Vehicle management control system in vehicle test area

Abstract

The invention discloses a vehicle management control system in a vehicle test area. The system comprises detection equipment, a data acquisition module, a data processing module, a display module and a control module, the data acquisition module is used for receiving vehicle data information acquired by the plurality of detection devices, registering identities of drivers in a waiting area and vehicle identification codes, and distributing queuing numbers for vehicles; the data processing module is used for receiving the vehicle data information uploaded by the data acquisition module and identifying the current state of each vehicle in the vehicle test area; the display module is used for displaying the vehicle queuing condition, indicating the vehicles to enter and exit and broadcasting; and the control module is used for acquiring the current state of each vehicle identified by the data processing module, controlling the display module to display, and releasing or forbidding the vehicle. According to the invention, automatic management control is carried out on the vehicles in the key vehicle test area, so that vehicle number management and control, safety track monitoring and related auxiliary operation service providing are realized.

Description

Technical Field

[0001] The invention relates to the field of vehicle management, and in particular to a vehicle management control system in an automobile testing area. Background Art

[0002] In the development of the automotive industry, modern testing technology plays an important role. The rapid development of the automotive industry has put forward higher requirements for the entire automotive industry.

[0003] The automobile test area refers to the area where ABS braking performance tests, moose tests and limit tests are conducted in closed test fields. In order to avoid too many vehicles entering the test area at the same time, it is necessary to strictly control the entry vehicle scheduling area, the departure area, and the return area. At the same time, any violations of the vehicles in the automobile test area may affect the safety of other vehicles.

[0004] To this end, an effective management and control system needs to be provided for the automobile testing area to achieve unmanned management and safety control of vehicles entering and leaving the area. Summary of the invention

[0005] In order to solve the above technical problems, the present invention provides a vehicle management and control system for an automobile testing area to achieve unmanned management and safety control of vehicles entering and leaving the area.

[0006] The present invention provides a vehicle management and control system for an automobile test area, the system comprising: a detection device, a data acquisition module, a data processing module, a display module and a control module;

[0007] The data acquisition module is used to receive vehicle data information collected by multiple detection devices, register the identity of the driver in the waiting area and the vehicle identification code, and assign a queue number to the vehicle; the detection equipment includes but is not limited to a road test unit, a laser radar, a millimeter wave radar and a high-definition camera;

[0008] The data processing module is used to receive the vehicle data information uploaded by the data acquisition module and identify the current status of each vehicle in the automobile test area;

[0009] The display module is used to display the vehicle queue situation, indicate the vehicle entry and exit and broadcast;

[0010] The control module is used to obtain the current status of each vehicle identified by the data processing module, control the display module to display, and release or prohibit the vehicle from passing.

[0011] Optionally, the detection equipment is distributed on both sides of the road in the automobile test area to collect vehicle data information, and uses ultra-wideband technology to send the collected vehicle data information to the data acquisition module;

[0012] The collection of vehicle data information includes: identifying the vehicle's on-board unit through the road test unit, capturing pictures and recording videos of the scene through high-definition cameras, and establishing a three-dimensional point cloud map of the test area through laser radar and millimeter-wave radar;

[0013] The data acquisition module includes an identification unit and a positioning unit. The identification unit confirms the queue number of the vehicle according to the vehicle-mounted unit. The positioning unit receives the position information of the detection equipment through ultra-wideband technology and obtains the positioning information of each vehicle according to the relative position between the detection equipment and each vehicle.

[0014] Optionally, the data processing module uses edge computing technology to count the entry and exit of vehicles in each area, and the steps are as follows:

[0015] Data classification: After obtaining the vehicle data information collected by the detection equipment, the vehicle data information of different vehicles is distinguished through different on-board units;

[0016] Model training: Pre-select vehicle data information in different states and perform feature extraction. The vehicle data information first enters the convolution kernel for convolution calculation, and then enters the global pooling for global pooling processing to obtain the features of vehicles in different states and establish a mapping relationship between vehicle states and vehicle features.

[0017] Feature recognition: Input the vehicle data collected by the detection equipment into the model to obtain the vehicle's features, and then use the mapping relationship to determine the vehicle's current status;

[0018] Data statistics: Count the number of vehicles and arrange for vehicles to enter or leave the corresponding area based on the current status of each vehicle and the capacity of the corresponding area;

[0019] The characteristics of vehicles in different states include, but are not limited to: entering the test area, undergoing testing, waiting in line, and leaving the test area.

[0020] Optionally, the data processing module is also used to establish lane line detection and vehicle detection, and collect the driving habits of drivers through big data, combine lane line detection and vehicle detection, predict the running trajectory of each vehicle according to the vehicle data information of each vehicle in the video, mark each vehicle before it violates the rules, and generate warning information for broadcast; the violations include but are not limited to arbitrary U-turns, parking on the road, cutting in, and driving in the wrong direction;

[0021] The lane line detection includes: capturing a picture of the lane line in the car test area by a high-definition camera, performing grayscale conversion on the picture, selecting a number of highlighted pixels in the grayscale picture, dividing a center line in the original lane line picture, the road line on the left side of the center line is the inner line, and the road line on the right side of the center line is the outer line, the corresponding pixel points on the left side of the center line in the grayscale picture belong to the inner line, and the pixel points on the right side of the center line belong to the outer line; connecting the adjacent pixel points in the inner line, connecting the adjacent pixel points in the outer line, removing abnormal points in the pixel points, so that the inner line and the outer line tend to be straight lines, and displaying them in the form of a top view and a front view;

[0022] The vehicle detection includes: capturing images of vehicles in the vehicle test area through a high-definition camera, extracting features of the vehicles, tracking the vehicles according to the features of the vehicles, and restoring the lane lines in the top view and the front view in the same proportion according to the features of the vehicles combined with the three-dimensional point cloud of the vehicles;

[0023] Based on the position of each vehicle in the overhead view and front view of the lane line, it is determined whether the vehicle has violated the regulations.

[0024] Furthermore, the data processing module is also used to monitor the illegal behaviors of each vehicle through AI recognition technology, and determine in real time whether the vehicle has any illegal behaviors, including:

[0025] Using a high-definition camera to obtain images of vehicle violations, a training set and a test set are established. Both the training set and the test set are derived from the images of vehicle violations and are randomly assigned. The training set is input to train the deep learning model. After the training is completed, the test set is input to the deep learning model to calculate the accuracy of the deep learning model on the test set.

[0026] The deep learning model is composed of a back-propagation neural network, which includes an input layer, multiple hidden layers and an output layer.

[0027] Optionally, the display module establishes communication with the data processing module, and the display module includes multiple display devices, including but not limited to display screens, signal lights, and intelligent voice devices;

[0028] The display screen displays the capacity of the vehicle testing area and queue information, including but not limited to driver information, queue number, vehicle identification number, and queue status;

[0029] The signal lights are used to indicate the entry and exit of vehicles in each area;

[0030] After receiving the processing result generated by the data processing module, the intelligent voice device notifies the vehicle to enter or exit, and after receiving the judgment from the data processing module that the vehicle has violated the regulations, the intelligent voice device announces the violation of the regulations.

[0031] Optionally, after receiving the processing result generated by the data processing module, the control module controls the gate to release or prohibit passage, and controls the display screen and signal lights to display.

[0032] Furthermore, the system also includes a database for storing driver information, vehicle information and vehicle violation information.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] The embodiment of the present invention receives vehicle data information collected by multiple detection devices through a data acquisition module, and the data processing module analyzes the vehicles in the automobile test area, and then arranges the passage of each vehicle, and determines whether the vehicle has any violations during the driving process of the vehicle, and then issues an early warning to each vehicle. The control module controls the display module to display or broadcast the vehicle's passage arrangements and violations, thereby realizing unmanned management and safety control of vehicles entering and leaving. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only preferred embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 It is a structural schematic diagram of a vehicle management and control system for an automobile test area provided in Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0037] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0038] Embodiment 1

[0039] Figure 1 1 is a schematic diagram of a vehicle management control system for a vehicle test area provided by the first embodiment of the present invention. This embodiment is applicable to the situation of unmanned vehicle entry and exit management in a vehicle test area. Figure 1 As shown, the system includes: a plurality of monitoring devices, a data acquisition module 10 , a data processing module 20 , a display module 30 and a control module 40 .

[0040] It should be noted that the automobile test area is divided into multiple test areas for different tests, and multiple detection devices or display devices are placed at the entrances and exits of each area or on both sides of the road line. The data acquisition module 10 will summarize the data collected by these detection devices, and the data processing module 20 will analyze these data and plan the vehicle traffic in each area. The display module 30 includes a variety of display devices to display the traffic conditions of each area, and the control module 40 controls the display module 30 and the gates at the entrances and exits of each area.

[0041] The data collection module is used to receive vehicle data information collected by multiple detection devices, register the identity of the driver in the waiting area and the vehicle identification code, and assign a queue number to the vehicle.

[0042] The vehicle identification number of a vehicle refers to the VIN code, which consists of 17 characters. The vehicle identification code is the ID number of the car, which includes the vehicle manufacturer, year, model, body type and code, engine code and assembly location. The VIN code is used to distinguish different vehicles and is displayed on the subsequent display screen.

[0043] Before entering the car testing area, the vehicle needs to register to confirm the identity of the driver and the vehicle, and assign a queue number to the vehicle. The display module will also display the queue number to inform the driver how many vehicles are waiting in line ahead of him.

[0044] The vehicle identification code is unique, but the queue number will change as the vehicle enters and exits. After the vehicle enters the current area, the display screen at the entrance of the current area will delete the vehicle information and adjust the current capacity of the area, the queue number of the vehicle behind, and the number of queues.

[0045] The detection equipment includes but is not limited to the road test unit, laser radar, millimeter wave radar and high-definition camera. The detection equipment is distributed on both sides of the road in the automobile test area to collect vehicle data information, and uses ultra-wideband technology to send the collected data information to the data acquisition module.

[0046] It should be noted that the road test unit refers to the RSU. The RSU is a device installed on the roadside in the ETC system that uses DSRC technology to communicate with the on-board unit OBU and realize vehicle identification and electronic deduction. The RSU is composed of a high-gain directional beam-controlled read-write antenna and a radio frequency controller. The high-gain directional beam-controlled read-write antenna is a microwave transceiver module that is responsible for sending / receiving, modulating / demodulating, encoding / decoding, and encrypting / decrypting signals and data; the radio frequency controller is a module that controls the transmission and reception of data and processes the transmission and reception of information to the upper computer.

[0047] Ultra-wideband technology refers to UWB. UWB technology is a wireless carrier communication technology that can integrate positioning and communication. It has strong penetration ability and can accurately locate indoors and underground. It does not use a sinusoidal carrier, but uses nanosecond non-sinusoidal narrow pulses to transmit data, so it occupies a wide spectrum range. UWB technology has the advantages of low system complexity, low power spectrum density of transmitted signals, insensitivity to channel fading, low interception capability, and high positioning accuracy. It is especially suitable for high-speed wireless access in dense multipath places such as indoors.

[0048] LiDAR and millimeter-wave radar are radar systems that emit lasers and millimeter waves to detect the position, speed and other characteristic quantities of the target. Among them, LiDAR can achieve extremely high resolution, ensure the accuracy of detection, and achieve precise measurement. Millimeter-wave radar, on the other hand, can work stably in different weather and lighting conditions by emitting electromagnetic waves with a frequency between microwaves and infrared, has strong penetration, and is not easily affected by environmental factors such as rain and fog.

[0049] The images taken by high-definition cameras are clear, with realistic color reproduction, strong environmental adaptability and longer life. One high-definition camera can achieve the monitoring effect of several ordinary cameras, with a wider and wider monitoring coverage, higher image resolution and richer information. Therefore, the use of high-definition monitoring can effectively reduce the scale of the system, avoid complicated video images, control and power supply transmission lines, and thus reduce the overall construction cost. High-definition cameras use advanced photosensitive devices to make image details clearer and more delicate, and can provide better image quality.

[0050] The collection of vehicle data information includes: identifying the vehicle's on-board unit through the road test unit, capturing pictures and recording videos of the scene through a high-definition camera, and establishing a three-dimensional point cloud map of the test area through lidar and millimeter-wave radar.

[0051] The on-board unit refers to OBU, which uses DSRC technology and is a microwave device that communicates with RSU. In the ETC system, OBU is placed on the vehicle and roadside unit RSU is set up on the roadside, and they communicate with each other through microwaves. When the vehicle passes through the road, the road test unit will identify the vehicle's on-board unit and confirm the vehicle's identity.

[0052] Before the vehicle enters the test area, the road test unit identifies the vehicle's on-board unit and registers the vehicle's vehicle identification code. The on-board unit is bound to the vehicle identification code, and the vehicle identification code is obtained synchronously when the detection equipment collects vehicle data information.

[0053] High-definition cameras, laser radars, and millimeter-wave radars collect data on passing vehicles and the surrounding environment, including video data, the shape and size of surrounding obstacles, and physical data of various vehicles, including but not limited to speed, shape, direction of travel, and distance between vehicles. Laser radars and millimeter-wave radars can create a visual 3D point cloud map that can not only display the moving vehicle, but also the surrounding environment and obstacles.

[0054] The data acquisition module includes an identification unit and a positioning unit. The identification unit confirms the queue number of the vehicle according to the vehicle's on-board unit. The positioning unit receives the location information of the detection equipment through ultra-wideband technology and obtains the positioning information of each vehicle according to the relative position between the detection equipment and each vehicle.

[0055] The data acquisition module includes an identification unit and a positioning unit. The identification unit is used to organize the video data recorded by the high-definition camera and identify the vehicle identity through the road test unit.

[0056] The positioning unit is used to obtain the positioning of each detection device, and obtain the positioning information of each vehicle based on the relative position of the detection device and each vehicle in the point cloud map provided by the laser radar or millimeter wave radar.

[0057] The positioning unit reports the vehicle positioning information in real time, making it easier for the platform to identify the vehicle and assist in determining the location.

[0058] The data processing module is used to receive the vehicle data information uploaded by the data acquisition module and identify the current status of each vehicle in the automobile test area.

[0059] Preferably, the data processing module can use edge computing technology to count the entry and exit of vehicles in each area, and the steps are as follows:

[0060] Data classification: After obtaining the vehicle data information collected by the detection equipment, the vehicle data information of different vehicles is distinguished through different on-board units;

[0061] Model training: Pre-select vehicle data information in different states and perform feature extraction. The vehicle data information first enters the convolution kernel for convolution calculation, and then enters the global pooling for global pooling processing to obtain the features of vehicles in different states and establish a mapping relationship between vehicle states and vehicle features.

[0062] Feature recognition: Input the vehicle data collected by the detection equipment into the model to obtain the vehicle's features, and then use the mapping relationship to determine the vehicle's current status;

[0063] Data statistics: Count the number of vehicles and arrange for vehicles to enter or leave the corresponding area based on the current status of each vehicle and the capacity of the corresponding area;

[0064] The characteristics of vehicles in different states include, but are not limited to: entering the test area, undergoing testing, waiting in line, and leaving the test area.

[0065] It should be noted that the current status of the vehicle refers to whether the vehicle is in a waiting state, driving state, or testing state. It can also include the state of vehicle failure, or the state of illegal parking on the line, etc. According to the different states of the vehicle, the entry and exit of the vehicle can be planned. For example, if the current vehicle is about to leave the test area, the exit gate can be opened to release it, and the entrance gate can be opened at the same time to remind the vehicles in the queue to enter the test area.

[0066] In addition, when the front and rear vehicles enter the detection range of the detection equipment, the identities of the drivers in each vehicle are identified and the driving habits of the drivers are obtained. For example, the driver's historical driving trajectory when turning is obtained to predict the driver's running trajectory at the next moment and determine whether there is a risk of collision between the two vehicles.

[0067] It is also possible to predict the vehicle's trajectory to determine whether it is in the process of making illegal behaviors such as making random U-turns, parking on the road, cutting in, driving in the wrong direction, etc.

[0068] It should be noted that the edge computing technology here preferably adopts multi-access edge computing MEC, which refers to a building block combination of connection, computing, capabilities, and applications. By providing the services and cloud computing functions required by users nearby, a service environment with high performance, low latency and high bandwidth is created, which accelerates the rapid download of various content, services and applications in the network, allowing consumers to enjoy an uninterrupted high-quality network experience.

[0069] In one embodiment, the data processing module is also used to establish lane line detection and vehicle detection, and collect the driver's driving habits through big data. According to the vehicle data information of each vehicle in the video, the lane line detection and vehicle detection are combined to predict the running trajectory of each vehicle, mark each vehicle before it violates the rules, and generate warning information for broadcast; violations include but are not limited to arbitrary U-turns, parking in the lane, cutting in, driving in the wrong direction, etc.

[0070] It should be noted that the combination of lane line detection and vehicle detection can establish a three-dimensional spatial map, which can monitor in real time whether the vehicle has collisions or other violations. On this basis, combined with big data analysis of the driver's driving habits, it is possible to predict the vehicle's driving trajectory at the next moment. These are two different processes. On the one hand, it can monitor whether a violation has occurred at the current moment, and on the other hand, it can predict whether a violation may occur at the next moment.

[0071] Furthermore, lane line detection includes: capturing an image of the lane lines in the car test area through a high-definition camera, converting the image into grayscale, selecting a number of highlighted pixels in the grayscale image, dividing a center line in the original lane line image, the road line to the left of the center line is the inner line, and the road line to the right of the center line is the outer line, the corresponding pixel points to the left of the center line in the grayscale image belong to the inner line, and the pixel points to the right of the center line belong to the outer line; connecting adjacent pixel points in the inner line, connecting adjacent pixel points in the outer line, removing abnormal points in the pixel points, making the inner line and the outer line tend to be straight lines, and displaying them in the form of a top view and a front view.

[0072] Generally speaking, lane lines on the road are depicted using high-brightness colors. Since lane lines belong to the highlight area, setting a threshold in the grayscale image can retain pixels with higher grayscale. Then extract the pixel points on the lane lines. By setting a center line in the original lane line image, different lane lines can be distinguished. This center line will distinguish the pixel points on different lane lines in the grayscale image. In this way, the lane lines can be extracted into three-dimensional space. Preferably, a three-dimensional coordinate system can also be established, and the three-dimensional coordinates of all pixel points can be marked. The pixel points on the same lane line should satisfy a linear relationship. If a pixel point does not satisfy the linear relationship, it is considered to be an abnormal point and is removed.

[0073] Furthermore, vehicle detection includes: capturing images of vehicles in the car test area with a high-definition camera, extracting features of the vehicle, tracking the vehicle based on the features of the vehicle, and restoring the lane lines in the same proportion in the top view and the front view based on the features of the vehicle combined with the three-dimensional point cloud map of the vehicle.

[0074] Based on the position of each vehicle in the overhead view and front view of the lane line, it is determined whether the vehicle has violated the regulations.

[0075] Lane detection and vehicle detection both belong to perception algorithms. Perception algorithms are a technology that uses sensor data to detect and measure objects. They are supervised learning algorithms that require a loss function to adjust the parameter model.

[0076] Lane line detection mainly involves image extraction and visualization. After inputting the image, the data processing module can convert the front view into a top view in real-world coordinates, restore the position of the lane line in the real world, and then extract and convert the region of interest into grayscale. After the conversion, the image is denoised by weighted filtering and averaging. Based on the high brightness characteristics of lane line pixels, a threshold is set in the grayscale image to retain pixels with higher grayscale and extract the pixels on the lane line. Finally, the relative position relationship between these curves and the vehicle is determined based on the center line of the image, as well as whether the lane line is the lane line of the lane.

[0077] Vehicle detection uses a high-definition camera to capture images of vehicles in the vehicle test area, extract vehicle features, track vehicles based on their features, and restore them in the top view and front view of the lane line in the same proportion based on the vehicle features combined with the vehicle's three-dimensional point cloud map. The purpose of vehicle detection is to track each vehicle.

[0078] Based on the position of each vehicle in the overhead view and front view of the lane line, determine whether the vehicle has violated the rules, including but not limited to making random U-turns, parking on the lane, cutting in, driving in the wrong direction, etc.

[0079] In another embodiment, the data processing module can also be used to monitor the violations of each vehicle through AI recognition technology, and determine in real time whether the vehicle has violated the regulations, including:

[0080] A large number of pictures of vehicle violations are obtained through high-definition cameras, and training sets and test sets are established. These training sets and test sets are all from pictures of vehicle violations and are randomly assigned. The training set is input to train the deep learning model. After the training is completed, the test set is input to the deep learning model to calculate the accuracy of deep learning on the test set.

[0081] In fact, deep models need to be trained and tested for accuracy, and a large number of pictures of violations need to be collected for artificial intelligence to learn. This method is similar to the feature extraction method mentioned above. Both require neurons to mine the deep features or data of the pictures. The advantage of artificial intelligence lies in iterative learning, which can improve the accuracy of judging violations through repeated identification.

[0082] Specifically, the deep learning model is composed of a back-propagation neural network, which includes an input layer, multiple hidden layers and an output layer.

[0083] Through AI recognition technology, real-time monitoring and early warning of illegal behaviors such as random U-turns, parking on the road, cutting in, and driving against traffic are carried out, effectively improving the efficiency of road vehicle transportation, reducing the occurrence of emergencies during road vehicle behavior, alleviating urban road traffic pressure, and ensuring traffic safety. AI recognition technology replaces human eyes, without the need for additional hardware, and real-time monitoring and identification of various traffic violations is quick and convenient.

[0084] After the model is built, the mean square error loss can be used to construct the loss function, and the calculation formula is:

[0085]

[0086] Where: n samples represents the total number of samples, y i is the true value of the i-th sample, is the predicted value of the ith sample, represents the square of the difference between the true value and the predicted value of the i-th sample, and the right side of the equation represents the sum of the differences between the true value and the predicted value of all samples and the average value.

[0087] Optionally, the accuracy, precision and recall of the model can also be calculated. The prediction results of the model can be divided into TP, TN, FP and FN according to whether they are correct or not. TP refers to the number of samples where the sample belongs to the vehicle with illegal behavior and the model determines that it is illegal; TN refers to the number of samples where the sample belongs to the vehicle with illegal behavior but the model determines that it is not illegal; FP refers to the number of samples where the sample does not belong to the vehicle with illegal behavior but the model determines that it is illegal; FN refers to the number of samples where the sample does not belong to the vehicle with illegal behavior and the model determines that it is not illegal.

[0088] Accuracy = (TP+TN) / (TP+FN+FP+TN);

[0089] Precision = TP / (TP+FP);

[0090] Recall = TP / (TP+FN).

[0091] Precision can be considered as the probability that the classifier does not make mistakes in classifying positive samples, and recall can be considered as the probability that the classifier does not miss positive samples.

[0092] In addition, the data processing module determines whether vehicles can be released and which area's vehicles can be released based on the perception data, scheduling status, test load rate, and vehicle queuing time of each braking test area, and links the facilities in the area for overall control.

[0093] Furthermore, the display module is used to display the vehicle queue situation, indicate the vehicle entry and exit and make announcements.

[0094] The display module establishes communication with the data processing module. The display module includes multiple display devices, including but not limited to display screens, signal lights and intelligent voice devices.

[0095] The display screen displays the capacity of the car testing area and queue information, which includes but is not limited to driver information, queue number, vehicle identification code, queue status, etc.

[0096] Traffic lights are used to indicate the entry and exit of vehicles in each area.

[0097] After receiving the processing results generated by the data processing module, the intelligent voice device notifies the vehicle of entry and exit, and after receiving the judgment of the data processing module that the vehicle has violated the regulations, it announces the violation of the vehicle.

[0098] Display screens are usually placed at the entrance of the test area to facilitate vehicles waiting for the test to understand the current queue situation, while signal lights are placed at intersections to prompt vehicles to pass. Signal lights usually operate synchronously with the gate. When a vehicle needs to enter or leave, the gate opens and the signal light flashes.

[0099] Intelligent voice devices are placed at various road sections and entrances and exits. They can be used to remind queuing vehicles of the current queue status of vehicles, and can also inform each vehicle that a vehicle on a certain road section has broken down during driving. In addition, they can also remind drivers of violations or remind drivers of vehicle crossing the line.

[0100] Furthermore, the control module is used to obtain the current status of each vehicle identified by the data processing module, control the display module to display, and release or prohibit the vehicle from passing.

[0101] After receiving the processing results generated by the data processing module, the control module controls the gate to release or prohibit passage, and controls the display screen and signal lights to display.

[0102] In the technical solution of the embodiment of the present invention, a lightweight and high-performance solution of UWB ultra-wideband technology can be adopted, which integrates ranging, 2D and 3D positioning, and data transmission functions. With high-precision ranging and positioning, it meets the needs of various high-performance application scenarios. When a vehicle violates the rules or a vehicle failure affects the test safety of other vehicles, the display screen and the mobile terminal can be linked in time to display the violation of the vehicle for reminder, and the gate can be linked to suspend release, and the signal light can be linked to warn. At the same time, it can be combined with perception data, electronic fences, and positioning data to determine that the vehicle is leaving the site and is ready to release the vehicles in the queue. There is no need to wait for the vehicle to completely leave the site before starting to release the waiting vehicle, which speeds up the vehicle test cycle and improves operational efficiency. At the same time, in order to consider the abnormal failures that may occur in the vehicle during the braking test, the test lane is released at 50% or 30% of the site load according to the volume of the test vehicle model and the braking test scenario, which can increase the efficiency by more than 3 times.

[0103] Furthermore, the system also includes a database for storing driver information, vehicle information, and vehicle violation information. The violation data is included in the database, and driver safety management, personal credit, and corporate credit management are strengthened to form a closed loop with the safety management system.

[0104] The technical solution of the embodiment of the present invention realizes the collection and analysis of vehicle data in the automobile testing area, realizes the control of vehicle quantity, monitoring of safety trajectories, and provides related auxiliary operation services through various devices such as road test units, lidars, millimeter-wave radars, high-definition cameras, display screens, traffic lights, intelligent voice devices, gates, and various technologies such as ultra-wideband technology, big data, multi-access edge computing, perception algorithms, and AI recognition technology.

[0105] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A vehicle management and control system for an automobile test area, characterized in that: The system includes: detection equipment, data acquisition module, data processing module, display module and control module; The data acquisition module is used to receive vehicle data information collected by multiple detection devices, register the identity of the driver in the waiting area and the vehicle identification code, and assign a queue number to the vehicle; the detection equipment includes but is not limited to a road test unit, a laser radar, a millimeter wave radar and a high-definition camera; The data processing module is used to receive the vehicle data information uploaded by the data acquisition module and identify the current status of each vehicle in the automobile test area; The display module is used to display the vehicle queue situation, indicate the vehicle entry and exit and broadcast; The control module is used to obtain the current status of each vehicle identified by the data processing module, control the display module to display, and release or prohibit the vehicle from passing.

2. A vehicle management and control system for an automobile test area as claimed in claim 1, characterized in that: The detection equipment is distributed on both sides of the road in the automobile test area to collect vehicle data information, and uses ultra-wideband technology to send the collected vehicle data information to the data acquisition module; The collection of vehicle data information includes: identifying the vehicle's on-board unit through the road test unit, capturing pictures and recording videos of the scene through high-definition cameras, and establishing a three-dimensional point cloud map of the test area through laser radar and millimeter-wave radar; The data acquisition module includes an identification unit and a positioning unit. The identification unit confirms the queue number of the vehicle according to the vehicle-mounted unit. The positioning unit receives the position information of the detection equipment through ultra-wideband technology and obtains the positioning information of each vehicle according to the relative position between the detection equipment and each vehicle.

3. A vehicle management and control system for an automobile test area as claimed in claim 2, characterized in that: The data processing module uses edge computing technology to count the entry and exit of vehicles in each area. The steps are as follows: Data classification: After obtaining the vehicle data information collected by the detection equipment, the vehicle data information of different vehicles is distinguished through different on-board units; Model training: Pre-select vehicle data information in different states and perform feature extraction. The vehicle data information first enters the convolution kernel for convolution calculation, and then enters the global pooling for global pooling processing to obtain the features of vehicles in different states and establish a mapping relationship between vehicle states and vehicle features. Feature recognition: Input the vehicle data collected by the detection equipment into the model to obtain the vehicle's features, and then use the mapping relationship to determine the vehicle's current status; Data statistics: Count the number of vehicles and arrange for vehicles to enter or leave the corresponding area based on the current status of each vehicle and the capacity of the corresponding area; The characteristics of vehicles in different states include, but are not limited to: entering the test area, undergoing testing, waiting in line, and leaving the test area.

4. A vehicle management and control system for an automobile test area as claimed in claim 3, characterized in that: The data processing module is also used to establish lane line detection and vehicle detection, and collect the driving habits of drivers through big data, combine lane line detection and vehicle detection, predict the running trajectory of each vehicle according to the vehicle data information of each vehicle in the video, mark each vehicle before it violates the rules, and generate warning information for broadcast; the violations include but are not limited to arbitrary U-turns, parking on the road, cutting in, and driving in the wrong direction; The lane line detection includes: capturing a picture of the lane line in the car test area by a high-definition camera, performing grayscale conversion on the picture, selecting a number of highlighted pixels in the grayscale picture, dividing a center line in the original lane line picture, the road line on the left side of the center line is the inner line, and the road line on the right side of the center line is the outer line, the corresponding pixel points on the left side of the center line in the grayscale picture belong to the inner line, and the pixel points on the right side of the center line belong to the outer line; connecting the adjacent pixel points in the inner line, connecting the adjacent pixel points in the outer line, removing abnormal points in the pixel points, so that the inner line and the outer line tend to be straight lines, and displaying them in the form of a top view and a front view; The vehicle detection includes: capturing images of vehicles in the vehicle test area through a high-definition camera, extracting features of the vehicles, tracking the vehicles according to the features of the vehicles, and restoring the lane lines in the top view and the front view in the same proportion according to the features of the vehicles combined with the three-dimensional point cloud of the vehicles; Based on the position of each vehicle in the overhead view and front view of the lane line, it is determined whether the vehicle has violated the regulations.

5. The vehicle management and control system for an automobile test area as claimed in claim 1, characterized in that: The data processing module is also used to monitor the violations of each vehicle through AI recognition technology and determine in real time whether the vehicle has violated the regulations, including: Using a high-definition camera to obtain images of vehicle violations, a training set and a test set are established. Both the training set and the test set are derived from the images of vehicle violations and are randomly assigned. The training set is input to train the deep learning model. After the training is completed, the test set is input to the deep learning model to calculate the accuracy of the deep learning model on the test set. The deep learning model is composed of a back-propagation neural network, which includes an input layer, multiple hidden layers and an output layer.

6. The vehicle management and control system for an automobile test area as claimed in claim 3, characterized in that: The display module establishes communication with the data processing module, and the display module includes multiple display devices, including but not limited to display screens, signal lights and intelligent voice devices; The display screen displays the capacity of the vehicle testing area and queue information, including but not limited to driver information, queue number, vehicle identification number, and queue status; The signal lights are used to indicate the entry and exit of vehicles in each area; After receiving the processing result generated by the data processing module, the intelligent voice device notifies the vehicle to enter or exit, and after receiving the judgment from the data processing module that the vehicle has violated the regulations, the intelligent voice device announces the violation of the regulations.

7. A vehicle management and control system for an automobile test area as claimed in claim 6, characterized in that: After receiving the processing result generated by the data processing module, the control module controls the gate to release or prohibit passage, and controls the display screen and signal lights to display.

8. The vehicle management and control system for an automobile test area as claimed in claim 1, characterized in that: The system also includes a database for storing driver information, vehicle information and vehicle violation information.

Images