Vehicle-mounted driving safety early warning system based on artificial intelligence data acquisition

Abstract

The invention relates to the technical field of vehicle-mounted driving safety early warning, in particular to a vehicle-mounted driving safety early warning system based on artificial intelligence data acquisition, which comprises a vehicle-mounted driving safety early warning platform, a database, a vehicle-mounted working condition aging unit, a driving safety evaluation unit, a vehicle working condition analysis unit, a driving line pressing early warning unit and a vehicle-mounted management unit, preliminary analysis is carried out from the perspective of vehicle-mounted information effectiveness, so that the vehicle-mounted monitoring performance risk condition of the target vehicle can be visually known, analysis is carried out from the driving point and the vehicle condition point of the target vehicle on the basis of information feedback and a progressive mode, the driving risk and the vehicle condition risk of the target vehicle can be known, and the driving risk and the vehicle condition risk of the target vehicle can be known. And on the basis of vehicle-mounted monitoring and normal driving, line pressing prediction and early warning are carried out on driving of the target vehicle through the driving line pressing distance prediction model, so that traffic accidents such as scraping caused by line crossing or line pressing driving are avoided.

Description

Technical Field

[0001] The present invention relates to the field of vehicle-mounted driving safety warning technology, and in particular to a vehicle-mounted driving safety warning system based on artificial intelligence data collection. Background Art

[0002] With the rapid improvement of industrial technology and the gradual improvement of policies and regulations, intelligent driving as the future direction of automobile development has become a consensus of the industry chain and even the entire society; in related technologies, smart cars are mostly equipped with active warning functions to warn drivers in the event of risks; Real-time monitoring of the vehicle's operation can help the driver understand the vehicle's operation status in a timely manner and avoid unexpected situations. However, in the existing technology, it is impossible to analyze the effectiveness of the vehicle's own monitoring information, which increases the vehicle's false alarm rate and is not conducive to the stable driving of the vehicle. At the same time, it is impossible to monitor the driver and the vehicle condition in real time, and thus it is impossible to remind the driver to respond in time, which increases the driving risk of the vehicle. At the same time, it is impossible to predict the risk of the vehicle driving over the line, which increases the risk of traffic accidents such as scratches caused by crossing the line or driving over the line; In view of the above technical defects, a solution is now proposed. Summary of the invention

[0003] The purpose of the present invention is to provide an on-board driving safety warning system based on artificial intelligence data collection to solve the above-mentioned technical defects. The present invention conducts a preliminary analysis from the perspective of the effectiveness of on-board information, so as to intuitively understand the on-board monitoring performance risk of the target vehicle, which is helpful to reasonably manage to improve the on-board monitoring performance of the target vehicle. Based on information feedback and progressive analysis from the two points of target vehicle driving and vehicle condition, on the one hand, it is helpful to understand the driving risk and vehicle condition risk of the target vehicle, and on the other hand, it is helpful to timely remind the driver to respond, thereby improving the vehicle driving safety. Based on on-board monitoring and normal driving, the surround driving information is evaluated and analyzed by constructing a line-crossing model, so as to predict and warn the driving of the target vehicle through a driving line-crossing distance prediction model, so as to facilitate the driver to understand in real time whether the target vehicle crosses the line when turning, so as to provide early warning feedback to avoid traffic accidents such as scratches caused by crossing the line or driving on the line, thereby helping to improve the driving safety and warning timeliness of the target vehicle.

[0004] The purpose of the present invention can be achieved through the following technical solutions: an on-board driving safety warning system based on artificial intelligence data collection, including an on-board driving safety warning platform, a database, an on-board working condition time effectiveness unit, a driving safety assessment unit, a vehicle working condition analysis unit, a driving line crossing warning unit and an on-board management unit; The vehicle-mounted driving safety warning platform retrieves the vehicle-mounted working conditions information of the target vehicle from the database and sends it to the vehicle-mounted working condition aging unit; The vehicle-mounted working condition aging unit is used to conduct vehicle-mounted reliability monitoring, evaluation and analysis on the received vehicle-mounted working conditions information, and conduct discrimination processing on the obtained vehicle monitoring risk index to obtain a qualified signal, a first-level impact signal or a second-level impact signal; The driving safety evaluation unit is used to respond to the qualified signal, collect the driving information of the target vehicle, conduct driving safety monitoring feedback evaluation and analysis on the driving information, and send the obtained normal signal or alarm signal to the vehicle-mounted management unit; The vehicle condition analysis unit is used to respond to the qualified signal, collect the driving state information of the target vehicle, conduct vehicle state abnormal monitoring and analysis on the driving state information, conduct discrimination processing on the obtained driving monitoring evaluation value, and obtain a stable signal or an alarm signal; The driving line crossing warning unit is used to respond to the normal signal and the stable signal, collect multiple groups of historical panoramic driving information of the target vehicle, conduct line crossing model construction evaluation and analysis on the panoramic driving information, and send the obtained deviation instruction or line crossing warning signal to the vehicle-mounted management unit.

[0005] Preferably, the vehicle-mounted reliability monitoring, evaluation and analysis process of the vehicle-mounted working condition aging unit is as follows: Collect the driving period of the target vehicle and set it as the time threshold, obtain the vehicle-mounted working conditions information of the target vehicle within the time threshold. The vehicle-mounted working conditions information includes an information offset value and a transmission response value. Compare and analyze the information offset value and the transmission response value with the stored preset information offset value threshold and threshold transmission response value threshold, obtain the number corresponding to the information offset value and the transmission response value that is greater than or equal to the preset information offset value threshold and threshold transmission response value threshold, and set it as the vehicle monitoring risk index, and conduct discrimination processing on the vehicle monitoring risk index to generate a qualified signal, a first-level impact signal or a second-level impact signal.

[0006] Preferably, the information offset value represents the number of deviations between the information collection cycle and the set information collection cycle of the sensors in the target vehicle during the operation period. The sensors in the target vehicle include a driving speed sensor and an acceleration sensor; the transmission response value represents the number of corresponding sensors in the sensors in the target vehicle whose transmission duration exceeds the preset transmission duration threshold during the operation period. The transmission duration represents the duration between the start time of collection and the end time of transmission.

[0007] Preferably, the driving safety monitoring feedback evaluation and analysis process of the driving safety evaluation unit is as follows: Obtain the driving information of the target vehicle within the time threshold. The driving information includes the driving fatigue value and the driving risk value. Compare and analyze the driving fatigue value and the driving risk value with the preset driving fatigue value threshold and the preset driving risk value threshold stored in it to generate a normal signal or an alarm signal. The driving fatigue value represents the duration between the end time of the target vehicle's last rest closest to the current driving moment; the driving risk value represents the frequency at which the driving speed of the target vehicle exceeds the speed limit of the driving section.

[0008] Preferably, the vehicle condition analysis unit's vehicle status abnormal monitoring and analysis process is as follows: Divide the time threshold into i sub-time periods, where i is a natural number greater than zero. Obtain the driving status information of the target vehicle within each sub-time period. The driving status information includes the vehicle temperature difference index and the heat exchange processing value. Compare and analyze the vehicle temperature difference index and the heat exchange processing value with the preset vehicle temperature difference index threshold and the preset heat exchange processing value threshold stored in it. Set the number of sub-time periods corresponding to the vehicle temperature difference index being less than the preset vehicle temperature difference index threshold and the heat exchange processing value being less than the preset heat exchange processing value threshold as the driving monitoring evaluation value, and perform discriminant processing on the driving monitoring evaluation value to generate a stable signal or an alarm signal.

[0009] Preferably, the vehicle temperature difference index represents the frequency at which the operating temperature of the target vehicle's engine exceeds the preset operating temperature; the heat exchange processing value represents the duration corresponding to the ratio of the ventilation fan speed of the target vehicle to the energy consumption value being greater than the preset threshold.

[0010] Preferably, the process of constructing and evaluating the driving line crossing warning unit's line crossing model is as follows: Obtain multiple groups of historical panoramic driving information of the target vehicle within the time threshold. The panoramic driving information includes video images of the front, left, right, and rear fields of view of the target vehicle. Preprocess the panoramic driving information and construct a driving line crossing distance prediction model. The preprocessing includes data cleaning and filling missing values; Obtain the current panoramic driving information of the target vehicle within the time threshold, and perform image validity analysis on the current panoramic driving information. Perform discriminant analysis on the obtained image invalidation value to generate a deviation instruction or a prediction instruction; When a prediction instruction is generated, substitute the current panoramic driving information into the driving line crossing prediction model for prediction, obtain the driving line crossing distance value of the target vehicle, and perform discriminant processing on the driving line crossing distance value to obtain a line crossing warning signal.

[0011] Preferably, the process of performing image validity analysis on the current panoramic driving information is as follows: Divide the images in the current surround-view driving information into several sub-region blocks, where g is a natural number greater than zero. Obtain the image information of each sub-region block. The image information includes clarity and brightness. Obtain the number of sub-region blocks whose image information deviates from the preset range or is lower than the preset threshold, and set it as the regional difference value. Compare and analyze the regional difference value with the preset regional difference value threshold stored, and set the ratio of the number of sub-regions corresponding to the regional difference value greater than the preset regional difference value threshold to the total number of sub-regions as the image failure value.

[0012] The beneficial effects of the present invention are as follows: The present invention conducts a preliminary analysis from the perspective of the effectiveness of vehicle-mounted information to intuitively understand the vehicle-mounted monitoring performance risk situation of the target vehicle, which helps to reasonably manage to improve the vehicle-mounted monitoring performance of the target vehicle. And based on the information feedback and progressive method, it analyzes from two points of the target vehicle's driving and vehicle condition. On the one hand, it helps to understand the driving risk and vehicle condition risk of the target vehicle. On the other hand, it helps to timely remind the driver to respond, thereby improving the driving safety of the vehicle. Based on vehicle-mounted monitoring and normal driving, the present invention constructs and evaluates a lane-crossing model for surround-view driving information, so as to predict and warn of lane-crossing during the driving of the target vehicle through the driving lane-crossing distance prediction model. Furthermore, it is convenient for the driver to know in real time whether the target vehicle crosses the line when turning, so as to give an early warning feedback and avoid traffic accidents such as scratches caused by crossing or pressing the line, thereby helping to improve the driving safety and warning timeliness of the target vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] The present invention will be further described below with reference to the accompanying drawings; Figure 1 is the system flow block diagram of the present invention; Figure 2 is the local analysis reference diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0014] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0015] Embodiment 1: Please refer to Figures 1 to 2As shown in the figure, the present invention is an in-vehicle driving safety warning system based on artificial intelligence data collection, including an in-vehicle driving safety warning platform, a database, an in-vehicle working condition aging unit, a driving safety assessment unit, a vehicle working condition analysis unit, a driving line crossing warning unit, and an in-vehicle management unit. The database is in one-way communication connection with the in-vehicle driving safety warning platform. The in-vehicle driving safety warning platform is in one-way communication connection with the in-vehicle working condition aging unit. The in-vehicle working condition aging unit is in one-way communication connection with both the driving safety assessment unit and the vehicle working condition analysis unit. Both the driving safety assessment unit and the vehicle working condition analysis unit are in one-way communication connection with the driving line crossing warning unit and the in-vehicle management unit. The driving line crossing warning unit is in one-way communication connection with the in-vehicle management unit; The in-vehicle driving safety warning platform retrieves the in-vehicle working condition information of the target vehicle from the database and sends it to the in-vehicle working condition aging unit. The in-vehicle working condition aging unit is used to conduct in-vehicle reliability monitoring and evaluation analysis on the received in-vehicle working condition information, so as to intuitively understand the in-vehicle monitoring performance risk situation of the target vehicle, so as to manage it reasonably and improve the in-vehicle monitoring performance of the target vehicle. The specific in-vehicle reliability monitoring and evaluation analysis process is as follows: Collect the driving time period of the target vehicle and set it as the time threshold. Obtain the in-vehicle working condition information of the target vehicle within the time threshold. The in-vehicle working condition information includes the information offset value and the transmission response value. Compare and analyze the information offset value and the transmission response value with the stored preset information offset value threshold and the threshold transmission response value threshold. Obtain the number corresponding to the information offset value and the transmission response value that is greater than or equal to the preset information offset value threshold and the threshold transmission response value threshold, and set it as the vehicle monitoring risk index, and conduct discriminant processing on the vehicle monitoring risk index: If the vehicle monitoring risk index is equal to zero, a qualified signal is generated; If the vehicle monitoring risk index is equal to 1, a first-level impact signal is generated; If the vehicle monitoring risk index is equal to 2, a second-level impact signal is generated. Among them, the in-vehicle monitoring impact degrees corresponding to the first-level impact signal and the second-level impact signal increase in sequence. Send the qualified signal, the first-level impact signal, or the second-level impact signal to the in-vehicle management unit. After receiving the qualified signal, the first-level impact signal, or the second-level impact signal, the in-vehicle management unit immediately performs the preset warning operation corresponding to the qualified signal, the first-level impact signal, or the second-level impact signal, so as to intuitively understand the in-vehicle monitoring performance risk situation of the target vehicle, so as to manage it reasonably and improve the in-vehicle monitoring performance of the target vehicle; In an embodiment of the present invention, the information offset value represents the number of sensors in the target vehicle whose information collection cycle during the operation period deviates from the set information collection cycle. The sensors in the target vehicle include a driving speed sensor, an acceleration sensor, etc. It should be noted that the larger the value of the information offset value, the greater the risk of abnormal on-vehicle monitoring of the target vehicle; In an embodiment of the present invention, the transmission response value represents the number of corresponding sensors in the sensors in the target vehicle whose transmission duration during the operation period exceeds the preset transmission duration threshold. The transmission duration represents the duration between the start time of collection and the end time of transmission. It should be noted that the transmission response value is an influence parameter reflecting the on-vehicle monitoring delay state; When a qualified signal is generated, the driving safety evaluation unit is used to respond to the qualified signal, collect the driving information of the target vehicle, and perform a driving safety monitoring feedback evaluation analysis on the driving information to determine whether the driving risk of the target vehicle is too high, so as to give a warning reminder in time, which helps to enhance the driving safety awareness and driving safety of the driver, and avoid fatigue driving and speeding. The specific driving safety monitoring feedback evaluation analysis process is as follows: Obtain the driving information of the target vehicle within the time threshold. The driving information includes a driving fatigue value and a driving risk value, and compare and analyze the driving fatigue value and the driving risk value with the preset driving fatigue value threshold and the preset driving risk value threshold stored in it: If the driving fatigue value is less than the preset driving fatigue value threshold and the driving risk value is less than the preset driving risk value threshold, a normal signal is generated; If the driving fatigue value is greater than or equal to the preset driving fatigue value threshold, or the driving risk value is greater than or equal to the preset driving risk value threshold, an alarm signal is generated, and the normal signal or the alarm signal is sent to the vehicle-mounted management unit. After receiving the normal signal or the alarm signal, the vehicle-mounted management unit immediately performs the preset warning operation corresponding to the normal signal or the alarm signal to timely remind the driver to drive safely; In an embodiment of the present invention, the driving fatigue value represents the duration between the end time of the target vehicle's last rest closest to the current driving moment. It should be noted that the larger the value of the driving fatigue value, the greater the driving risk; In an embodiment of the present invention, the driving risk value represents the frequency at which the driving speed of the target vehicle exceeds the speed limit of the driving section. It should be noted that the larger the value of the driving risk value, the greater the driving risk.

[0016] Embodiment 2: When a qualified signal is generated, the vehicle condition analysis unit is used to respond to the qualified signal, collect the driving state information of the target vehicle, and perform vehicle state abnormal monitoring and analysis on the driving state information to determine whether the vehicle condition is normal during the driving process of the target vehicle, which helps to improve the timeliness and monitoring efficiency of vehicle condition monitoring and early warning during the operation of the target vehicle, so as to timely remind the driver to manage the vehicle. The specific vehicle state abnormal monitoring and analysis process is as follows: Divide the time threshold into i sub-time periods, where i is a natural number greater than zero. Obtain the driving state information of the target vehicle in each sub-time period. The driving state information includes the vehicle temperature difference index and the heat exchange processing value. Compare and analyze the vehicle temperature difference index and the heat exchange processing value with the preset vehicle temperature difference index threshold and the preset heat exchange processing value threshold stored in it. Set the number of sub-time periods corresponding to the vehicle temperature difference index being less than the preset vehicle temperature difference index threshold and the heat exchange processing value being less than the preset heat exchange processing value threshold as the driving monitoring evaluation value, and perform discrimination processing on the driving monitoring evaluation value: If the driving monitoring evaluation value is less than or equal to the preset driving monitoring evaluation value threshold, generate a stable signal; If the driving monitoring evaluation value is greater than the preset driving monitoring evaluation value threshold, generate an alarm signal, and send the stable signal or the alarm signal to the in-vehicle management unit. After receiving the stable signal or the alarm signal, the in-vehicle management unit immediately performs the preset warning operation corresponding to the stable signal or the alarm signal, so as to timely remind the driver to manage the vehicle and improve the driving safety of the vehicle; In the embodiment of the present invention, the vehicle temperature difference index represents the frequency at which the operating temperature of the engine of the target vehicle exceeds the preset operating temperature. It should be noted that the vehicle temperature difference index is an influence parameter reflecting the state of the target vehicle itself; In the embodiment of the present invention, the heat exchange processing value represents the duration corresponding to the ratio of the ventilation fan speed to the energy consumption value of the target vehicle being greater than the preset threshold. It should be noted that the heat exchange processing value is an influence parameter reflecting the heat dissipation state of the target vehicle itself; Based on the normal signal and the stable signal, the driving lane crossing warning unit is used to respond to the normal signal and the stable signal, collect multiple groups of historical panoramic driving information of the target vehicle, and perform lane crossing model construction and evaluation analysis on the panoramic driving information, so as to perform lane crossing prediction and warning on the driving of the target vehicle through the driving lane crossing distance prediction model, and then facilitate the driver to know in real time whether the target vehicle crosses the line when turning, so as to perform early warning feedback and avoid traffic accidents such as scratches caused by crossing or crossing the line, which helps to improve the driving safety and warning timeliness of the target vehicle. The specific lane crossing model construction and evaluation analysis process is as follows: Obtain multiple groups of historical surround driving information of the target vehicle within the time threshold. The surround driving information includes video images of the front, left, right, and rear fields of view of the target vehicle. Preprocess the surround driving information and construct a driving line-crossing distance prediction model. The preprocessing includes data cleaning, filling missing values, etc.; In an embodiment of the present invention, obtain the current surround driving information of the target vehicle within the time threshold, and perform image validity analysis on the current surround driving information. Perform discriminant analysis on the obtained image failure values: If there is an image failure value greater than the preset perspective image threshold, generate a deviation instruction; If there is no image failure value greater than the preset perspective image threshold, generate a prediction instruction. Send the deviation instruction to the vehicle-mounted management unit. After receiving the deviation instruction, the vehicle-mounted management unit immediately performs the preset warning operation corresponding to the deviation instruction, thereby ensuring the validity of the collected images and reducing the impact of the images on the line-crossing prediction; In an embodiment of the present invention, the process of performing image validity analysis on the current surround driving information is as follows: Divide the images in the current surround driving information into g sub-region blocks, where g is a natural number greater than zero. Obtain the image information of each sub-region block. The image information includes clarity, brightness, etc. Obtain the number corresponding to the image information of each sub-region block deviating from the preset range or being lower than the preset threshold, and set it as the regional difference value. Compare and analyze the regional difference value with the stored preset regional difference value threshold. Set the ratio of the number of sub-regions corresponding to the regional difference value greater than the preset regional difference value threshold to the total number of sub-regions as the image failure value; When generating a prediction instruction, substitute the current surround driving information into the driving line-crossing prediction model for prediction, obtain the driving line-crossing distance value of the target vehicle, and perform discriminant processing on the driving line-crossing distance value: If the driving line-crossing distance value is greater than the preset driving line-crossing distance value threshold, do not generate any signal; If the driving line-crossing distance value is less than or equal to the preset driving line-crossing distance value threshold, generate a line-crossing warning signal. Send the line-crossing warning signal to the vehicle-mounted management unit. After receiving the line-crossing warning signal, the vehicle-mounted management unit immediately performs the preset warning operation corresponding to the line-crossing warning signal, facilitating the driver to understand in real time whether the target vehicle is crossing the line when turning, so as to perform early warning feedback and avoid traffic accidents such as scratches caused by crossing or straddling the line, thereby contributing to improving the driving safety and warning timeliness of the target vehicle; In summary, the present invention conducts a preliminary analysis from the perspective of the effectiveness of in-vehicle information, so as to intuitively understand the risk situation of the in-vehicle monitoring performance of the target vehicle, which helps to reasonably manage to improve the in-vehicle monitoring performance of the target vehicle. By analyzing from two aspects of the target vehicle's driving and vehicle condition based on information feedback and a progressive manner, on the one hand, it helps to understand the driving risk and vehicle condition risk of the target vehicle, and on the other hand, it helps to timely remind the driver to take countermeasures, thereby improving the driving safety of the vehicle. That is, it conducts a driving safety monitoring feedback evaluation analysis on the driving information to determine whether the driving risk of the target vehicle is too high, so as to timely give a warning reminder, which helps to enhance the driver's driving safety awareness and improve driving safety, and avoid fatigue driving and speeding. It conducts a vehicle condition abnormal monitoring analysis on the driving state information to determine whether the vehicle condition of the target vehicle is normal during driving, which helps to improve the timeliness and monitoring efficiency of vehicle condition monitoring and warning during the operation of the target vehicle, so as to timely remind the driver to manage the vehicle. Based on in-vehicle monitoring and normal driving, it conducts a construction evaluation analysis of a line-crossing model for panoramic driving information, so as to predict and give a warning of line-crossing for the driving of the target vehicle through a driving line-crossing distance prediction model, and then facilitate the driver to know in real time whether the target vehicle crosses the line when turning, so as to give an early warning feedback and avoid traffic accidents such as scratches caused by crossing or crossing the line, thereby helping to improve the driving safety and warning timeliness of the target vehicle.

[0017] The setting of the threshold value is for the convenience of comparison. Regarding the size of the threshold value, it depends on the amount of sample data and the number of base values set by those skilled in the art for each group of sample data; as long as it does not affect the proportional relationship between the parameters and the quantified values.

[0018] The above formulas are all obtained by collecting a large amount of data for software simulation and selecting a formula close to the true value. The coefficients in the formulas are set by those skilled in the art according to the actual situation. As described above, it is only a preferred specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. A vehicle-mounted driving safety warning system based on artificial intelligence data collection, characterized in that: It includes an on-board driving safety warning platform, a database, an on-board operating condition time effectiveness unit, a driving safety assessment unit, a vehicle operating condition analysis unit, a driving line crossing warning unit, and an on-board management unit; The vehicle-mounted driving safety warning platform retrieves the vehicle-mounted working condition information of the target vehicle from the database and sends it to the vehicle-mounted working condition time-effect unit; The vehicle operating condition time-effectiveness unit is used to perform vehicle reliability monitoring, evaluation and analysis on the received vehicle operating condition information, and to perform discrimination processing on the obtained vehicle monitoring risk index to obtain a qualified signal or a first-level impact signal or a second-level impact signal; The driving safety assessment unit is used to respond to the qualified signal, collect the driving information of the target vehicle, conduct driving safety monitoring feedback assessment and analysis on the driving information, and send the obtained normal signal or alarm signal to the vehicle management unit; The vehicle operating condition analysis unit is used to respond to the qualified signal, collect the driving state information of the target vehicle, perform vehicle state abnormality monitoring and analysis on the driving state information, perform judgment processing on the obtained driving monitoring evaluation value, and obtain a stable signal or an alarm signal; The driving line crossing warning unit is used to respond to normal signals and stable signals, collect multiple groups of historical surround driving information of the target vehicle, and build an evaluation and analysis of the surround driving information to cross the line model, and send the obtained deviation instructions or cross the line warning signals to the on-board management unit.

2. The vehicle-mounted driving safety warning system based on artificial intelligence data collection according to claim 1 is characterized in that: The on-board reliability monitoring, evaluation and analysis process of the on-board operating condition aging unit is as follows: The driving time period of the target vehicle is collected and set as the time threshold, and the on-board operating condition information of the target vehicle within the time threshold is obtained. The on-board operating condition information includes an information offset value and a transmission response value. The information offset value and the transmission response value are compared and analyzed with the stored preset information offset value threshold and the threshold transmission response value threshold, and the number of information offset values ​​and transmission response values ​​that are greater than or equal to the preset information offset value threshold and the threshold transmission response value threshold is obtained, and set as the vehicle monitoring risk index, and the vehicle monitoring risk index is discriminated and processed to generate a qualified signal or a first-level impact signal or a second-level impact signal.

3. The vehicle-mounted driving safety warning system based on artificial intelligence data collection according to claim 2 is characterized in that: The information offset value indicates the number of deviations between the information collection cycle during the operation period and the set information collection cycle in the sensors in the target vehicle, and the sensors in the target vehicle include a driving speed sensor and an acceleration sensor; The transmission response value indicates the number of corresponding sensors in the target vehicle whose transmission duration exceeds a preset transmission duration threshold during the operation period, and the transmission duration indicates the duration between the start time of collection and the end time of transmission.

4. The vehicle-mounted driving safety warning system based on artificial intelligence data collection according to claim 1 is characterized in that: The driving safety monitoring feedback evaluation and analysis process of the driving safety evaluation unit is as follows: Acquire driving information of the target vehicle within the time threshold, the driving information including driving fatigue value and driving risk value, compare and analyze the driving fatigue value and driving risk value with the preset driving fatigue value threshold and preset driving risk value threshold recorded and stored internally, and generate a normal signal or generate an alarm signal; The driving fatigue value represents the time between the target vehicle's last rest end time and the current driving time; The driving risk value indicates the frequency at which the driving speed of the target vehicle exceeds the speed limit of the driving section.

5. The vehicle-mounted driving safety warning system based on artificial intelligence data collection according to claim 1 is characterized in that: The vehicle state abnormality monitoring and analysis process of the vehicle operating condition analysis unit is as follows: The time threshold is divided into i sub-time periods, where i is a natural number greater than zero, and the driving status information of the target vehicle in each sub-time period is obtained, the driving status information including the vehicle temperature difference index and the heat exchange treatment value, the vehicle temperature difference index and the heat exchange treatment value are compared and analyzed with the preset vehicle temperature difference index threshold and the preset heat exchange treatment value threshold entered and stored internally, the number of sub-time periods corresponding to the vehicle temperature difference index being less than the preset vehicle temperature difference index threshold and the heat exchange treatment value being less than the preset heat exchange treatment value threshold is set as the driving monitoring evaluation value, and the driving monitoring evaluation value is discriminated and processed to generate a stable signal or an alarm signal.

6. The vehicle-mounted driving safety warning system based on artificial intelligence data collection according to claim 5 is characterized in that: The vehicle temperature difference index indicates the frequency at which the operating temperature of the target vehicle engine exceeds the preset operating temperature; the heat exchange processing value indicates the duration corresponding to the ratio between the ventilation fan speed and the energy consumption value of the target vehicle being greater than the preset threshold.

7. The vehicle-mounted driving safety warning system based on artificial intelligence data collection according to claim 1 is characterized in that: The construction, evaluation and analysis process of the line crossing model of the line crossing warning unit is as follows: Obtain multiple sets of historical surround driving information of the target vehicle within the time threshold. The surround driving information includes video images of the front, left, right and rear fields of view of the target vehicle. Preprocess the surround driving information and build a driving line distance prediction model. The preprocessing includes data cleaning and filling missing values. The current surround driving information of the target vehicle within the time threshold is obtained, and image validity analysis is performed on the current surround driving information, and a discriminant analysis is performed on the obtained image failure value to generate a deviation instruction or a prediction instruction; When a prediction instruction is generated, the current surround driving information is substituted into the driving line crossing prediction model for prediction, the driving line crossing distance value of the target vehicle is obtained, and the driving line crossing distance value is judged and processed to obtain a line crossing warning signal.

8. The vehicle-mounted driving safety warning system based on artificial intelligence data collection according to claim 7 is characterized in that: The image validity analysis process of the current surround driving information is as follows: The image in the current surround driving information is divided into g sub-region blocks, where g is a natural number greater than zero, and image information of each sub-region block is obtained, the image information including clarity and brightness, and the number of sub-region blocks whose image information deviates from a preset range or is lower than a preset threshold is obtained, and is set as a regional difference value. The regional difference value is compared and analyzed with a stored preset regional difference value threshold, and the ratio of the number of sub-regions corresponding to the regional difference value greater than the preset regional difference value threshold to the total number of sub-regions is set as an image failure value.

Images