Expressway vehicle centimeter-level trajectory tracking system based on 5G-U Beidou fusion

Abstract

The invention discloses an expressway vehicle centimeter-level trajectory tracking system based on 5G-U Beidou fusion, and relates to the technical field of 5G-Beidou traffic trajectory tracking, and the system comprises a multi-source heterogeneous data collection module which collects Beidou positioning, 5G vehicle state and visual data; the multi-frequency multi-constellation fusion positioning module is used for improving the algorithm and improving the ionosphere correction precision; the dynamic environment perception and error correction module is used for constructing a dynamic electronic fence and optimizing an error correction model; the centimeter-level track generation and visualization module is used for improving an interpolation algorithm and smoothing a track; and the cloud collaboration and edge computing module adopts federal learning to realize model collaboration optimization, and also relates to a plurality of innovation modules such as safety guarantee and intelligent service. The system integrates 5G-U and Beidou technologies, realizes centimeter-level high-precision positioning, can dynamically sense the environment and correct errors, improves the data transmission processing efficiency and safety, integrates semantic enhancement, intelligent service and visualization functions, and assists the intelligent development of expressways.

Description

Technical Field

[0001] The present invention relates to the field of 5G-Beidou traffic trajectory tracking technology, and in particular to a centimeter-level trajectory tracking system for highway vehicles based on 5G-U Beidou fusion. Background Art

[0002] With the rapid development of smart transportation, highway vehicle trajectory tracking technology is crucial for traffic safety management, traffic flow optimization, and the provision of intelligent services. Traditional vehicle trajectory tracking primarily relies on a single global navigation satellite system (GNSS), such as Beidou or GPS. However, satellite signals are susceptible to factors such as weather, terrain, and building obstruction. In complex environments such as mountainous areas and urban overpasses, signal loss and reduced positioning accuracy can occur, making it difficult to meet the centimeter-level positioning requirements for vehicles on highways. Furthermore, the low data update frequency of a single satellite system prevents it from accurately reflecting the dynamic driving status of vehicles in real time, limiting the ability to track vehicles in detail.

[0003] With the application of 5G technology, cellular network-based vehicle positioning has brought new directions to trajectory tracking. However, 5G faces problems such as limited coverage and insufficient signal penetration in highway scenarios. This is especially true in remote sections of road or in special areas such as tunnels, where continuous and stable signal transmission is difficult to ensure. Furthermore, 5G positioning alone still cannot achieve centimeter-level accuracy, failing to meet the urgent need for high-precision positioning in applications such as autonomous vehicle trajectory monitoring and accurate accident tracing. Furthermore, existing 5G and Beidou integration solutions often remain at the simple data overlay level, lacking the deep integration and coordinated processing of multi-source data, making it difficult to fully leverage the advantages of both.

[0004] In terms of trajectory tracking applications, existing systems generally lack the ability to perceive and adapt to dynamic environmental changes. Highway environments are complex and ever-changing, with factors such as vehicle driving status, surrounding traffic conditions, and weather conditions constantly changing. However, traditional systems are unable to detect these changes and correct trajectory tracking errors in a timely manner. Furthermore, existing technologies have significant shortcomings in data security, intelligent service generation, and visualization, failing to meet the comprehensive safety, functionality, and usability requirements of smart highway vehicle trajectory tracking systems. Summary of the Invention

[0005] The present invention proposes a centimeter-level trajectory tracking system for highway vehicles based on 5G-U Beidou fusion to solve the problems mentioned in the above-mentioned prior art.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions: A centimeter-level trajectory tracking system for highway vehicles based on 5G-U BeiDou fusion, including the following modules: Multi-source heterogeneous data acquisition module: obtains vehicle original positioning data through BeiDou-3 satellite receiver ,in, are three-dimensional coordinates, Timestamp; deployed on 5G-U roadside units to collect vehicle driving status data , where v is velocity, a is acceleration, is the heading angle, is the pitch angle; the vehicle visual feature data is obtained using the roadside camera , build a multi-source data set ; Multi-frequency and multi-constellation fusion positioning module: using improved ionospheric-free combined observation equation Calculate the ionospheric-free combination observation values ​​of BeiDou B1C, B2a and GPS L1, L5 bands. The equation adds an ionospheric residual correction term on the basis of the traditional ionospheric-free combination. , is the ionospheric residual correction coefficient; Dynamic environment perception and error correction module: building dynamic electronic fences on highways , through the space-time distance formula Determine whether the vehicle has crossed the boundary, where is the spatial position, is the velocity component, is the time interval; Centimeter-level trajectory generation and visualization module: based on improved B-spline interpolation algorithm Generate a smooth trajectory, where d i is the control point, is a p-order canonical B-spline basis function that satisfies the endpoint interpolation condition , through the constrained optimization algorithm Identify control points; Cloud collaboration and edge computing modules: Building a distributed computing system under the federated learning architecture , through the knowledge distillation algorithm To achieve model collaborative optimization, the algorithm transforms the teacher model The knowledge is distilled to the student model through the temperature parameter T , while protecting privacy, the accuracy of the edge node model is improved.

[0007] Furthermore, it also includes a multi-frequency and multi-constellation fusion positioning module: using an adaptive robust Kalman filter algorithm , where the robustness gain , is the weight matrix, , is the residual, To achieve the mean error, the window size is dynamically adjusted through the window sliding mechanism. , the mechanism is based on the residual variance Automatically adjust the filter window.

[0008] Furthermore, it also includes a dynamic environment perception and error correction module: using an improved multipath error correction model , this model adds a direction-sensitive term based on the traditional exponential decay model , where D and E are new parameters, , For reference value.

[0009] Furthermore, the 5G-U roadside unit in the multi-source heterogeneous data acquisition module adopts spatial modulation multi-input multi-output technology, through the formula Calculate channel capacity by activating different antennas to transmit additional information, deployed at intervals along the highway Satisfy the formula ,in As a safety factor, this formula adds the speed standard deviation to the traditional calculation based on vehicle speed and update cycle. .

[0010] Furthermore, the centimeter-level trajectory generation and visualization module uses the space-time coordinate conversion formula Convert WGS-84 coordinates to the local space-time coordinate system, where T st is a 4×4 spatiotemporal transformation matrix containing rotation, translation, and time scaling parameters ,The parameters are estimated by nonlinear least squares algorithm.,The space-time transformation model introduces the time dimension based on the traditional three-dimensional space,transformation.

[0011] Furthermore, the cloud collaboration and edge computing module designs a dynamic resource allocation algorithm , the constraints are , where w i is the task weight, r ij Allocate benefits to resources and optimize through reinforcement learning algorithms , the algorithm models the resource allocation problem as a Markov decision process by maximizing the long-term cumulative reward Realize dynamic resource allocation.

[0012] Furthermore, it also includes a spatiotemporal semantic enhancement module: building a spatiotemporal semantic model for highways ,in is a semantic object, For object relations, For time reasons, Design semantic enhancement positioning algorithm for spatial relationship , where the semantic correction is the semantic feature, w i is the weight, f i is the mapping function.

[0013] Furthermore, it also includes a multi-dimensional security assurance module: building a multi-dimensional security assessment model ,in Security indicators for each dimension, weight; design a security risk prediction algorithm ,in is the safety index change rate, For historical security records, predict future security risks through long short-term memory networks.

[0014] Furthermore, it also includes an intelligent service generation module: building a service demand understanding model ,in For user queries, To understand user intent using the pre-trained language model BERT for contextual information, we design trajectory prediction and service matching algorithms. ,in is the prediction trajectory based on LSTM, Serving candidates, For user preferences, matching services is performed through cosine similarity.

[0015] Furthermore, it also includes a digital twin visualization module: using a multi-resolution rendering algorithm , where R i Rendering results for different resolutions, is the distance weight function , d is the observation distance, d i The threshold for resolution switching is used. The algorithm dynamically adjusts the rendering resolution according to the observation distance, thereby improving the rendering frame rate while ensuring the visual effect.

[0016] Compared with the existing technology, the beneficial effects of the present invention are: By innovatively integrating 5G-U and Beidou technologies and deeply processing multi-source heterogeneous data, the system can achieve centimeter-level high-precision positioning even in complex environments. Whether it is signal-blocked sections in mountainous areas or elevated bridge areas in cities, it can accurately capture the vehicle's position, effectively solving the problem of inaccurate positioning caused by traditional single technology.

[0017] The system boasts powerful dynamic environmental perception and error correction capabilities, enabling real-time detection of vehicle driving status and surrounding environmental changes, allowing for timely adjustment of tracking strategies. This significantly improves the real-time and accuracy of tracking, providing reliable data support for traffic safety management. For example, it can consistently output accurate vehicle trajectory information even in situations such as lane changes and weather emergencies.

[0018] In terms of data transmission and computing, the 5G-U roadside unit utilizes advanced technologies to increase channel capacity. Combined with the coordinated optimization of cloud and edge computing, it achieves efficient data transmission and processing, reduces system latency, and ensures real-time data delivery. Furthermore, the system's multi-dimensional security system effectively enhances the security of data transmission and storage, preventing data leakage and tampering.

[0019] The system also integrates functions such as spatiotemporal semantic enhancement, intelligent service generation, and digital twin visualization. Spatiotemporal semantic enhancement enables positioning to better meet actual scenario requirements; intelligent service generation provides personalized services based on vehicle trajectories and user needs; and digital twin visualization intuitively displays vehicle trajectories, facilitating monitoring and decision-making by managers, ultimately driving the development of intelligent and safe highways. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a schematic block diagram of the centimeter-level trajectory tracking system for highway vehicles based on 5G-U Beidou fusion proposed by the present invention; Figure 2 The following is a bar chart comparing the positioning accuracy of the traditional solution and this solution in different environments; Figure 3 The line graph of data transmission delay of the traditional solution and this solution changing with vehicle density; Figure 4 The radar charts of system resource utilization for the traditional solution and this solution are shown below. DETAILED DESCRIPTION

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 efforts are within the scope of protection of the present invention.

[0022] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise" and the like to indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as limiting the present invention.

[0023] In addition, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the said features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined. In addition, the terms "installed", "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The present invention will be further described in detail below with reference to the accompanying drawings.

[0024] Reference Figures 1 to 4 : A centimeter-level trajectory tracking system for highway vehicles based on 5G-U BeiDou fusion, including the following modules: Multi-source heterogeneous data acquisition module: Install a BeiDou-3 satellite receiver on the vehicle to obtain raw positioning data at a frequency of 10Hz ,in is the three-dimensional coordinate in the WGS-84 coordinate system, 5G-U roadside units (RSUs) are deployed every 500 meters along the highway, using spatial modulation multiple-input multiple-output (SM-MIMO) technology. Calculate the channel capacity, where B is the system bandwidth (Hz), SNR is the received signal-to-noise ratio (dB), and M is the number of activated antennas. is the complex gain of the i-th channel. RSU collects vehicle driving status data at a frequency of 20Hz , where v is the driving speed (m / s), a is the acceleration (m / s²), is the heading angle (°), is the pitch angle (°). 8-megapixel cameras are installed every 200 meters on both sides of the road to obtain vehicle visual feature data , and finally build a multi-source data set .

[0025] Multi-frequency and multi-constellation fusion positioning module: for Beidou B1C (1575.42MHz), B2a (1176.45MHz) and GPS L1 (1575.42MHz), L5 (1176.45MHz) frequency bands, using the improved ionospheric-free combined observation equation Calculate the observed value where , is the dual-frequency carrier phase observation value, , is the corresponding frequency, is the ionospheric residual correction coefficient (the value for this section is 0.8), and I1 is the ionospheric delay in the L1 band.

[0026] Dynamic environment perception and error correction module: building dynamic electronic fences on highways ,in , is the space-time coordinate of the fence boundary point. Determine whether the vehicle has crossed the boundary, where P(t) is the vehicle position, is the velocity component, is the time interval (take 1 second).

[0027] Centimeter-level trajectory generation and visualization module: based on improved B-spline interpolation algorithm Generate a smooth trajectory, where d i are the coordinates of the control points, is the p-order canonical B-spline basis function (p is 3), and the constrained optimization algorithm is used. Determine the optimal control point, where P i are discrete observation points, is the smoothing factor (taken as 0.1). While ensuring the trajectory accuracy, the trajectory curvature is continuously changed, and the resulting trajectory smoothness is improved compared to the traditional method.

[0028] Cloud collaboration and edge computing module: Build a distributed computing system consisting of 1 cloud center and 10 edge computing nodes. Achieve collaborative model optimization, where is the weight coefficient (take 0.6), is the cross entropy loss, is the KL divergence, p,q are the predicted distributions, T is the temperature parameter (take 4), the algorithm takes the teacher model The knowledge is distilled to the student model through the temperature parameter T , while protecting privacy, the accuracy of the edge node model is improved.

[0029] The present invention also includes a multi-frequency multi-constellation fusion positioning module: using an adaptive robust Kalman filter algorithm ,in is the estimated value of the state, is the observed value, is the measurement matrix, robustness gain , is the weight matrix, , is the residual, To achieve the mean error, the window size is dynamically adjusted through the window sliding mechanism. , is the basic window size (take 10), is the adjustment coefficient (taken as 0.5).

[0030] The present invention also includes a dynamic environment perception and error correction module: using an improved multipath error correction model , where A, B, C are traditional parameters (taken as 0.5, 0.2, 0.3 respectively), D, E are new direction-sensitive parameters (taken as 0.2, 0.3 respectively), is the reference direction angle (take 90°), is the reference height (take 5 meters). At the same time, according to the weather station data, use the formula Corrected tropospheric delay, where are constants (0.002277, 1255, and 0.00227, respectively), is the air pressure (hPa), is the absolute temperature (K), is the average temperature (K), is the water vapor pressure (hPa), and h is the receiver altitude (m).

[0031] In the present invention, the 5G-U roadside unit (RSU) in the multi-source heterogeneous data acquisition module adopts spatial modulation multiple input multiple output (SM-MIMO) technology, and is calculated by formula Calculate channel capacity. This technology increases channel capacity by 30%-40% under the same spectrum resources by activating different antennas to transmit additional information. It is deployed at intervals along highways. Satisfy the formula ,in As a safety factor, this formula adds the speed standard deviation to the traditional calculation based on vehicle speed and update cycle. , making the RSU deployment spacing more reasonable and improving network coverage.

[0032] In the present invention, the centimeter-level trajectory generation and visualization module is based on the space-time coordinate conversion formula Convert WGS-84 coordinates to the local space-time coordinate system, where T st is a 4×4 spatiotemporal transformation matrix containing rotation, translation, and time scaling parameters , is the rotation matrix, is the translation vector, s is the time scaling factor (take 1). Multi-resolution rendering algorithm is used ,in Rendering results for different resolutions, is the distance weight function , d is the observation distance, d i is the resolution switching threshold (100m, 500m, 1000m when i=1, 2, 3 respectively), is the attenuation coefficient (taken as 0.01).

[0033] In the present invention, the cloud collaboration and edge computing module design a dynamic resource allocation algorithm , the constraints are , where w i is the task weight, r ij Allocate benefits to resources, To assign decision variables, For resource consumption, Optimize resource capacity through reinforcement learning algorithm ,in is the discount factor (take 0.9), is the reward at time step t. The algorithm models the resource allocation problem as a Markov decision process by maximizing the long-term cumulative reward Realize dynamic resource allocation to improve the overall system throughput.

[0034] The present invention also includes a spatiotemporal semantic enhancement module: constructing a spatiotemporal semantic model of a highway ,in is a semantic object, For object relations, For time reasons, Design semantic enhancement positioning algorithm for spatial relationship , where the semantic correction is a semantic feature, such as lane lines, guardrails, etc., wi is the weight, f i The system uses semantic information to assist in correcting GNSS positioning errors, and improves positioning accuracy by 40%-50% in environments such as urban canyons.

[0035] The present invention also includes a multi-dimensional security assurance module: building a multi-dimensional security assessment model ,in Security indicators for various dimensions, including positioning accuracy, communication quality, system stability, etc. weight; design a security risk prediction algorithm ,in is the safety index change rate, For historical safety records, the long short-term memory network (LSTM) is used to predict future safety risks, which can provide early warning of potential safety hazards 30 seconds in advance with a false alarm rate of less than 5%.

[0036] The present invention also includes an intelligent service generation module: building a service demand understanding model ,in For user queries, To understand user intent using the pre-trained language model BERT, the accuracy rate is over 92%. We also designed trajectory prediction and service matching algorithms. ,in The prediction trajectory based on LSTM has a prediction accuracy of over 85%. Serving candidates, Based on user preferences, the most suitable service is matched through cosine similarity, with a matching accuracy of over 88%.

[0037] The present invention also includes a digital twin visualization module: using a multi-resolution rendering algorithm , where R i Rendering results for different resolutions, is the distance weight function , d is the observation distance, d i The algorithm dynamically adjusts the rendering resolution according to the observation distance, which ensures the visual effect while increasing the rendering frame rate by 40%-50%, thus achieving real-time and smooth rendering of large-scale highway scenes.

[0038] The above are only preferred specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solutions and inventive concepts of the present invention, should be covered by the scope of protection of the present invention.

Claims

1. A centimeter-level trajectory tracking system for highway vehicles based on 5G-U BeiDou fusion, characterized in that: Includes the following modules: Multi-source heterogeneous data acquisition module: obtains vehicle original positioning data through BeiDou-3 satellite receiver ,in, are three-dimensional coordinates, Timestamp; deployed on 5G-U roadside units to collect vehicle driving status data , where v is velocity, a is acceleration, is the heading angle, is the pitch angle; the vehicle visual feature data is obtained using the roadside camera , build a multi-source data set ; Multi-frequency and multi-constellation fusion positioning module: Using the improved ionospheric-free combined observation equation Calculate the ionospheric-free combination observation values ​​of BeiDou B1C, B2a and GPS L1, L5 bands. The equation adds an ionospheric residual correction term on the basis of the traditional ionospheric-free combination. , is the ionospheric residual correction coefficient; Dynamic environment perception and error correction module: building dynamic electronic fences on highways , through the space-time distance formula Determine whether the vehicle has crossed the boundary, where is the spatial position, is the velocity component, is the time interval; Centimeter-level trajectory generation and visualization module: based on improved B-spline interpolation algorithm Generate a smooth trajectory, where d i is the control point, is a p-order canonical B-spline basis function that satisfies the endpoint interpolation condition , through the constrained optimization algorithm Identify control points; Cloud collaboration and edge computing modules: Building a distributed computing system under the federated learning architecture , through the knowledge distillation algorithm To achieve model collaborative optimization, the algorithm transforms the teacher model The knowledge is distilled to the student model through the temperature parameter T , while protecting privacy, the accuracy of the edge node model is improved.

2. The centimeter-level trajectory tracking system for highway vehicles based on 5G-U BeiDou fusion according to claim 1 is characterized in that: It also includes a multi-frequency and multi-constellation fusion positioning module: using adaptive robust Kalman filtering algorithm , where the robustness gain , is the weight matrix, , is the residual, To achieve the mean error, the window size is dynamically adjusted through the window sliding mechanism. , the mechanism is based on the residual variance Automatically adjust the filter window.

3. The centimeter-level trajectory tracking system for highway vehicles based on 5G-U BeiDou fusion according to claim 1 is characterized in that: It also includes a dynamic environment perception and error correction module: using an improved multipath error correction model , this model adds a direction-sensitive term based on the traditional exponential decay model , where D and E are new parameters, , For reference value.

4. The centimeter-level trajectory tracking system for highway vehicles based on 5G-U BeiDou fusion according to claim 1 is characterized in that: The 5G-U roadside unit in the multi-source heterogeneous data acquisition module adopts spatial modulation multi-input multi-output technology, and the formula Calculate channel capacity by activating different antennas to transmit additional information, deployed at intervals along the highway Satisfy the formula ,in As a safety factor, this formula adds the speed standard deviation to the traditional calculation based on vehicle speed and update cycle. .

5. The centimeter-level trajectory tracking system for highway vehicles based on 5G-U BeiDou fusion according to claim 1 is characterized in that: The centimeter-level trajectory generation and visualization module uses the space-time coordinate conversion formula Convert WGS-84 coordinates to the local space-time coordinate system, where T st is a 4×4 spatiotemporal transformation matrix containing rotation, translation, and time scaling parameters ,The parameters are estimated by nonlinear least squares algorithm.,The space-time transformation model introduces the time dimension based on the traditional three-dimensional space,transformation.

6. The centimeter-level trajectory tracking system for highway vehicles based on 5G-U BeiDou fusion according to claim 1 is characterized in that: The cloud collaboration and edge computing module designs a dynamic resource allocation algorithm , the constraints are , where w i is the task weight, r ij Allocate benefits to resources and optimize through reinforcement learning algorithms , the algorithm models the resource allocation problem as a Markov decision process by maximizing the long-term cumulative reward Realize dynamic resource allocation.

7. The centimeter-level trajectory tracking system for highway vehicles based on 5G-U BeiDou fusion according to claim 1 is characterized in that: It also includes a spatiotemporal semantic enhancement module: building a spatiotemporal semantic model for highways ,in is a semantic object, For object relations, For time reasons, Design semantic enhancement positioning algorithm for spatial relationship , where the semantic correction is the semantic feature, w i is the weight, f i is the mapping function.

8. The centimeter-level trajectory tracking system for highway vehicles based on 5G-U BeiDou fusion according to claim 1 is characterized in that: It also includes a multi-dimensional security assurance module: building a multi-dimensional security assessment model ,in Security indicators for each dimension, weight; design a security risk prediction algorithm ,in is the safety index change rate, For historical security records, predict future security risks through long short-term memory networks.

9. The centimeter-level trajectory tracking system for highway vehicles based on 5G-U BeiDou fusion according to claim 1 is characterized in that: Also includes intelligent service generation module: building a service demand understanding model ,in For user queries, To understand user intent, we use the pre-trained language model BERT to obtain contextual information. Design trajectory prediction and service matching algorithms ,in is the prediction trajectory based on LSTM, Serving candidates, For user preferences, matching services is performed through cosine similarity.

10. The centimeter-level trajectory tracking system for highway vehicles based on 5G-U BeiDou fusion according to claim 1 is characterized in that: Also includes a digital twin visualization module: using multi-resolution rendering algorithm , where R i Rendering results for different resolutions, is the distance weight function , d is the observation distance, d i The threshold for resolution switching is used. The algorithm dynamically adjusts the rendering resolution according to the observation distance, thereby improving the rendering frame rate while ensuring the visual effect.

Images