Tunnel driving vehicle monitoring method and system, terminal and storage medium
By building simulation models in the tunnel driving vehicle monitoring system and using Kalman prediction models, combining image recognition technology to mark and display special vehicles, the problem that existing systems cannot prevent hazards to other vehicles is solved, and the safety of tunnel driving is improved and management efficiency.
Patent Information
- Application Number
- CN202411352709.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-26
- Publication Date
- 2025-01-14
AI Technical Summary
The existing tunnel vehicle monitoring system can only detect violations of the vehicle itself, which can easily lead to direct harm to other vehicles and fail to effectively prevent potential risks.
The simulation model is constructed based on tunnel geometric data, combined with Kalman prediction model and image recognition technology, the real-time position, speed and acceleration of the vehicle are obtained, the vehicle trajectory and behavior are predicted, special vehicles are marked, and displayed in the simulation model.
It has achieved accurate prediction of vehicle driving trends and timely discovery of potential risks, improved the safety of tunnel driving and traffic management efficiency, and enhanced the emergency response capabilities for special circumstances.
Smart Images
Figure CN119314326A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of tunnel driving, and in particular to a method, system, terminal and storage medium for monitoring vehicles driving in a tunnel. Background Art
[0002] Tunnels are underground passages with various types and characteristics. Many tunnels use traditional sensors such as radars and cameras to monitor vehicles and obtain the real-time status of vehicles.
[0003] Radar sensors can be used to detect vehicle distance, speed and other information. They transmit electromagnetic waves and receive reflected waves, and obtain relevant dynamic data of the vehicle based on time difference and Doppler effect. Cameras are mainly used for visual monitoring, which can intuitively capture the appearance, license plate and other information of the vehicle. With the help of image recognition technology, they can also analyze the behavior of the vehicle to a certain extent, such as whether it has violated the lane change regulations.
[0004] The current vehicle inspection can only detect the violations that the vehicle itself has already committed. When a vehicle commits a violation, it is easy to cause direct harm to other vehicles. Summary of the invention
[0005] In order to solve the problem that the current vehicle detection can only detect the illegal behaviors that have already occurred in the vehicle itself, and when the vehicle commits illegal behaviors, it is easy to cause direct harm to other vehicles, the present invention provides a tunnel vehicle monitoring method, system, terminal and storage medium.
[0006] In a first aspect, the present invention provides a method for monitoring vehicles traveling in a tunnel, which adopts the following technical solution: A method for monitoring vehicles traveling in a tunnel, comprising the following steps: S1, constructing a tunnel simulation model based on the tunnel’s geometric data; S2, obtaining the vehicle's real-time position, real-time speed, real-time acceleration and special identification; S3, establishing a Kalman prediction model to predict the vehicle position, speed and acceleration; S4, combining the predicted position, speed, acceleration with the real-time position, real-time speed, and real-time acceleration to analyze and mark special vehicles, wherein the special vehicles include: special track vehicles and special speed vehicles; S5, displaying the marked special vehicle and / or the vehicle with the special identification at a corresponding position of the simulation model.
[0007] Optionally, in step S3, the construction of the Kalman prediction model includes: Constructing a state vector based on the position coordinates of the vehicle in the plane, the vehicle speed, and the vehicle acceleration; According to the kinematic principle of the vehicle, the state transfer matrix A is determined, and the state vector prediction at the next moment is:
[0008]
[0009] in, is the predicted state vector at time t+1, is the state vector at time t, B is the control input matrix, is the external control input vector at time t, is the random disturbance at time t; According to the state transfer matrix A and the random interference at time t Calculate the covariance matrix of the predicted state vector at time t+1 ; The Kalman prediction model is modified according to the actual state vector newly acquired at time t+1.
[0010] Optionally, the step of correcting the Kalman prediction model according to the newly acquired actual state vector at time t+1 includes: Construct an actual model between the actual state vector at the next moment and the predicted state vector at the next moment:
[0011] in, is the actual state vector at time t+1, H is the actual matrix, is the actual noise at time t+1, is a Gaussian noise with a mean of 0 and an actual noise covariance matrix of R; Calculate the Kalman gain based on the actual matrix and the actual noise covariance matrix:
[0012] in, is the Kalman gain at time t+1, is the covariance matrix of the predicted state vector at time t+1; Combine the Kalman gain, the actual state vector and the actual matrix to correct the predicted state vector and the covariance matrix of the predicted state vector:
[0013]
[0014] in, is the corrected predicted state vector at time t+1, is the covariance matrix of the corrected predicted state vector at time t+1.
[0015] Optionally, marking a special vehicle includes marking a special vehicle on a track, specifically including: Calculate the actual direction vector of the current moment actual position relative to the position at the previous moment and the predicted direction vector of the current moment predicted position relative to the position at the previous moment; The deviation angle between the actual direction vector and the predicted direction vector is calculated by the vector dot product formula; Determine whether the deviation angle exceeds the angle threshold. If it exceeds the angle threshold, it is determined to be a special track vehicle and the vehicle is marked.
[0016] Optionally, in step S4, marking a special vehicle includes marking a vehicle with a special speed, specifically including: Determine whether the predicted speed exceeds the current speed and / or the calibrated speed. If it is determined to be speeding, the vehicle is marked as a special speed vehicle; It is determined whether the predicted acceleration is greater than a threshold. If it is determined to be greater than the threshold, the vehicle is a speed-special vehicle and the vehicle is marked.
[0017] Optionally, obtaining the special identifier in step S2 includes: Pre-establish a keyword library related to special logos; Acquire appearance image information of the vehicle; Keywords are extracted based on image recognition algorithms and compared with the keyword library to obtain special identification.
[0018] Optionally, the special trajectory vehicles, special speed vehicles and vehicles with special identification are displayed in the simulation model using different colors respectively.
[0019] In a second aspect, a tunnel vehicle monitoring system is provided, wherein when the system is executed, the tunnel vehicle monitoring method described above is applied, including: Information acquisition module, which obtains the vehicle's real-time position, real-time speed, real-time acceleration and special identification; Model building module, builds tunnel simulation model based on tunnel geometry data; establishes Kalman prediction model; The vehicle marking module predicts the vehicle's driving trajectory, speed and acceleration based on the Kalman prediction model; combines the predicted position, speed, acceleration with the real-time position, real-time speed and real-time acceleration to analyze and mark special vehicles, including: special trajectory vehicles and special speed vehicles.
[0020] In a third aspect, a tunnel vehicle monitoring terminal is provided, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the memory is coupled to the processor, and the processor implements the tunnel vehicle monitoring method as described above when executing the computer program.
[0021] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned method for monitoring vehicles traveling in a tunnel.
[0022] In summary, the present invention has the following beneficial technical effects: The simulation model built based on tunnel geometry data provides an intuitive and accurate basic framework for the entire monitoring, which helps to fully understand the impact of the tunnel environment on vehicle driving. Secondly, obtaining the real-time position, speed and special identification of the vehicle can achieve accurate grasp of the vehicle status in the tunnel, providing detailed data support for subsequent analysis. Furthermore, the Kalman prediction model's prediction of vehicle driving trajectory, speed and acceleration can provide early insight into vehicle driving trends and help to discover potential risks in a timely manner. Then, by combining predictions with actual data to mark special vehicles, it is possible to accurately screen out vehicles with abnormal driving, abnormal speed and special identification, which is convenient for targeted management and response. Finally, the marked special vehicles are displayed at the corresponding position of the simulation model, so that the monitor can quickly locate and handle special vehicle-related situations, improving the safety of tunnel driving, the efficiency of traffic management and the emergency response capabilities for special situations. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] Figure 1 The present invention is a flow chart of a method for monitoring vehicles traveling in a tunnel according to an embodiment of the present invention.
[0024] Figure 2 The present invention is a schematic structural diagram of a tunnel vehicle monitoring system according to an embodiment of the present invention. DETAILED DESCRIPTION
[0025] The following is combined with Figure 1 -Attached Figure 2 The present invention is described in further detail.
[0026] Example 1 Reference Figure 1 A method for monitoring vehicles traveling in a tunnel according to this embodiment includes the following steps: S1, constructing a tunnel simulation model based on the tunnel’s geometric data; Determine the cross-sectional shape and size of the tunnel, such as the diameter of a circular tunnel, the width and height of a rectangular tunnel, etc. For irregularly shaped tunnels, multiple cross-sectional data points are required to describe their shape. Record the curve information of the tunnel, including the radius of the curve, the starting and ending angles of the curve, and the position of the curve in the tunnel (such as the distance from the starting point of the tunnel). If the tunnel has a slope, measure the size of the slope (usually expressed as a percentage) and the range of slope changes.
[0027] Use SUMO (Simulation of Urban MObility), which is an open source traffic simulation software suitable for simulating large-scale traffic networks. For tunnel models, the geometry of the tunnel can be accurately constructed, and the input and output of traffic flow can be flexibly defined. The interface with external programs is also relatively rich, which facilitates secondary development to meet specific simulation needs, such as simulating the driving of special vehicles (such as dangerous goods transport vehicles) in tunnels. General 3D modeling software combined with programming.
[0028] S2, obtaining the vehicle's real-time position, real-time speed, real-time acceleration and special identification; Acquire the real-time position and speed of the vehicle based on radar sensors and cameras; Radar detects targets by emitting high-frequency electromagnetic waves and receiving reflected waves. In a tunnel, radar sensors can be installed on the tunnel wall or at specific locations. The radar emits electromagnetic waves to the vehicle, and based on the time delay of the reflected waves and the propagation speed of the electromagnetic waves, the distance between the vehicle and the radar can be calculated (divided by 2 because of the round-trip distance of the electromagnetic waves). If the coordinates of the installation location of the radar are known, and assuming that the vehicle is located in a straight line within the detection range of the radar, the position coordinates of the vehicle relative to the radar can be obtained through simple geometric relationships.
[0029] Radar uses the Doppler effect to measure the speed of a vehicle. When a vehicle moves toward the radar, the frequency of the reflected wave increases; when the vehicle moves away from the radar, the frequency of the reflected wave decreases. Based on the difference between the received reflected wave frequency and the transmitted wave frequency, the vehicle's speed can be calculated using a formula.
[0030] A camera is installed in the tunnel, and the camera captures an image of the vehicle. By analyzing the image, using the known camera installation position, viewing angle, reference objects in the image (such as markings on the tunnel wall, lane lines, etc.), and the pixel position of the vehicle in the image, computer vision technology can be used to calculate the approximate position of the vehicle. For example, if the height of the camera and the horizontal distance from the tunnel wall are known, and the pixel distance between the vehicle and the tunnel wall in the image is, the horizontal distance between the vehicle and the camera can be estimated based on the principle of similar triangles, and then the position of the vehicle in the tunnel can be determined.
[0031] Speed acquisition: The camera can calculate the speed by analyzing the position change of the vehicle in consecutive image frames. Assuming the frame rate of the camera is (the number of frames shot per second), the displacement of the vehicle in two adjacent frames is (the pixel displacement of the vehicle in the image is obtained by image analysis, and then converted to actual displacement according to the method mentioned above), then the speed of the vehicle. However, the accuracy of this method may be affected by factors such as image resolution, frame rate, and the relative motion direction between the vehicle and the camera.
[0032] Real-time and comprehensive monitoring of tunnel traffic helps with traffic flow management, accident prevention, and can provide personalized services to drivers, thereby increasing tunnel capacity, preventing dangerous situations and improving the driving experience.
[0033] Obtaining special identification includes: Pre-establish a keyword library related to special logos; Acquire appearance image information of the vehicle; Keywords are extracted based on image recognition algorithms and compared with the keyword library to obtain special identification.
[0034] S3, establishing a Kalman prediction model to predict the vehicle position, speed and acceleration; The Kalman prediction model explicitly considers the process noise and actual noise in the system. Process noise is used to describe the uncertainty in the dynamic changes of the system itself, such as the random interference factors such as uneven road surface and wind when the vehicle is driving; actual noise reflects the error in sensor measurement, such as the error in radar measuring vehicle distance and speed, and the error in camera identifying vehicle position.
[0035] By setting the process noise covariance matrix and the actual noise covariance matrix in the model, the Kalman prediction model can incorporate this noise information into the prediction and update process, thereby more realistically reflecting the actual system situation.
[0036] In the presence of noise, the Kalman prediction model can provide the best estimate of the system state (in the sense of minimum mean square error) based on the dynamic characteristics of the system and the actual data. This means that it can infer the true situation of the system (such as the driving state of the vehicle) as accurately as possible under noise interference. For example, even if there is a certain error fluctuation in the measurement data of the sensor, the Kalman filter can continuously update the estimate so that the final result is close to the true value.
[0037] The Kalman prediction model uses a recursive algorithm for calculation. At each time step, the current state estimate can be calculated by using only the state estimate value, covariance estimate value and the actual value at the previous moment. This recursive feature makes it very suitable for real-time application scenarios, such as real-time monitoring of vehicle driving status in tunnels.
[0038] The construction of the Kalman prediction model includes: S3-1, based on the vehicle's position coordinates in the plane , vehicle speed and vehicle acceleration Constructing the state vector ; S3-2, determine the state transfer matrix A according to the kinematic principle of the vehicle, and the state vector prediction at the next moment is:
[0039]
[0040] in, is the predicted state vector at time t+1, is the state vector at time t, B is the control input matrix, is the external control input vector at time t, is the random disturbance at time t, is a Gaussian noise with zero mean and random interference covariance matrix Q; In this embodiment, external control includes braking, acceleration and steering. When the vehicle only generates acceleration or braking behavior, the control input matrix B= , at this time, the state vector prediction at the next moment is specifically expanded as: ; S3-3, according to the state transfer matrix A and the random interference at time t Calculate the covariance matrix of the predicted state vector at time t+1 ; The specific prediction is:
[0041] in, is the covariance matrix of the state vector at time t, is the covariance matrix of the predicted state vector at time t+1; the covariance matrix determines the confidence and uncertainty of the system's state estimation. The initial covariance matrix is determined based on the error range of the position, velocity, and acceleration detection equipment, for example: The uncertainty of the initial position is 10 meters, and the uncertainty of the initial velocity is 2 meters / second, so the initial covariance matrix is Can be set to:
[0042] It means that the confidence in the position is low (large error) at the beginning, while the confidence in the velocity is high (small error); S3-4, correcting the Kalman prediction model according to the actual state vector newly acquired at time t+1, the steps include: Construct an actual model between the actual state vector at the next moment and the predicted state vector at the next moment:
[0043] in, is the actual state vector at time t+1, H is the actual matrix, is the actual noise at time t+1, is a Gaussian noise with a mean of 0 and an actual noise covariance matrix of R; Calculate the Kalman gain based on the actual matrix and the actual noise covariance matrix:
[0044] in, is the Kalman gain at time t+1, is the covariance matrix of the predicted state vector at time t+1; Combine the Kalman gain, the actual state vector and the actual matrix to correct the predicted state vector and the covariance matrix of the predicted state vector:
[0045]
[0046] in, is the corrected predicted state vector at time t+1, is the covariance matrix of the corrected predicted state vector at time t+1.
[0047] S4, combining the predicted position, speed, acceleration with the real-time position, real-time speed, and real-time acceleration to analyze and mark special vehicles, wherein the special vehicles include: special track vehicles and special speed vehicles; Marking special vehicles includes marking special vehicles on the track, including: Calculate the actual direction vector of the current moment actual position relative to the position at the previous moment and the predicted direction vector of the current moment predicted position relative to the position at the previous moment; The deviation angle between the actual direction vector and the predicted direction vector is calculated by the vector dot product formula; Determine whether the deviation angle exceeds the angle threshold. If it exceeds the angle threshold, it is determined to be a special track vehicle and the vehicle is marked; Marking special vehicles includes marking special vehicles on the track, and also includes: The actual direction vector and the predicted direction vector are used to determine whether the vehicle is going in the wrong direction. If the vehicle is going in the wrong direction, it is determined to be a special trajectory vehicle and the vehicle is marked.
[0048] Marking of special vehicles includes marking of vehicles with special speeds, including: Determine whether the predicted speed exceeds the current speed and / or the calibrated speed. If it is determined to be speeding, the vehicle is marked as a special speed vehicle; It is determined whether the predicted acceleration is greater than a threshold. If it is determined to be greater than the threshold, the vehicle is a speed-special vehicle and the vehicle is marked.
[0049] S5, displaying the marked special vehicle and / or the vehicle with the special identification at a corresponding position of the simulation model.
[0050] The special track vehicles, special speed vehicles and vehicles with special markings are displayed in the simulation model using different colors. In this embodiment, red, yellow and blue are used to represent the colors of the three special vehicles for marking.
[0051] Example 2 Reference Figure 2 The difference between this embodiment and embodiment 1 is that this embodiment provides a tunnel vehicle monitoring system, and when the system is executed, the tunnel vehicle monitoring method described above is applied, including: Information acquisition module, which obtains the vehicle's real-time position, real-time speed, real-time acceleration and special identification; Model building module, builds tunnel simulation model based on tunnel geometry data; establishes Kalman prediction model; The vehicle marking module predicts the vehicle's driving trajectory, speed and acceleration based on the Kalman prediction model; combines the predicted position, speed, acceleration with the real-time position, real-time speed and real-time acceleration to analyze and mark special vehicles, including special trajectory vehicles and special speed vehicles; The display module displays the marked special vehicle and / or the vehicle with the special identification at the corresponding position of the simulation model.
[0052] Example 3 The difference between this embodiment and Embodiment 1 is that this embodiment provides a tunnel vehicle monitoring terminal, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, the memory is coupled to the processor, and the processor implements the tunnel vehicle monitoring method as described above when executing the computer program.
[0053] Example 4 The difference between this embodiment and embodiment 1 is that this embodiment provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the tunnel vehicle monitoring method as described above.
[0054] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. As for the method disclosed in the embodiment, since it corresponds to the system disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.
[0055] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.
[0056] The above disclosure is only a preferred embodiment of the present invention, but the present invention is not limited thereto. Any non-creative changes that can be thought of by a person skilled in the art, as well as several improvements and modifications made without departing from the principle of the present invention, should fall within the protection scope of the present invention.
Claims
1. A method for monitoring vehicles traveling in a tunnel, characterized in that: The following steps are involved: S1, constructing a tunnel simulation model based on the tunnel’s geometric data; S2, obtaining the vehicle's real-time position, real-time speed, real-time acceleration and special identification; S3, establishing a Kalman prediction model to predict the vehicle position, speed and acceleration; S4, combining the predicted position, speed, acceleration with the real-time position, real-time speed, and real-time acceleration to analyze and mark special vehicles, wherein the special vehicles include: special track vehicles and special speed vehicles; S5, displaying the marked special vehicle and / or the vehicle with the special identification at a corresponding position of the simulation model.
2. The method for monitoring vehicles traveling in a tunnel according to claim 1, characterized in that: In step S3, the construction of the Kalman prediction model includes: Constructing a state vector based on the position coordinates of the vehicle in the plane, the vehicle speed, and the vehicle acceleration; According to the kinematic principle of the vehicle, the state transfer matrix A is determined, and the state vector prediction at the next moment is: in, is the predicted state vector at time t+1, is the state vector at time t, B is the control input matrix, is the external control input vector at time t, is the random disturbance at time t; According to the state transfer matrix A and the random interference at time t Calculate the covariance matrix of the predicted state vector at time t+1 ; The Kalman prediction model is modified according to the actual state vector newly acquired at time t+1.
3. The method for monitoring vehicles traveling in a tunnel according to claim 2, characterized in that: The steps of correcting the Kalman prediction model according to the actual state vector newly acquired at time t+1 include: Construct an actual model between the actual state vector at the next moment and the predicted state vector at the next moment: in, is the actual state vector at time t+1, H is the actual matrix, is the actual noise at time t+1, is a Gaussian noise with a mean of 0 and an actual noise covariance matrix of R; Calculate the Kalman gain based on the actual matrix and the actual noise covariance matrix: in, is the Kalman gain at time t+1, is the covariance matrix of the predicted state vector at time t+1; Combine the Kalman gain, the actual state vector and the actual matrix to correct the predicted state vector and the covariance matrix of the predicted state vector: in, is the corrected predicted state vector at time t+1, is the covariance matrix of the corrected predicted state vector at time t+1.
4. The method for monitoring vehicles traveling in a tunnel according to claim 1, characterized in that: Marking special vehicles includes marking special vehicles on the track, including: Calculate the actual direction vector of the current moment actual position relative to the position at the previous moment and the predicted direction vector of the current moment predicted position relative to the position at the previous moment; The deviation angle between the actual direction vector and the predicted direction vector is calculated by the vector dot product formula; Determine whether the deviation angle exceeds the angle threshold. If it exceeds the angle threshold, it is determined to be a special track vehicle and the vehicle is marked.
5. The method for monitoring vehicles traveling in a tunnel according to claim 1, characterized in that: In step S4, marking a special vehicle includes marking a vehicle with a special speed, specifically including: Determine whether the predicted speed exceeds the current speed and / or the calibrated speed. If it is determined to be speeding, the vehicle is marked as a special speed vehicle; It is determined whether the predicted acceleration is greater than a threshold. If it is determined to be greater than the threshold, the vehicle is a speed-special vehicle and the vehicle is marked.
6. The method for monitoring vehicles traveling in a tunnel according to claim 1, characterized in that: Acquiring the special identification in step S2 includes: Pre-establish a keyword library related to special logos; Acquire appearance image information of the vehicle; Keywords are extracted based on image recognition algorithms and compared with the keyword library to obtain special identification.
7. The method for monitoring vehicles traveling in a tunnel according to claim 1, characterized in that: The special track vehicles, special speed vehicles and vehicles with special identification are displayed in the simulation model using different colors respectively.
8. A tunnel vehicle monitoring system, characterized in that: When the system is executed, the method for monitoring vehicles traveling in a tunnel as claimed in claims 1 to 7 is applied, including: Information acquisition module, which obtains the vehicle's real-time position, real-time speed, real-time acceleration and special identification; Model building module, builds tunnel simulation model based on tunnel geometry data; establishes Kalman prediction model; The vehicle marking module predicts the vehicle's driving trajectory, speed and acceleration based on the Kalman prediction model; combines the predicted position, speed, acceleration with the real-time position, real-time speed and real-time acceleration to analyze and mark special vehicles, including special trajectory vehicles and special speed vehicles; The display module displays the marked special vehicle and / or the vehicle with the special identification at the corresponding position of the simulation model.
9. A tunnel vehicle monitoring terminal, characterized in that: The invention comprises a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, wherein the memory is coupled to the processor, and when the processor executes the computer program, the method for monitoring vehicles traveling in a tunnel as claimed in any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the method for monitoring vehicles traveling in a tunnel according to any one of claims 1 to 7.