Vehicle positioning management method and system

Abstract

The invention discloses a vehicle positioning management method and system, and the method comprises the steps: collecting vehicle driving state data in real time, collecting a vehicle surrounding environment image through a vehicle-mounted camera, and carrying out the preprocessing of the image, and obtaining multi-modal data; performing fusion analysis on the multi-modal data, identifying a dangerous driving behavior of the vehicle through a vehicle anti-fatigue early warning system DSM and performing real-time positioning processing on the vehicle, and generating early warning information if the dangerous driving behavior is identified; the early warning information is converted into voice based on an advanced driving assistance system (ADAS), voice early warning is played to a driver through a vehicle-mounted loudspeaker, the driver is reminded to take measures in time, meanwhile, the early warning information is displayed on a vehicle-mounted display screen in a text mode, and corresponding measures are taken through a management center platform to manage a vehicle. According to the invention, real-time positioning and management of the running state of the vehicle are realized, and the driving safety is effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle intelligent management, and in particular to a vehicle positioning management method and system. Background Art

[0002] In the vehicle positioning management system, how to achieve real-time warning and evidence collection of dangerous driving behaviors of vehicles is a key technical issue. First, the system needs to collect the driving status data of the vehicle in real time through on-board sensors and cameras, including vehicle speed, acceleration, steering angle, lane departure, etc., and analyze and judge these data through intelligent algorithms to identify dangerous driving behaviors of vehicles, such as lane departure, too close distance, and sudden braking. Secondly, after identifying dangerous driving behaviors, the system needs to issue voice warning prompts to the driver in time to remind the driver to pay attention to safety and take timely measures. At the same time, the system also needs to upload the relevant data and evidence of dangerous driving behaviors to the management platform for subsequent accident analysis and responsibility determination. Finally, considering that the vehicle may be traveling across the country, the terminal equipment and the management center also need to have a stable and reliable voice intercom function to facilitate timely communication and coordination in emergency situations. How to achieve efficient, accurate, and real-time data collection, analysis, and warning in a vehicle-mounted environment, and ensure the stability and reliability of data upload and voice intercom, is a complex technical challenge faced by the vehicle positioning management system. Summary of the invention

[0003] In order to solve the technical problems existing in the above-mentioned prior art, the present invention proposes a vehicle positioning management method and system to achieve real-time monitoring of the vehicle operation status and risk warning.

[0004] On the one hand, to achieve the above-mentioned purpose, the present invention provides a vehicle positioning management method, comprising:

[0005] Collect vehicle driving status data in real time, and use the on-board camera to collect images of the vehicle's surrounding environment, pre-process the images, and obtain multi-modal data;

[0006] Perform fusion analysis on the multimodal data, identify dangerous driving behaviors of the vehicle through the vehicle fatigue warning system DSM and perform real-time positioning processing on the vehicle, and generate warning information if dangerous driving behaviors are identified;

[0007] Based on the advanced driving assistance system ADAS, the warning information is converted into voice, and the voice warning is played to the driver through the vehicle speaker to remind the driver to take timely measures. At the same time, the warning information is displayed in text form on the vehicle display screen, and corresponding measures are taken through the management center platform to manage the vehicle.

[0008] On the other hand, to achieve the above object, the present invention also provides a vehicle positioning management system, comprising:

[0009] The data acquisition module is used to collect vehicle driving status data in real time, and at the same time use the vehicle-mounted camera to collect images of the vehicle's surrounding environment, pre-process the images, and obtain multi-modal data;

[0010] An intelligent analysis and warning module is used to perform fusion analysis on the multimodal data, identify dangerous driving behaviors of the vehicle through the vehicle fatigue warning system DSM, and perform real-time positioning processing on the vehicle. If dangerous driving behaviors are identified, warning information is generated;

[0011] The vehicle management module is used to convert the warning information into voice based on the advanced driving assistance system ADAS, play the voice warning to the driver through the vehicle speaker, remind the driver to take timely measures, and display the warning information in text form on the vehicle display screen, and take corresponding measures to manage the vehicle through the management center platform.

[0012] Compared with the prior art, the present invention has the following advantages and technical effects:

[0013] (1) The present invention collects vehicle driving status data and environmental images in real time through the vehicle-mounted terminal device, and uses deep learning algorithms to integrate and analyze the data to identify dangerous driving behaviors. When dangerous behaviors are detected, the system automatically generates voice and text warning information to remind the driver, and uploads relevant data to the management center platform for further analysis. The present invention also supports real-time voice intercom function between the vehicle-mounted terminal and the management center, which facilitates timely communication in emergency situations.

[0014] (2) Through the distributed architecture that combines edge computing and cloud computing, the present invention realizes real-time positioning and management of vehicle operating status, effectively improves driving safety, and provides a reliable basis for accident analysis and handling. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0016] Figure 1 A flow chart of a vehicle positioning management method according to an embodiment of the present invention;

[0017] Figure 2 The figure is a schematic diagram of the structure of a vehicle positioning management system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0018] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0019] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0020] The present invention proposes a vehicle positioning management method, such as Figure 1 ,include:

[0021] Collect vehicle driving status data in real time, and use the on-board camera to collect images of the vehicle's surrounding environment, pre-process the images, and obtain multi-modal data;

[0022] Fusion analysis of multimodal data is performed to identify dangerous driving behaviors of vehicles through the vehicle fatigue warning system DSM and perform real-time positioning of the vehicle. If dangerous driving behaviors are identified, warning information is generated;

[0023] Based on the advanced driving assistance system ADAS, the warning information is converted into voice, and the voice warning is played to the driver through the vehicle speaker to remind the driver to take timely measures. At the same time, the warning information is displayed in text form on the vehicle display screen, and corresponding measures are taken through the management center platform to manage the vehicle.

[0024] Furthermore, real-time collection of vehicle driving status data includes:

[0025] The vehicle speed, acceleration, steering angle and lane departure status data are acquired from the vehicle electronic control unit ECU in real time through the CAN bus interface of the vehicle terminal device, and the vehicle surrounding environment images are collected through the vehicle camera;

[0026] The collected environmental images are preprocessed, including image denoising, enhancement and correction processing. The preprocessed image data is time-synchronized with the driving status data obtained from the CAN bus to construct multimodal data.

[0027] Specifically, the vehicle terminal device obtains driving status data including vehicle speed, acceleration, steering angle, lane departure, etc. from the vehicle ECU in real time through the CAN bus interface, and the vehicle camera collects images of the vehicle's surrounding environment. The collected environmental images are preprocessed, including image denoising, enhancement, correction, etc., to improve image quality and provide high-quality input for subsequent intelligent algorithm analysis. The preprocessed image data is synchronized with the driving status data obtained from the CAN bus to construct multimodal data. The method of constructing multimodal data can provide richer and more accurate information for subsequent analysis, which helps to improve the analysis effect of the algorithm. The synchronized data can reflect the vehicle's status and surrounding environment at the same time, thereby providing support for accurate judgment of the vehicle's driving conditions.

[0028] Furthermore, the data is integrated and analyzed, and the vehicle's dangerous driving behavior is identified through the vehicle fatigue warning system DSM, including:

[0029] Acquiring vehicle speed, steering wheel angle, acceleration and other driving parameters based on the vehicle's driving status is the core of the intelligent driving assistance system ADAS. The speed sensor measures the wheel speed and calculates that the vehicle's current speed is 80 km / h; the steering wheel angle sensor records that the steering wheel has turned 15 degrees; and the accelerometer senses that the vehicle's horizontal acceleration is 0.5g. These parameters are collected at a frequency of multiple times per second to ensure the real-time and accuracy of the data.

[0030] The vehicle anti-fatigue warning system DSM uses a convolutional neural network algorithm based on deep learning to perform feature extraction and fusion analysis on the multimodal data to obtain a comprehensive feature vector;

[0031] Input the comprehensive feature vector into the pre-trained dangerous driving behavior recognition model for classification and judgment, and output the judgment result;

[0032] According to the judgment result of the dangerous driving behavior identification model, it is determined whether the vehicle has dangerous driving behavior.

[0033] Specifically, the vehicle fatigue warning system DSM uses a deep learning-based convolutional neural network algorithm to extract and fuse the preprocessed vehicle driving status data and environmental image data to obtain a comprehensive feature vector. The extracted comprehensive feature vector is input into a pre-trained dangerous driving behavior recognition model for classification and judgment. The model is trained with a large amount of dangerous driving behavior sample data. According to the judgment results of the dangerous driving behavior recognition model, it is determined whether the vehicle has dangerous driving behaviors such as lane deviation, too close distance, and sudden braking.

[0034] The convolutional neural network (CNN) based on deep learning can automatically extract features in images, such as edges, textures, and shapes, through multi-layer convolution and pooling operations. For example, when identifying lane lines, CNN can extract the color, shape, and position features of lane lines. At the same time, driving status data such as speed and acceleration can also be converted into numerical feature vectors through feature extraction. These feature vectors are fused to form a comprehensive feature vector containing rich multi-dimensional information. The extracted comprehensive feature vector is input into a pre-trained dangerous driving behavior recognition model for classification and judgment. The model is trained with a large amount of dangerous driving behavior sample data and can recognize a variety of dangerous behaviors. For example, the model may contain multiple classifiers for identifying behaviors such as lane departure, too close distance, and sudden braking. Each classifier is trained based on a large amount of labeled data to ensure the accuracy of recognition. According to the judgment results of the dangerous driving behavior recognition model, it is determined whether the vehicle has dangerous driving behaviors such as lane departure, too close distance, and sudden braking. For example, if the model determines that the vehicle currently has lane departure behavior, it may be because the steering wheel angle sensor data shows that the steering wheel deviates from the normal driving trajectory during the vehicle's driving, and the lane line recognition results in the image data also show that the vehicle has deviated from the lane. Combined with this information, the model can accurately determine lane departure behavior. If a dangerous driving behavior is identified, a warning message of the corresponding level is generated according to the degree of danger of the behavior. Warning messages are divided into multiple levels, such as low risk, medium risk and high risk. For example, a slight lane deviation may generate a low-risk warning, while a close distance and high speed may generate a high-risk warning. The generation of warning information is based on preset thresholds and rules to ensure the timeliness and accuracy of the warning.

[0035] Furthermore, real-time positioning of the vehicle is performed, including:

[0036] The GPS positioning terminal can be used to check patrols in areas where dangerous driving behaviors occur in real time, and dispatch and command orders can be quickly issued through the management center platform.

[0037] Furthermore, a voice warning is played to the driver through the vehicle speaker, including:

[0038] For the identified dangerous driving behavior, the warning information text matching the type of dangerous driving behavior is obtained from the preset warning information library, and the urgency of the warning information is determined according to the degree of danger;

[0039] The ADAS converts the acquired warning information text into a voice signal through speech synthesis technology, and adjusts the volume and speech speed of the voice signal according to the urgency of the warning information;

[0040] The synthesized warning voice signal is transmitted to the vehicle-mounted speaker, and the warning voice is played through the vehicle-mounted speaker. At the same time, the warning information text is transmitted to the vehicle-mounted display screen and displayed.

[0041] Specifically, the generated warning information is pushed to the driver in real time through various means such as the on-board display screen and voice broadcast system. Low-risk warnings can be prompted by icons on the display screen, while high-risk warnings sound an alarm through the voice broadcast system to remind the driver to pay attention. Warnings in various ways ensure that drivers can receive warning information in a timely manner in different situations and take corresponding measures to avoid accidents. In this way, the on-board terminal equipment can comprehensively utilize multiple sensor data and intelligent algorithms to identify dangerous driving behaviors in real time, generate warning information, and effectively improve driving safety. The application of this multimodal data fusion and deep learning algorithm not only improves the accuracy of recognition, but also provides reliable warnings in complex driving environments and reduces the probability of traffic accidents.

[0042] Furthermore, if the risk level of the identified dangerous driving behavior exceeds a preset safety threshold, the vehicle's active safety control strategy is triggered, and corresponding safety measures are taken according to the type of dangerous driving behavior;

[0043] Continuously monitor the vehicle's driving status, obtain driving parameters in real time and input them into the random forest model to identify dangerous driving behaviors, and dynamically adjust the warning strategy based on the identification results.

[0044] Specifically, the vehicle terminal device collects vehicle driving status data in real time through sensors, including vehicle speed sensors, acceleration sensors, and steering angle sensors. For example, the vehicle speed sensor can monitor the vehicle speed in real time through wheel speed or GPS signals, the acceleration sensor can sense the acceleration changes of the vehicle in all directions, and the steering angle sensor records the steering wheel angle information. The acquisition frequency of these data usually ranges from tens to hundreds of times per second to ensure the real-time and accuracy of the data. According to the preset dangerous driving behavior judgment rules, the system will analyze the collected data in real time. For example, if the vehicle speed rises rapidly to 120% of the legal speed limit in a short period of time, or the acceleration exceeds the safety threshold (in this embodiment, the acceleration safety threshold is 0.5), the system will judge it as speeding or rapid acceleration; if the steering angle changes by more than 45 degrees in a very short period of time, it may be regarded as a sharp turn. These rules are based on a large amount of driving behavior data and traffic accident statistical analysis, and are intended to identify dangerous driving behaviors that may lead to accidents. Once dangerous driving behavior is identified, the vehicle camera will immediately start to collect environmental image data inside and outside the vehicle. The in-car camera can capture the driver's status, such as whether he is distracted or tired, while the out-car camera records the environmental conditions around the vehicle, such as road conditions and the location of other vehicles. These image data provide important visual evidence for subsequent analysis and processing. The generated warning information will be customized according to the type and degree of dangerous driving behavior. For example, if the system recognizes that the driver frequently changes lanes while driving at high speed, the warning message may be "Please note that there is a greater risk of changing lanes at high speeds, please stay in the lane." The generation of warning information is based on a preset warning information library, which contains warning texts corresponding to a variety of dangerous driving behaviors to ensure the accuracy and timeliness of the information.

[0045] Furthermore, active safety control is used to pre-set driving routes for vehicles with dangerous driving behaviors. When the task begins, the vehicle's route and status begin to be monitored and recorded. When the vehicle does not follow the preset route or drives out of the set area, the system will automatically alarm and notify the management center platform to take corresponding measures based on the actual situation.

[0046] Furthermore, when dangerous driving behavior is identified, the vehicle driving status data, environmental image data and warning information collected by the vehicle terminal device are packaged in a predetermined format to generate a data packet;

[0047] Encrypt the generated data packets, use an asymmetric encryption algorithm, and use the public key of the management center platform to encrypt the data packets;

[0048] The encrypted data packet is uploaded to the management center platform through the wireless communication module of the vehicle-mounted terminal device. After receiving the uploaded data packet, the management center platform uses the private key to decrypt the data packet, restore the original vehicle driving status data, environmental image data and warning information, and store and analyze the decrypted data. Through big data analysis technology and machine learning algorithms, dangerous driving behaviors are counted and analyzed to provide decision support for relevant departments.

[0049] Specifically, the vehicle terminal device packages the collected vehicle driving status data, environmental image data and warning information in a predetermined format. The format of the data packet usually includes a data header, a data body and a check code. The data header contains information such as the data type and timestamp, and the data body contains specific numerical values ​​and image data. The check code is used to verify whether the data has been tampered with during transmission. In order to ensure the security of data transmission, the system will encrypt the generated data packet using an asymmetric encryption algorithm, such as the RSA algorithm. The data packet is encrypted using the public key of the management center platform, so that only the management center platform with the corresponding private key can decrypt the data packet to ensure that the data will not be illegally intercepted and tampered with during transmission. The encrypted data packet is uploaded to the management center platform through the wireless communication module of the vehicle terminal device. The wireless communication module usually supports multiple communication protocols, such as 4G / 5G, Wi-Fi, etc. Reliable transmission protocols, such as TCP protocol, are used during the upload process to ensure the integrity and reliability of data transmission. The TCP protocol establishes a connection through a three-way handshake, and performs data verification and retransmission mechanisms during transmission, effectively avoiding data loss and disorder problems. After receiving the uploaded data packet, the management center platform uses the private key to decrypt the data packet and restore the original vehicle driving status data, environmental image data and warning information. The decrypted data will be stored in the database for subsequent analysis and processing. By storing and analyzing these data, using big data analysis technology and machine learning algorithms, dangerous driving behaviors can be counted and analyzed. For example, cluster analysis can find that dangerous driving behaviors are high in certain sections or time periods, and classification algorithms can predict the probability of dangerous driving under specific conditions for specific drivers. These analysis results can provide decision support for traffic management departments, such as optimizing traffic light settings and strengthening patrols on specific sections. In addition, through the analysis of a large amount of data, the preset dangerous driving behavior judgment rules and warning information database can be continuously optimized to improve the recognition accuracy and warning effect of the system. For example, if the warning information of a certain type of dangerous driving behavior fails to effectively reduce the accident rate through analysis, the expression of the warning information can be adjusted or the urgency of the voice prompt can be increased to enhance the driver's vigilance. Through this comprehensive data collection, transmission, analysis and feedback mechanism, the on-board terminal equipment and the management center platform jointly build an efficient safety warning and active safety control system, which effectively improves the level of driving safety and reduces the occurrence of traffic accidents.

[0050] Furthermore, a voice intercom channel is established between the vehicle-mounted terminal equipment and the management center platform through the wireless communication network, VoIP technology is used to achieve real-time voice communication, and voice quality is guaranteed by echo cancellation and noise suppression algorithms. When an emergency occurs, the driver can communicate and coordinate with the management center in a timely manner through the voice intercom function.

[0051] Specifically, the vehicle terminal device is connected to the mobile communication network through the wireless network module and establishes a data communication link with the management center platform. The management center platform obtains the IP address and port number of the vehicle terminal according to its registration information, and sends the voice intercom request to the target vehicle terminal. After receiving the voice intercom request, the vehicle terminal starts the VoIP module and establishes an RTP voice transmission channel with the management center platform. During the voice transmission process, the vehicle terminal and the management center platform respectively use echo cancellation algorithms to process the collected voice signals and remove echo interference. The vehicle terminal and the management center platform also use noise suppression algorithms to reduce noise on the voice signals and improve the signal-to-noise ratio and clarity of the voice. When the vehicle terminal detects an emergency, such as a collision or sudden braking, it automatically triggers the voice intercom function and initiates a call request to the management center. After receiving the emergency call, the management center communicates with the driver in real time through the voice intercom channel to understand the situation and provide necessary remote assistance. At the same time, it coordinates rescue forces to rush to the scene for disposal.

[0052] This embodiment also provides a vehicle positioning management system, such as Figure 2 ,include:

[0053] The data acquisition module is used to collect vehicle driving status data in real time, and at the same time use the vehicle-mounted camera to collect images of the vehicle's surrounding environment, pre-process the images, and obtain multi-modal data;

[0054] An intelligent analysis and warning module is used to perform fusion analysis on the multimodal data, identify dangerous driving behaviors of the vehicle through the vehicle fatigue warning system DSM, and perform real-time positioning processing on the vehicle. If dangerous driving behaviors are identified, warning information is generated;

[0055] The vehicle management module is used to convert the warning information into voice based on the advanced driving assistance system ADAS, play the voice warning to the driver through the vehicle speaker, remind the driver to take timely measures, and display the warning information in text form on the vehicle display screen, and take corresponding measures to manage the vehicle through the management center platform.

[0056] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A vehicle positioning management method, characterized in that: include: Collect vehicle driving status data in real time, and use the on-board camera to collect images of the vehicle's surrounding environment, pre-process the images, and obtain multi-modal data; Perform fusion analysis on the multimodal data, identify dangerous driving behaviors of the vehicle through the vehicle fatigue warning system DSM and perform real-time positioning processing on the vehicle, and generate warning information if dangerous driving behaviors are identified; Based on the advanced driving assistance system ADAS, the warning information is converted into voice, and the voice warning is played to the driver through the vehicle speaker to remind the driver to take timely measures. At the same time, the warning information is displayed in text form on the vehicle display screen, and corresponding measures are taken through the management center platform to manage the vehicle.

2. The vehicle positioning management method according to claim 1, characterized in that: The vehicle driving status data is collected in real time, including: The vehicle speed, acceleration, steering angle and lane departure status data are obtained from the vehicle electronic control unit ECU in real time through the CAN bus interface of the vehicle terminal device, and the vehicle surrounding environment image is collected through the vehicle camera; The collected environmental images are preprocessed, including image denoising, enhancement and correction processing. The preprocessed image data is time-synchronized with the driving status data obtained from the CAN bus to construct multimodal data.

3. The vehicle positioning management method according to claim 2, characterized in that: The data is integrated and analyzed, and the dangerous driving behavior of the vehicle is identified through the vehicle fatigue warning system DSM, including: The vehicle anti-fatigue warning system DSM uses a convolutional neural network algorithm based on deep learning to perform feature extraction and fusion analysis on the multimodal data to obtain a comprehensive feature vector; Inputting the comprehensive feature vector into a pre-trained dangerous driving behavior recognition model for classification and judgment, and outputting the judgment result; According to the judgment result of the dangerous driving behavior identification model, it is determined whether the vehicle has dangerous driving behavior.

4. The vehicle positioning management method according to claim 3, characterized in that: Performing real-time positioning processing on the vehicle includes: The GPS positioning terminal can be used to check patrols in areas where dangerous driving behaviors occur in real time, and dispatch and command orders can be quickly issued through the management center platform.

5. The vehicle positioning management method according to claim 1, characterized in that: The warning information is converted into voice, and the voice warning is played to the driver through the vehicle speaker, including: For the identified dangerous driving behavior, the warning information text matching the type of dangerous driving behavior is obtained from the preset warning information library, and the urgency of the warning information is determined according to the degree of danger; The ADAS converts the acquired warning information text into a voice signal through speech synthesis technology, and adjusts the volume and speech speed of the voice signal according to the urgency of the warning information; The synthesized warning voice signal is transmitted to the vehicle-mounted speaker, and the warning voice is played through the vehicle-mounted speaker. At the same time, the warning information text is transmitted to the vehicle-mounted display screen and displayed.

6. The vehicle positioning management method according to claim 1, characterized in that: The method further includes: if the risk level of the identified dangerous driving behavior exceeds a preset safety threshold, triggering an active safety control strategy of the vehicle and taking corresponding safety measures according to the type of dangerous driving behavior; Continuously monitor the vehicle's driving status, obtain driving parameters in real time and input them into the random forest model to identify dangerous driving behaviors, and dynamically adjust the warning strategy based on the identification results.

7. The vehicle positioning management method according to claim 6, characterized in that: The active safety control is used to pre-set driving routes for vehicles with dangerous driving behaviors. When the task starts, the vehicle's route and status begin to be monitored and recorded. When the vehicle does not drive according to the preset route and drives out of the set area, the system will automatically alarm and notify the management center platform to take corresponding measures according to the actual situation.

8. The vehicle positioning management method according to claim 2, characterized in that: When dangerous driving behavior is identified, the vehicle driving status data, environmental image data and warning information collected by the vehicle terminal device are packaged in a predetermined format to generate a data packet; Encrypt the generated data packets, use an asymmetric encryption algorithm, and use the public key of the management center platform to encrypt the data packets; The encrypted data packet is uploaded to the management center platform through the wireless communication module of the vehicle-mounted terminal device. After receiving the uploaded data packet, the management center platform uses the private key to decrypt the data packet, restore the original vehicle driving status data, environmental image data and warning information, and store and analyze the decrypted data. Through big data analysis technology and machine learning algorithms, dangerous driving behaviors are counted and analyzed to provide decision support for relevant departments.

9. The vehicle positioning management method according to claim 8, characterized in that: A voice intercom channel is established between the vehicle-mounted terminal device and the management center platform through a wireless communication network, VoIP technology is used to achieve real-time voice communication, and voice quality is guaranteed by echo cancellation and noise suppression algorithms. When an emergency occurs, the driver can communicate and coordinate with the management center in a timely manner through the voice intercom function.

10. A vehicle positioning management system, characterized in that: include: The data acquisition module is used to collect vehicle driving status data in real time, and at the same time use the vehicle-mounted camera to collect images of the vehicle's surrounding environment, pre-process the images, and obtain multi-modal data; An intelligent analysis and warning module is used to perform fusion analysis on the multimodal data, identify dangerous driving behaviors of the vehicle through the vehicle fatigue warning system DSM, and perform real-time positioning processing on the vehicle. If dangerous driving behaviors are identified, warning information is generated; The vehicle management module is used to convert the warning information into voice based on the advanced driving assistance system ADAS, play the voice warning to the driver through the vehicle speaker, remind the driver to take timely measures, and display the warning information in text form on the vehicle display screen, and take corresponding measures to manage the vehicle through the management center platform.

Images