting the synchronous multi-source basic positioning data into the constructed multi-source lane positioning model to obtain the lane information of the host vehicle includes: S311, dynamically determining participating calculation data based on synchronized multi-source basic positioning data; the participating calculation data includes roadside basic positioning data and GPS-side basic positioning data, or vehicle-side basic positioning data and GPS-side basic positioning data; the dynamically determining participating calculation data based on synchronized multi-source basic positioning data includes: S3111, obtaining the occlusion area parameters of the main vehicle in the current traffic environment based on any multi-source basic positioning data; S3112: Preset an occlusion area parameter threshold. When the occlusion area parameter exceeds the occlusion area parameter threshold, the main vehicle side basic positioning data and the GPS side basic positioning data are used as participating calculation data; otherwise, the road side basic positioning data and the GPS side basic positioning data are used as participating calculation data; S312: Perform lane positioning fusion based on the participating calculation data to obtain lane positioning information of the host vehicle; specifically, the method includes: S3121, preprocessing the basic positioning data on the GPS side based on the Kalman filter algorithm; S3122: Considering the motion characteristics of the host vehicle, a hidden Markov model based on lidar data and GPS data is established; S3123: Based on a hidden Markov fusion model, the radar positioning data is fused with the pre-processed GPS-side basic positioning data to obtain lane positioning information of the host vehicle. The radar positioning data includes: the host vehicle-side basic positioning data or the road-side basic positioning data. The method for constructing the hidden Markov fusion model includes: The lane is used as the hidden state, and radar positioning data and GPS-based positioning data are used as the observed state. The timestamp of the observed state is used as the initial state probability distribution vector, and state transitions are performed based on the distance relationship between adjacent frames. The relative position, global position, and driving parameter similarity of the main vehicle are used as reference factors in the observation probability matrix to construct a hidden Markov fusion model. The observed state includes the main vehicle's speed and direction, while the hidden state includes the main vehicle's lane position and driving direction. S3124, dynamically optimize the state transition probability in the state transition matrix according to the confidence W: ； Among them, N1 represents the number of point clouds in the current frame; N2 represents the threshold of the number of valid inspection point clouds; K represents the point cloud number coefficient; e represents the natural base; For weather data: perform warning judgment to obtain weather warning information; S32, using the denoised collaborative data, the synchronized multi-source basic positioning data, the lane information of the host vehicle, and the weather warning information as the primary calculation results; The third-party service platform constructs a dynamic scene around the main vehicle based on the primary calculation results, and displays the dynamic scene around the main vehicle in real time on the navigation platform based on digital twin technology; the dynamic scene around the main vehicle includes: weather warning information, environmental information around the main vehicle and traffic warning information.

2. The navigation method based on vehicle-road cloud and digital twin technology according to claim 1 is characterized in that: The third-party service platform constructs a dynamic scene around the main vehicle based on the primary calculation results, and displays the dynamic scene around the main vehicle in real time on the navigation platform based on digital twin technology; Included methods: Identify traffic objects around the main vehicle based on synchronized multi-source basic positioning data; Obtain a map of the target road section and render a static scene based on Unity or Unreal Engine software; The static scene includes: lane lines, road shoulders and green belts; Constructing a dynamic scene in a static scene based on the lane information of the main vehicle, the traffic objects around the main vehicle, and the denoised collaborative data; the dynamic scene includes: a 3D model of the main vehicle and a model of the traffic objects around the main vehicle; Weather warning information and traffic warning information are displayed separately in dynamic scenes.

3. The navigation method based on vehicle-road cloud and digital twin technology according to claim 1 is characterized in that: The weather warning information and traffic warning information are also displayed in real-time twin mode in text or voice form; the dynamic scene around the main vehicle is displayed in twin mode in full-screen mode or split-screen mode based on the navigation data of the navigation platform.

4. The navigation system based on vehicle-road-cloud and digital twin technology is characterized by: Used to implement the navigation method based on vehicle-road-cloud and digital twin technology as described in any one of claims 1-3; the system includes: a third-party service platform, a navigation platform and a vehicle-road-cloud platform; The third-party service platform is used to request the navigation platform to start navigation; it is also used to request to start the vehicle-road cloud platform after the navigation platform successfully starts navigation; The navigation platform is used to display navigation data in real time; The vehicle-road cloud platform is used to collect collaborative data and perform primary calculations, and also to send the results of these calculations to a third-party service platform. These primary calculations include determining weather warning information and the host vehicle's lane information, and denoising the collaborative data. The third-party service platform is also used to construct a dynamic scene around the main vehicle based on the primary calculation results, and based on the digital twin technology, the dynamic scene around the main vehicle is displayed in real time on the navigation platform; the dynamic scene around the main vehicle includes: weather warning information, environmental information around the main vehicle and traffic warning information.

5. A computer-readable medium having a computer program stored thereon, characterized in that: The computer program is executed by a processor to implement the navigation method based on vehicle-road cloud and digital twin technology as described in any one of claims 1 to 3.

Images