Highway three-dimensional reconstruction method and system based on GNSS high-precision positioning and orientation

Abstract

The invention relates to the technical field of intelligent traffic, and discloses an expressway three-dimensional reconstruction method and system based on GNSS high-precision positioning and orientation, and the method comprises the following steps: 1, obtaining the real-time position and speed information of a vehicle through employing at least one GNSS receiver with advanced signal processing capability, high positioning precision can be maintained under the coverage of urban canyons and trees; 2, at least one I MU is utilized, the IMU adopts a machine learning algorithm to conduct self-calibration so as to adapt to dynamic changes in long-time operation, and future drift of the sensor can be predicted and compensated. Through an advanced GNSS receiver and a multi-frequency multi-system receiving technology, the system can realize high-precision positioning even in complex environments such as urban canyons and tree coverage, the I MU is equipped with a machine learning algorithm, self-calibration can be realized, dynamic change can be adapted, future sensor drift can be predicted and compensated, and long-term stable operation is ensured.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a method and system for three-dimensional reconstruction of a highway based on GNSS high-precision positioning and orientation. Background Art

[0002] With the continuous improvement of transportation infrastructure construction and the development of intelligent transportation systems, the demand for refined management of highways and high-precision maps is growing. Highway 3D reconstruction is a technology that collects and processes geographic information data and uses these data to generate a 3D model of the highway. This technology helps to improve the accuracy and efficiency of traffic management, road planning and vehicle navigation. The method and system based on GNSS (Global Navigation Satellite System) high-precision positioning and orientation can provide more accurate location information, thereby improving the quality and accuracy of 3D reconstruction.

[0003] Traditional highway 3D reconstruction methods mainly rely on laser scanners (L iDAR) or photogrammetry technology, which cannot achieve high accuracy and cannot guarantee long-term operation, but have limitations such as high equipment cost and long data acquisition and processing cycle. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides a method and system for three-dimensional reconstruction of highways based on GNSS high-precision positioning and orientation, which solves the problem that traditional three-dimensional reconstruction methods of highways mainly rely on laser scanners (L iDAR) or photogrammetry technology, cannot achieve high accuracy and cannot guarantee long-term operation.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: a highway three-dimensional reconstruction method based on GNSS high-precision positioning and orientation, comprising the following steps:

[0006] Step 1: Use at least one GNSS receiver with advanced signal processing capabilities to obtain real-time position and velocity information of the vehicle and maintain high positioning accuracy in urban canyons and tree cover;

[0007] Step 2: Utilize at least one IMU, wherein the IMU uses a machine learning algorithm to self-calibrate to adapt to dynamic changes during long-term operation and to be able to predict and compensate for future drift of the sensor;

[0008] Step 3: Collecting images and point cloud data captured by at least one vehicle-mounted camera or laser radar, wherein the camera is equipped with an intelligent exposure control system;

[0009] Step 4: Fuse the data in the above steps and use advanced multi-sensor data fusion algorithms to achieve accurate positioning and attitude estimation of the vehicle;

[0010] Step 5: Use a 3D modeling algorithm based on feature extraction and matching to convert the fused data in step 4 into a 3D model of the highway and its surrounding environment.

[0011] Preferably, in step 4, the positioning accuracy can be maintained by using visual inertial odometry (VIO) technology when GPS signals are unavailable.

[0012] Preferably, in step one, the GNSS receiver adopts multi-frequency and multi-system receiving technology to improve signal tracking capability and positioning accuracy in complex environments.

[0013] Preferably, in step three, the lidar uses advanced signal processing technology to improve detection capabilities in rain, fog and other low visibility conditions.

[0014] Preferably, in step five, the three-dimensional modeling algorithm further includes an adaptive learning mechanism that can learn from continuous measurement data and improve the accuracy and details of the model.

[0015] Preferably, in step 4, vehicle traffic patterns and historical maintenance records are combined to achieve highway service life prediction and maintenance plan optimization.

[0016] Preferably, the vehicle-mounted camera or lidar in step three is also used for dynamic traffic monitoring to capture vehicle behavior and traffic status in real time, providing real-time data support for traffic management.

[0017] A system for a highway three-dimensional reconstruction method based on GNSS high-precision positioning and orientation includes a central processing unit, which includes a communication unit and a user interface. The unit is responsible for coordinating the synchronous collection of data from various sensors, processing a multi-sensor fusion algorithm, and realizing real-time three-dimensional modeling.

[0018] Preferably, the central processing unit is connected to a communication unit, and the communication unit is used to upload the fused data and three-dimensional model to a cloud server for remote monitoring and data analysis.

[0019] Preferably, the central processing unit includes a user interface for displaying the three-dimensional model and traffic analysis results, and allowing users to perform interactive queries and export data.

[0020] The present invention provides a highway 3D reconstruction method and system based on GNSS high-precision positioning and orientation, which has the following beneficial effects:

[0021] 1. The present invention uses advanced GNSS receivers and multi-frequency multi-system receiving technology. The system can achieve high-precision positioning even in complex environments such as urban canyons and tree cover. The IMU is equipped with a machine learning algorithm that can self-calibrate and adapt to dynamic changes, predict and compensate for future sensor drift, and ensure long-term stable operation.

[0022] 2. The present invention combines data from GNSS, IMU, cameras or lidar through a multi-sensor fusion algorithm to accurately locate the vehicle and estimate its posture. It can also maintain positioning accuracy through visual inertial odometry (VIO) technology even when the GPS signal is lost.

[0023] 3. The present invention improves the performance of the system under different lighting and visibility conditions through an intelligent exposure control system and advanced lidar signal processing technology, ensuring that it can work effectively even in low visibility conditions such as rain and fog. By using a three-dimensional modeling algorithm of feature extraction and matching, combined with an adaptive learning mechanism, the system can continuously improve the accuracy and details of the model, and support the optimization of service life prediction and maintenance plan.

[0024] 4. The present invention realizes three-dimensional reconstruction and real-time traffic monitoring of highways through the system, captures vehicle behavior and traffic status, and provides strong data support for traffic management. The central processing unit cooperates with various sensors to synchronously collect data, and uploads it to the cloud through the communication module to realize remote monitoring and analysis of data. The user interface displays the three-dimensional model and traffic analysis results, supports users to conduct interactive queries and export data, and improves user experience and data practicality. DETAILED DESCRIPTION

[0025] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0026] Embodiment 1:

[0027] The embodiment of the present invention provides a method for three-dimensional reconstruction of a highway based on GNSS high-precision positioning and orientation, comprising the following steps:

[0028] Step 1: Use at least one GNSS receiver with advanced signal processing capabilities to obtain real-time position and velocity information of the vehicle and maintain high positioning accuracy in urban canyons and tree cover;

[0029] Step 2: Utilize at least one IMU, wherein the IMU uses a machine learning algorithm to self-calibrate to adapt to dynamic changes during long-term operation and to be able to predict and compensate for future drift of the sensor;

[0030] Step 3: Collecting images and point cloud data captured by at least one vehicle-mounted camera or laser radar, wherein the camera is equipped with an intelligent exposure control system;

[0031] Step 4: Fuse the data in the above steps and use advanced multi-sensor data fusion algorithms to achieve accurate positioning and attitude estimation of the vehicle;

[0032] Step 5: Use a 3D modeling algorithm based on feature extraction and matching to convert the fused data in step 4 into a 3D model of the highway and its surrounding environment.

[0033] In step 4, visual inertial odometry (VIO) technology can be used to maintain positioning accuracy when GPS signals are unavailable.

[0034] Specifically, visual sensors: The camera continuously captures a sequence of images of the surrounding environment. By identifying feature points between consecutive frames, the relative motion between adjacent frames can be estimated. The matching of these feature points allows the system to calculate the movement path of the camera.

[0035] Inertial Measurement Unit (IMU): An IMU is a device that contains an accelerometer and a gyroscope that can measure the linear acceleration and angular velocity of a device. IMU data provides instant information about the device's motion, which can be used to predict changes in the device's position and attitude.

[0036] In one embodiment, in tunnels, urban canyons, or other environments where GPS signals are unreliable, VIO technology can maintain the device's positioning capability, ensuring the continuous operation of the system and the integrity of the data. By fusing visual and inertial data, VIO provides more accurate positioning information than using IMU or cameras alone, reduces cumulative errors, and can offset the impact of sensor noise to a certain extent.

[0037] In step one, the GNSS receiver uses multi-frequency and multi-system reception technology to improve signal tracking capabilities and positioning accuracy in complex environments.

[0038] In one embodiment, multi-frequency reception technology: GNSS systems such as GPS, Galileo, BeiDou, and GLONASS usually provide signals in multiple frequency bands. Multi-frequency receivers are able to simultaneously receive signals of multiple frequencies from the same satellite. Signals of different frequencies help to more accurately correct the delay effect of the ionosphere, because the ionosphere affects signals of different frequencies differently. By comparing these signals, the receiver can calculate a more accurate position.

[0039] In step three, the lidar uses advanced signal processing techniques to improve detection capabilities in rain, fog and other low-visibility conditions.

[0040] In one embodiment, pulse coding: By encoding the transmitted laser pulses, the signal-to-noise ratio (SNR) can be improved by decoding them when they are received. This allows the system to more effectively distinguish between valid signals and background noise, especially in low-visibility environments.

[0041] In step five, the 3D modeling algorithm further includes an adaptive learning mechanism that can learn from continuous measurement data and improve the accuracy and details of the model.

[0042] In one embodiment, the adaptive learning mechanism is typically data-driven, meaning it relies on a large amount of input data to train and adjust internal parameters. In the case of 3D modeling, this data may come from historical measurements, lidar scans updated in real time, or other sensor information.

[0043] In step 4, the vehicle traffic patterns and historical maintenance records are combined to achieve the service life prediction and maintenance plan optimization of the highway.

[0044] In one embodiment, this integrated analysis and prediction method can more effectively manage highway assets, extend road service life, and reduce safety risks and maintenance costs due to road damage. This also facilitates more sustainable infrastructure management, improving the efficiency and effectiveness of maintenance work through accurate prediction and optimized resource allocation.

[0045] The vehicle-mounted camera or lidar in step three is also used for dynamic traffic monitoring, capturing vehicle behavior and traffic status in real time, and providing real-time data support for traffic management.

[0046] In one embodiment, through the application of these technologies, vehicle-mounted cameras and lidar play an important role in dynamic traffic monitoring, which can not only improve traffic efficiency and safety, but also promote the development of intelligent transportation systems and provide strong technical support for modern urban traffic management.

[0047] Embodiment 2:

[0048] Based on the previous embodiment, this embodiment provides a system for a highway three-dimensional reconstruction method based on GNSS high-precision positioning and orientation, including a central processing unit, which includes a communication unit and a user interface. The unit is responsible for coordinating the synchronous collection of data from various sensors, processing multi-sensor fusion algorithms, and realizing real-time three-dimensional modeling.

[0049] In one embodiment, through real-time monitoring and analysis of traffic flow, traffic can be effectively managed, congestion can be reduced, and traffic efficiency can be improved. The three-dimensional model and traffic analysis results can provide data support for traffic planning and decision-making, such as where to build new roads and how to optimize the existing traffic network. By uploading data and models to a cloud server, remote monitoring and data analysis can be conveniently performed without being restricted by geographical location.

[0050] The central processing unit is connected to the communication unit, which is used to upload the fused data and three-dimensional model to the cloud server for remote monitoring and data analysis.

[0051] The central processing unit includes a user interface for displaying the three-dimensional model and traffic analysis results, and allowing users to perform interactive queries and export data.

[0052] In one embodiment, through real-time three-dimensional modeling and traffic analysis, traffic problems such as congestion and accidents can be discovered in a timely manner, and corresponding measures can be taken. Through real-time monitoring and analysis of traffic flow, traffic can be effectively managed, congestion can be reduced, and traffic efficiency can be improved. The three-dimensional model and traffic analysis results can provide data support for traffic planning and decision-making. By uploading data and models to a cloud server, remote monitoring and data analysis can be conveniently carried out without being restricted by geographical location.

[0053] Comparative Example 1:

[0054] First, a mobile device (such as a vehicle) equipped with a GNSS receiver drives on the highway, collecting positioning data and time data at the same time. This data includes longitude, latitude, altitude and timestamp. At the same time, the on-board camera also collects image data of the road surface. This image data can be used to identify features of the road surface, such as road signs and lane lines. The GNSS data and image data will be sent to the central processing unit for processing. The main purpose of the processing is to match and fuse the GNSS data and image data to generate the corresponding 3D model. This process usually requires the use of some computer vision and machine learning algorithms, such as feature extraction, feature matching, and deep learning. The processed data will be used to generate a 3D model of the highway. This model is usually a point cloud model, where each point represents a geographical location, and the color and shape of the point represent the terrain and road surface characteristics of that location. Finally, the 3D model will be stored in the database, and users can view and analyze this model through GIS software to understand the actual situation of the highway.

[0055]

[0056] Table 1

[0057] As shown in Table 1, Example 1 provides a comprehensive highway 3D reconstruction method, which has the advantages of high positioning accuracy, strong environmental adaptability, self-calibration ability, excellent low visibility performance and high-precision 3D modeling, and has a significant role in promoting the development of modern transportation infrastructure management and intelligent transportation systems.

[0058] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A highway 3D reconstruction method based on GNSS high-precision positioning and orientation, characterized in that: The following steps are involved: Step 1: Use at least one GNSS receiver with advanced signal processing capabilities to obtain real-time position and velocity information of the vehicle and maintain high positioning accuracy in urban canyons and tree cover; Step 2: Utilize at least one IMU, wherein the IMU uses a machine learning algorithm to self-calibrate to adapt to dynamic changes during long-term operation and to be able to predict and compensate for future drift of the sensor; Step 3: Collecting images and point cloud data captured by at least one vehicle-mounted camera or laser radar, wherein the camera is equipped with an intelligent exposure control system; Step 4: Fuse the data in the above steps and use advanced multi-sensor data fusion algorithms to achieve accurate positioning and attitude estimation of the vehicle; Step 5: Use a 3D modeling algorithm based on feature extraction and matching to convert the fused data in step 4 into a 3D model of the highway and its surrounding environment.

2. The method for 3D reconstruction of a highway based on GNSS high-precision positioning and orientation according to claim 1, characterized in that: In the step 4, the positioning accuracy can be maintained by using the visual inertial odometry (VIO) technology when the GPS signal is unavailable.

3. The highway 3D reconstruction method based on GNSS high-precision positioning and orientation according to claim 1 is characterized in that: In step 1, the GNSS receiver uses multi-frequency and multi-system receiving technology to improve signal tracking capability and positioning accuracy in complex environments.

4. The method for 3D reconstruction of a highway based on GNSS high-precision positioning and orientation according to claim 1, characterized in that: In step three, the lidar uses advanced signal processing technology to improve detection capabilities in rain, fog and other low visibility conditions.

5. The method for 3D reconstruction of a highway based on GNSS high-precision positioning and orientation according to claim 1, characterized in that: In step 5, the three-dimensional modeling algorithm further includes an adaptive learning mechanism that can learn from continuous measurement data and improve the accuracy and details of the model.

6. The method for 3D reconstruction of a highway based on GNSS high-precision positioning and orientation according to claim 1, characterized in that: In step 4, vehicle traffic patterns and historical maintenance records are combined to achieve highway service life prediction and maintenance plan optimization.

7. The method for 3D reconstruction of a highway based on GNSS high-precision positioning and orientation according to claim 1, characterized in that: The vehicle-mounted camera or lidar in step three is also used for dynamic traffic monitoring, capturing vehicle behavior and traffic status in real time, and providing real-time data support for traffic management.

8. A system for a highway three-dimensional reconstruction method based on GNSS high-precision positioning and orientation, comprising a central processing unit according to any one of claims 1 to 7, characterized in that: The central processing unit includes a communication unit and a user interface. The unit is responsible for coordinating the synchronous collection of data from various sensors, processing multi-sensor fusion algorithms, and realizing real-time three-dimensional modeling.

9. The system of the highway 3D reconstruction method based on GNSS high-precision positioning and orientation according to claim 8, characterized in that: The central processing unit is connected to a communication unit, and the communication unit is used to upload the fused data and the three-dimensional model to a cloud server for remote monitoring and data analysis.

10. The system of the highway 3D reconstruction method based on GNSS high-precision positioning and orientation according to claim 8, characterized in that: The central processing unit includes a user interface for displaying the three-dimensional model and traffic analysis results, and allowing users to perform interactive queries and export data.

Images