AI auxiliary teaching driving practice safety management method and system

Abstract

The embodiment of the invention provides an AI auxiliary teaching driving practice safety management method and system, and aims to solve the problem that the driving teaching using an intelligent coach module still has safety risks due to the large driving level difference of driving students. The method comprises the following steps: acquiring image data of a current driving student, and determining sight line information of the driving student based on the image data; under the condition that the sight line information indicates that the driving student watches a display terminal of an AI driving coach, acquiring environment image data of the current vehicle; and displaying the environment image data on the display terminal together.

Description

Technical Field

[0001] This application relates to the field of computer technology, and particularly to an AI-assisted teaching driving practice safety management method and system. Background Art

[0002] With the rapid development of artificial intelligence, autonomous driving, and intelligent perception technologies, traditional driving training modes have gradually exposed problems such as high teaching costs, shortage of coach resources, uneven teaching quality, and difficulty in quantifying training effects. Therefore, the "intelligent coach module" leveraging AI technology has gradually become one of the mainstream solutions to improve the efficiency and quality of driving training.

[0003] The intelligent coach module is usually installed in training vehicles. By integrating multiple sensors (such as cameras, lidar, millimeter-wave radars, GPS positioning, inertial navigation, etc.), it can real-time perceive the driving behavior of trainees and the vehicle state, and automatically evaluate the driving habits, safety awareness, and skill levels of drivers based on big data analysis and deep learning algorithms.

[0004] However, due to the large differences in the driving levels of driving trainees, there are still safety risks when using the intelligent coach module for driving teaching. Summary of the Invention

[0005] The embodiments of this application provide an AI-assisted teaching driving practice safety management method and system, which can solve the problem that there are still safety risks when using the intelligent coach module for driving teaching due to the large differences in the driving levels of driving trainees.

[0006] The first aspect of the embodiments of this application provides an AI-assisted teaching driving practice safety management method, including:

[0007] Collect image data of the current driving trainee to determine the line-of-sight information of the driving trainee based on the image data;

[0008] When the line-of-sight information indicates that the driving trainee is looking at the display terminal of the AI driving coach, obtain the environmental image data of the current vehicle;

[0009] Display the environmental image data on the display terminal together.

[0010] Optionally, it further includes:

[0011] When the line-of-sight information indicates that the driving trainee is looking at the display terminal of the AI driving coach, wake up the assisted driving function to perform assisted driving custody on the vehicle.

[0012] Optionally, it further includes:

[0013] Collecting driving status data of the vehicle at the start time of assisted driving trusteeship takeover, and driving status data and position relationship of surrounding vehicles;

[0014] When it is determined based on the driving state data of the vehicle at the time of takeover of assisted driving, the driving state data of surrounding vehicles, and their positional relationships that a collision would have occurred if assisted driving had not been taken over, simulating a theoretical collision scenario between the vehicle and surrounding vehicles based on the driving state data of the vehicle at the time of takeover, the driving state data of surrounding vehicles, and their positional relationships;

[0015] The theoretical collision picture is displayed on the display terminal.

[0016] Optionally, also include:

[0017] determining the type of the surrounding vehicles based on the environmental image data;

[0018] Determining theoretical structural parameters of the surrounding environment vehicles according to the types of the surrounding environment vehicles;

[0019] When it can be determined based on the driving state data of the vehicle at the time of takeover of assisted driving, the driving state data of surrounding vehicles, and their positional relationship that a collision would have occurred if assisted driving trusteeship had not been performed, a theoretical collision scene between the vehicle and surrounding vehicles is simulated based on the driving state data of the vehicle at the time of takeover, the driving state data of surrounding vehicles, their positional relationship, and theoretical structural parameters of the surrounding vehicles, the theoretical collision scene including a theoretical degree of damage after the collision between the vehicle and surrounding vehicles;

[0020] The theoretical collision picture is displayed on the display terminal.

[0021] Optionally, also include:

[0022] When the line of sight information indicates that the driving student has returned to the normal driving field of view from the display terminal where the driver is watching the AI driving instructor, the assisted driving function is exited to release the vehicle from the assisted driving trusteeship.

[0023] Optionally, also include:

[0024] When the line of sight information indicates that the driver trainee has returned to a normal driving field of view from the display terminal viewing the AI driving instructor, obtaining the driver trainee's driving control information;

[0025] When the driving control information matches the current driving environment, the assisted driving function is exited to remove the vehicle from the assisted driving trusteeship.

[0026] Optionally, also include:

[0027] When the vehicle is under assisted driving custody, an attention warning message is generated.

[0028] In a second aspect of the embodiments of the present application, an AI-assisted teaching driving practice safety management device is provided, including:

[0029] An acquisition unit, configured to acquire image data of the current driving trainee to determine the line-of-sight information of the driving trainee based on the image data;

[0030] An acquisition unit, configured to acquire the environmental image data of the current vehicle when the line-of-sight information indicates that the driving trainee is watching the display terminal of the AI driving instructor;

[0031] A display unit, configured to display the environmental image data on the display terminal together.

[0032] In a third aspect of the embodiments of the present application, an electronic system is provided, including a memory and a processor. When the processor executes a computer program stored in the memory, the steps of the above-mentioned AI-assisted teaching driving practice safety management method are implemented.

[0033] In a fourth aspect of the embodiments of the present application, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned AI-assisted teaching driving practice safety management method are implemented.

[0034] In summary, the AI-assisted teaching driving practice safety management method provided by the embodiments of the present application acquires image data of the current driving trainee to determine the line-of-sight information of the driving trainee based on the image data; when the line-of-sight information indicates that the driving trainee is watching the display terminal of the AI driving instructor, acquires the environmental image data of the current vehicle; and displays the environmental image data on the display terminal together. Thus, the fusion of real-time environmental images can prevent trainees from ignoring the external situation due to excessive focus on the screen, effectively reducing the probability of accidents caused by distracted attention. Even if the trainees focus on the screen information, they can intuitively and real-time see the external environmental status, realizing the synchronization of driving prompts and environmental perception, and effectively improving the driver's intuitive understanding and judgment ability of the surrounding traffic environment. This method relies on real-time face recognition and line-of-sight detection technologies to achieve the synchronous update of trainee behavior monitoring and environmental image fusion. The interaction process is natural and smooth, enhancing the trainee's training experience. Trainees can see AI prompts and real-time external conditions on the screen simultaneously without frequently shifting their line of sight back and forth, significantly reducing the cognitive burden during driving and reducing tension and anxiety.

[0035] Correspondingly, the AI-assisted teaching driving practice safety management device, electronic system, and computer-readable storage medium provided by the embodiments of the present invention also have the above technical effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a schematic flowchart of a possible AI-assisted teaching driving practice safety management method provided by an embodiment of the present application;

[0037] Figure 2 It is a schematic structural block diagram of a possible AI-assisted teaching driving practice safety management device provided by an embodiment of the present application;

[0038] Figure 3 It is a schematic hardware structure diagram of a possible AI-assisted teaching driving practice safety management device provided by an embodiment of the present application;

[0039] Figure 4 It is a schematic structural block diagram of a possible electronic system provided by an embodiment of the present application;

[0040] Figure 5 It is a schematic structural block diagram of a possible computer-readable storage medium provided by an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0041] The embodiments of the present application provide an AI-assisted teaching driving practice safety management method and system, which can solve the problem that there are still safety risks in driving teaching using an intelligent coach module due to the large differences in the driving levels of driving students.

[0042] The terms "first", "second", "third", "fourth", etc. (if any) in the specification, claims and above-mentioned drawings of the present application are used to distinguish similar objects, and do not necessarily need to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments described here can be implemented in an order different from that shown or described here. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product or device comprising a series of steps or units does not necessarily need to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products or devices. The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments.

[0043] Please refer to Figure 1 , which is a flowchart of an AI-assisted teaching driving practice safety management method provided by an embodiment of the present application, and specifically may include: S110-S130.

[0044] S110. Collect the image data of the current driving trainee to determine the line-of-sight information of the driving trainee based on the image data.

[0045] S120. When the line-of-sight information indicates that the driving trainee is watching the display terminal of the AI driving instructor, obtain the environmental image data of the current vehicle.

[0046] S130. Display the environmental image data on the display terminal together.

[0047] It can be understood that by real-time monitoring the facial image of the driving trainee, inferring the position of the trainee's line-of-sight landing point, if it is judged that the line of sight turns to the screen to watch the AI prompt, the real-time video image of the vehicle external environment is quickly launched and integrated into the AI instructor display screen, so that the trainee can watch the AI prompt and at the same time not lose the perception of the surrounding environment, avoiding accident risks.

[0048] Exemplarily, inside the teaching vehicle, a high-resolution high-definition camera (recommended resolution of 1920×1080 pixels or above) is installed. The camera should be fixed directly in front of the driver's console, with a field of view that can completely cover the trainee's face to ensure unobstructed acquisition of facial information. After the vehicle starts, the camera captures the driver's dynamic facial images in real time and transmits the image data to the AI computing terminal at a frequency of no less than 30 frames per second to ensure the continuity and stability of the driver's line of sight detection. For example, an infrared camera is installed directly in front of the driver's seat, which can effectively capture clear facial images even when the lighting conditions are insufficient. Subsequently, the collected image data is transmitted to the driving behavior monitoring module, and the images are processed and analyzed in real time through deep learning algorithms to achieve driver line of sight detection. This can be achieved through the following algorithm: First, accurately locate the driver's facial area in the image through a face detection model (such as YOLOv8 or SSD). Then, based on the face key point detection technology, extract facial features, locate the positions of the eyes, pupils, and head pose data. Next, use the line of sight tracking algorithm to calculate the combination of the eyeball line of sight direction and the head pose to determine whether the driver's line of sight falls within the area of the in-vehicle AI coach display terminal. When the angle between the driver's line of sight direction and the display terminal is less than 10°, it is determined that the trainee is watching the display screen. After determining that the driver is watching the AI coach screen, the vehicle intelligent management system triggers the external environment image acquisition process. Multiple wide-angle high-definition cameras are installed at key positions outside the vehicle (such as directly in front of the vehicle head, the rear of the vehicle, and the areas of the left and right side rearview mirrors) to provide a comprehensive coverage of the 360° environment around the vehicle. Camera configuration examples: The front camera is installed in the middle upper area of the vehicle's front windshield; a side high-definition camera is installed under each of the left and right rearview mirrors; a reverse camera is installed at the rear. The above cameras capture the environmental image data around the vehicle in real time. After being transmitted to the in-vehicle intelligent terminal, real-time image stitching and distortion correction processing are performed to construct a blind spot-free environmental perspective image stream. To ensure clear and smooth images, the camera capture frame rate can be set above 30 frames per second, and the image quality is 720P or higher to provide accurate and intuitive environmental information. The last step is to fuse and display the collected real-time environmental images with the driving prompt information of the AI coach: The central area of the main display screen is the display area of the AI coach module, which is used to provide real-time text or graphic driving prompts, such as "Turn left ahead, slow down", "Your current speed is too fast, please slow down", etc.; when it is detected that the driver's line of sight moves to this screen area, a smaller external real-time image window automatically pops up in a corner of the screen (such as the lower right corner), presenting the image of the road ahead, the blind spot, or the environment of the side and rear of the vehicle in real time; if the line of sight moves back to the road, the environmental image is automatically hidden to prevent too much information on the screen from interfering with driving. When the driver is training to turn on an urban road and watches the AI coach prompt "Please slow down at the intersection ahead", if the line of sight stays for too long, the image of the intersection ahead is automatically displayed in a corner of the screen in real time to remind the trainee to pay attention to pedestrians and non-motor vehicles that may cross the intersection, preventing collision accidents caused by ignoring the traffic conditions.When a driver is undergoing reverse training and pays too much attention to the screen prompts while ignoring the external rear environment, a small window at the bottom of the screen automatically displays the real-time image captured by the reverse camera, intuitively prompting the reverse environment and preventing collisions with obstacles.

[0049] In summary, the AI-assisted teaching driving practice safety management method provided by the above embodiments collects the image data of the current driving student, and determines the line-of-sight information of the driving student based on the image data; when the line-of-sight information indicates that the driving student is watching the display terminal of the AI driving instructor, obtains the environmental image data of the current vehicle; and displays the environmental image data on the display terminal together. Thus, the fusion of real-time environmental images can prevent students from ignoring the external situation due to excessive attention to the screen, effectively reducing the probability of accidents caused by distracted attention. Even if the student pays attention to the screen information, they can intuitively and real-time see the external environmental state, realizing the synchronization of driving prompts and environmental perception, and effectively improving the driver's intuitive understanding and judgment ability of the surrounding traffic environment. This method relies on real-time face recognition and line-of-sight detection technologies to achieve the synchronous update of student behavior monitoring and environmental image fusion. The interaction process is natural and smooth, enhancing the student's training experience. The student can see the AI prompts and the real-time external situation on the screen at the same time without frequently shifting the line of sight back and forth, significantly reducing the cognitive burden during driving and reducing tension and anxiety.

[0050] In one embodiment, it further includes:

[0051] When the line-of-sight information indicates that the driving student is watching the display terminal of the AI driving instructor, wakes up the assisted driving function to perform assisted driving custody on the vehicle.

[0052] It is understandable that it may also include automatically triggering the assisted driving function of the vehicle when it is detected that the sight of the driving trainee stays on the AI driving instructor display terminal, so as to temporarily take over the vehicle control right and enhance driving safety. When the in-vehicle intelligent instructor module determines that the driver is focusing on watching the in-vehicle AI instructor display terminal by real-time monitoring of the driver's sight direction (such as the attention is distracted beyond the set threshold, such as not looking at the road surface continuously for more than 1 second), the system will automatically trigger the configured assisted driving function of the vehicle and implement a short-term assisted driving trusteeship. When the sight monitoring algorithm detects that the driver's sight stays continuously in the display screen area, the system immediately sends an instruction to wake up the existing assisted driving (ADAS or L2-level assisted driving) function of the vehicle. First, the system sends a wake-up signal to the vehicle's driving control unit (such as EPS electric power steering control unit, ESP electronic stability system, ACC adaptive cruise control module) through the CAN bus or Ethernet communication interface. After receiving the wake-up instruction, the assisted driving system completes the quick start procedure by itself and quickly enters the standby takeover mode. For example, when the trainee is driving on a road at a speed of 60 km / h and the attention stays on the AI instructor screen for a long time, the assisted driving system immediately wakes up the ACC adaptive cruise function and activates the lane keeping assist function, ready to take over the throttle, brake and steering wheel at any time. After the assisted driving function is woken up, the real-time environment perception sensors (such as lidar, millimeter wave radar, camera) will automatically start to scan the environment and build a real-time dynamic environment model. When the system determines that the trainee continues to focus on the AI instructor screen and pays insufficient attention to the road environment (the sight deviation duration exceeds the set safety threshold, such as 1 second), the assisted driving system will actively execute the vehicle temporary trusteeship: speed control, start the automatic speed limit or automatic deceleration measures to ensure that the vehicle decelerates smoothly or maintains a safe speed; lane keeping, identify the lane lines through the front camera or millimeter wave radar, and the vehicle actively adjusts the steering wheel angle to keep the vehicle driving in the center of the lane; collision warning and emergency braking: real-time monitor the distance of obstacles in front of and behind the vehicle, and if a potential collision risk is found, actively control the vehicle braking. For example, during the period when the driver observes the prompt content on the display screen, the vehicle in front suddenly decelerates, and the assisted driving system immediately controls the brake to decelerate and corrects the driving trajectory through the active steering system to avoid a rear-end collision. When the trainee is driving on a curve and ignores the curve ahead due to focusing on the screen, the assisted driving system quickly activates the lane keeping and steering assist functions to guide the vehicle to pass through the curve smoothly. When the sight monitoring detects that the driver's sight returns to the road again, indicating that the driver regains the road condition perception, the assisted driving system automatically gradually releases the control of the vehicle, emits a prompt sound or a screen prompt to clearly inform the driver that the control right has been returned. The driver resumes control of the vehicle, and the assisted driving function returns to the standby state for activation again when the next safety risk occurs. For example, after the driver's sight returns to the road, the system automatically pops up a prompt in the AI display terminal: "The assisted driving trusteeship has been exited, please continue driving", prompting the driver to resume control of the vehicle.Therefore, when the driver's line of sight is diverted for a long time, the assisted driving system can actively intervene in vehicle control, effectively avoiding the risk of safety accidents caused by distracted attention, especially suitable for novice learners or learners with relatively weak driving skills. Through the linkage of AI assistance and the assisted driving system, accidents caused by human negligence can be prevented, ensuring a safe practice environment for learners. During the training of learners, they understand that the vehicle's intelligent assisted driving is always on standby and can automatically provide safety protection in case of emergencies, reducing the training tension at the psychological level and improving the driving practice efficiency.

[0053] In one embodiment, it further includes:

[0054] Collect the driving state data of the vehicle, the driving state data of surrounding environmental vehicles, and the positional relationship at the starting moment of assisted driving takeover.

[0055] When it can be determined based on the driving state data of the vehicle, the driving state data of surrounding environmental vehicles, and the positional relationship at the assisted driving takeover moment that there is a possibility of collision if assisted driving takeover is not performed, simulate the theoretical collision scenario of the vehicle and surrounding environmental vehicles according to the driving state data of the vehicle, the driving state data of surrounding environmental vehicles, and the positional relationship at the takeover moment.

[0056] Display the theoretical collision scenario on the display terminal.

[0057] It can be understood that not only can the driver's attention be monitored in real time through driver line of sight detection and the environmental image be displayed in a timely manner, but also the assisted driving system can be actively triggered through intelligent judgment to take over vehicle control, further improving training safety. In addition, this embodiment particularly adds a vehicle collision risk prediction and simulation function, that is: when the line of sight monitoring system determines that the driver is watching the AI terminal and the attention is diverted, the vehicle assisted driving takeover is automatically started; at the moment of takeover, the AI system automatically collects the real-time state information of the vehicle's own motion state (speed, direction, acceleration) and other vehicles in the environment (position, speed, driving direction); the system analyzes and simulates in real time the possible collision situation that may occur if the vehicle is not taken over in time, and shows the consequences of the collision risk through simulation derivation; if it is determined that a collision accident may occur, the collision risk is dynamically and intuitively displayed on the display terminal through simulation to strengthen the safety awareness and driving sense of responsibility of learners.

[0058] Exemplarily, a high-definition camera facing the driver inside the vehicle cockpit captures the facial image data of the trainee in real time and transmits it to the in-vehicle AI computing terminal. A convolutional neural network (CNN) model can be used to implement driver face detection and gaze tracking; calculate the driver's gaze direction angle to determine whether the in-vehicle AI coach display screen is watched for a long time; if the line of sight deviates from the road by more than a preset threshold (for example, the line of sight deviation lasts for more than 2 seconds), the vehicle's Advanced Driver Assistance System (ADAS) is immediately automatically awakened to briefly take over the control of the vehicle's steering wheel, brakes or accelerator to avoid potential risks. When the ADAS function is activated and takes over the vehicle, the following key data is synchronously recorded in real time: the status information of the vehicle itself; speed (read through the vehicle CAN bus); driving trajectory information (GPS / INS data); real-time dynamic information such as acceleration, braking status, and steering angle. And the status information of surrounding environment vehicles (detected by radar or vision sensors): the relative position, speed, and movement direction of surrounding vehicles; road traffic conditions, including the predicted trajectories of other vehicles, possible vehicle collision positions or distances. For example, if the trainee is moving forward at a speed of 50 km / h and the attention is transferred to the AI coach prompt, and the distance to the vehicle in front is only 20 meters and it is braking and decelerating at this moment, the radar senses this information in real time and transmits the data to the intelligent control system immediately. While taking over the vehicle, the system quickly completes the simulation and determination of the collision risk. The vehicle and environmental vehicle status data collected in real time can be used to quickly conduct virtual simulations based on vehicle-to-everything (V2X) information fusion technology or real-time dynamic path planning algorithms (such as vehicle trajectory prediction algorithms, time-to-collision (TTC) safety model algorithms). The AI system, based on a collision model, such as a model based on vehicle kinematic simulation, such as the CarSim simulation model, quickly simulates whether the current driving state is likely to cause a collision with environmental vehicles if the ADAS function is not enabled. For example, if the simulation results show that at the current vehicle speed of 50 km / h, the distance to the vehicle in front is 10 m and it is decelerating, and if the driver is not assisted in braking, a rear-end collision accident will occur after 1.5 seconds, then this scenario is immediately determined to be a high-risk scenario. When the simulation results indicate an obvious collision risk, the system automatically presents the risk simulation screen intuitively in the area of the AI coach screen watched by the driver. The simulated collision scenario can be displayed on the screen in the form of 3D animation or real-time AR (augmented reality): including the simulated dynamic demonstration of the distance between the current vehicle and the vehicle in front quickly shortening until the collision occurs; clearly marking the countdown time, distance, impact position, and possible damage degree (such as collision intensity, highlighted in red warning) of the possible collision; at the same time, accompanied by sound and light alarm prompts, so that the driver realizes that if the system does not take over in time, a serious accident may occur at this moment, thus improving the driver's vigilance. For example, when the trainee's line of sight falls on the screen prompt "The vehicle in front is decelerating, please pay attention", the collision risk simulation image of the vehicle in front is immediately superimposed on the screen, with the prompt: "The current distance is too close, there is a serious risk of rear-end collision! The system has activated emergency braking!"When the driver regains attention to the road ahead, the assisted driving system automatically prompts the driver to resume control of the vehicle. After the driver resumes active control of the vehicle, the system resumes monitoring the driver's attention status to respond to the next possible risk situation. Thus, not only is the triggering of the assisted driving system achieved, but also the potential accident risks are visually displayed through simulation, giving trainees a more direct visual impact and a profound understanding of driving risks, effectively cultivating the driver's safety awareness and emergency response capabilities. By simulating collision scenarios in real time, the dangerous consequences of failing to pay attention to the road in a timely manner are vividly demonstrated, intuitively strengthening the concept of safe driving for trainees. Combining vision detection technology with the real-time status of the vehicle and environmental data for integrated analysis, the vehicle is actively and accurately taken over, improving the safety of driving training and the intelligence and accuracy of vehicle control. A complete intelligent closed-loop process for driving training safety management is formed, effectively ensuring the safety of trainees during the training process.

[0059] In one embodiment, it further includes:

[0060] Determine the type of the surrounding environment vehicle based on the environmental image data;

[0061] Determine the theoretical structural parameters of the surrounding environment vehicle according to the type of the surrounding environment vehicle;

[0062] When it can be determined based on the driving state data of the vehicle, the driving state data of the surrounding environment vehicle, and the positional relationship at the moment of assisted driving takeover that there is a possible impact if assisted driving takeover is not performed, simulate the theoretical impact picture of the vehicle and the surrounding environment vehicle according to the driving state data of the vehicle, the driving state data of the surrounding environment vehicle, the positional relationship, and the theoretical structural parameters of the surrounding environment vehicle at the takeover moment. The theoretical impact picture includes the theoretical damage degree after the vehicle and the surrounding environment vehicle collide;

[0063] Display the theoretical impact picture on the display terminal.

[0064] It can be understood that in order to further enhance the safety of driving training and the safety awareness of trainees, during the AI intelligent driving teaching process, when a driving trainee's line of sight is shifted due to viewing the prompt information of the AI intelligent coach, there is a potential collision risk. For this scenario, this embodiment innovatively proposes that while the driving trainee's attention is transferred to the AI coach display screen, the assisted driving system is automatically awakened to take over vehicle control; furthermore, this embodiment additionally realizes accurate collision risk simulation on the original basis, precisely simulates the possible collision scenarios between vehicles according to the type characteristics of environmental vehicles, and intuitively displays them on the display terminal, comprehensively improving the safety awareness and risk awareness of driving trainees.

[0065] Exemplarily, when conducting AI intelligent driving teaching in a vehicle, first, a high-definition camera device (such as a high-definition RGB or infrared camera) is installed at a suitable position in front of the in-vehicle instrument panel to capture the facial image data of the driving trainee in real-time and continuously, and transmit it to the in-vehicle AI computing terminal for analysis and processing. In the in-vehicle computing terminal, a face recognition and gaze tracking algorithm based on a convolutional neural network (CNN) is deployed to monitor the gaze direction and head pose of the trainee in real-time and accurately determine the gaze landing area of the driving trainee. For example, based on the fusion calculation of the driver's eye feature points and head pose data, when it is determined that the gaze of the driving trainee continuously focuses on the in-vehicle AI coach screen for more than a preset threshold time (such as 1.5 seconds), the system automatically determines that the state of the trainee's distracted attention is established.

[0066] After determining the state of the driver's attention transfer, the system automatically triggers the rapid wake-up of the vehicle's advanced driver assistance system (ADAS) module to actively take over the vehicle driving control and perform short-term vehicle safety custody. Specifically, the intelligent assistance driving system sends control commands to the vehicle control system through the in-vehicle CAN bus or Ethernet communication interface to quickly manage the throttle, brake, and steering of the vehicle. Taking an actual scenario as an example, when the vehicle is driving on the road at a certain speed (such as 50 km / h) and the trainee's line of sight leaves the road for more than the set threshold due to watching the screen, the assisted driving system quickly activates the vehicle's automatic deceleration and automatic lane keeping functions, actively controls the steering wheel to correct the driving route to ensure that the vehicle is in a safe and controllable state and avoid accidents.

[0067] At the same time, at the starting moment when the assisted driving takes over and controls the vehicle, the system further automatically collects the real-time driving state data of the vehicle, including vehicle speed, acceleration, braking state, steering angle, and GPS real-time positioning data, as well as the driving state data (speed, acceleration, motion trajectory, driving direction) and real-time position relationship of the surrounding environment vehicles. Meanwhile, the intelligent system further uses a pre-trained deep learning object detection model, such as the YOLOv8 or SSD algorithm, to process the environmental image data captured by the vehicle's external camera in real-time, automatically identify the types of surrounding vehicles (such as cars, SUVs, buses, trucks, vans, motorcycles, etc.), and extract the corresponding vehicle feature parameters from the pre-stored vehicle parameter database according to the identified vehicle type, including but not limited to the length, width, height, weight range, collision intensity parameters, etc. of the vehicle. For example, when it is recognized that the vehicle in front is a large truck, the collision intensity parameters and size information of the large truck are automatically matched to improve the subsequent collision simulation accuracy.

[0068] Based on the real-time state data of the vehicle (such as vehicle speed 50 km / h, deceleration 2 m / s 2, when the brake pedal status is not activated), environmental vehicle status data (such as the vehicle in front suddenly braking and decelerating), and the positional relationship between vehicles, the intelligent system conducts precise vehicle collision simulation in real time. The system calls a vehicle dynamics simulation tool or model (such as the CarSim or PreScan simulation platform), inputs the above data and vehicle characteristic parameters, and quickly simulates the possible collision accidents that might occur if the assisted driving takeover is not performed, including precisely calculating the collision point, collision angle, collision speed, as well as the theoretical deformation degree of the vehicle and the possible injuries to the driver. During the simulation process, theoretical collision image animations in the form of 3D or augmented reality (AR) are generated in real time. For example, if a trainee approaches the vehicle in front at a speed of 50 km / h, the system simulation shows that if deceleration is not taken over in time, a rear-end collision will occur within 2 seconds, and clearly demonstrates the deformation degree of the vehicle structure and the compression condition of the cockpit at the moment of collision.

[0069] Finally, the real-time generated collision simulation images are dynamically displayed on the in-vehicle AI coach display terminal that the driver is paying attention to, visually presenting the simulation results of the possible collision accidents that might occur if the vehicle is not taken over in time. The displayed collision images can use red-highlighted warning images to mark the risk areas, and clearly present the process and results of the possible accidents in the form of 3D animations or augmented reality (AR), thereby visually strengthening the safety awareness of driving trainees. For example, the screen shows the impact process between the vehicle driven by the trainee and the large truck in front through animations or augmented reality, including the cockpit being squeezed, the damage of the vehicle body structure, and the theoretical injury status of the driver, strongly and intuitively reminding the trainee to pay attention to driving safety and the awareness of active prevention. Thus, the vehicle assisted driving function is automatically activated to effectively avoid the immediate collision risk caused by the driver's distracted attention. Through accurate and dynamic collision risk image simulation, it clearly and intuitively informs the driver of the serious accidents that might occur if not intervened in time, and the visual impact significantly enhances the sense of responsibility for safe driving. The automatic recognition of vehicle types and the corresponding collision model parameters are added to make the simulation more accurate and realistic, and intuitively improve the trainee's perception ability of driving risks. A complete technical closed-loop is achieved from driver attention monitoring to vehicle active takeover, real-time collision risk simulation, and intuitive result presentation, making the safety management of driving teaching reach a more intelligent, accurate, and efficient level.

[0070] In one embodiment, it further includes:

[0071] In the case where the line-of-sight information indicates that the driving trainee returns from watching the display terminal of the AI driving coach to the normal driving vision, the assisted driving function is exited to release the vehicle from the assisted driving takeover.

[0072] It can be understood that the current AI intelligent driving coach is gradually entering the driving training field. However, due to the uneven driving levels of driving trainees, they may frequently check the display terminal of the AI driving coach during training to obtain guidance information. This behavior of distracted attention will significantly increase the safety hazards during the driving training process. To effectively address this problem, this embodiment proposes an AI driving coach-assisted driving safety management method. This method monitors the trainee's line-of-sight status in real time, automatically determines whether the trainee is paying attention to the display terminal information. If the line of sight is shifted, the vehicle's assisted driving function is activated in real time for short-term custody, and further intelligent simulation and visual display of the vehicle collision risk are carried out. When the trainee refocuses on the road, the assisted driving function is smoothly exited, and the control right is returned to the driving trainee, thus realizing a complete closed-loop of active monitoring, automatic response, visual warning, and active intervention of safety risks during the driving training process. Through real-time line-of-sight monitoring, the active assisted driving system takes over vehicle control, accurate collision simulation, and precise identification of the environmental vehicle type, combined with real-time theoretical collision image display and the dynamic and smooth exit mechanism of the assisted driving system, a comprehensive closed-loop of active safety management for driving training is achieved. It significantly improves the active safety during the driving training process. Through automatic takeover control, the accident risk caused by driver distraction is effectively reduced; through collision simulation and real-time risk visual display, the driver's safety awareness is effectively improved, and the intuitive understanding of driving risks is enhanced; the introduction of environmental vehicle type and structural parameters further improves the accuracy of collision risk simulation and makes the risk assessment more credible; realizing the intelligent and smooth exit of the vehicle's assisted driving function effectively avoids the instability risk caused by the sudden exit of the system, and improves the system's stability and safety; it improves the driver's safe driving awareness, risk prediction ability, and driving skills, which is conducive to cultivating the trainee's correct driving concept and improving the long-term safety of driving behavior.

[0073] According to some embodiments, it further includes:

[0074] In the case where the line-of-sight information indicates that the driving trainee returns from watching the display terminal of the AI driving coach to the normal driving vision, obtain the driving control information of the driving trainee;

[0075] In the case where the driving control information matches the current driving environment, exit the assisted driving function to release the vehicle from the assisted driving custody.

[0076] It is understandable that when the driver's line-of-sight detection system determines that the line of sight of the driving trainee returns from the AI driving instructor display terminal to the normal driving field of view, the system then enters the stage of smoothly handing over control. While the trainee's line of sight returns to the road, the system further collects and judges the current driving control state data of the trainee in real time (such as the state of holding the steering wheel, the state of the accelerator pedal, the state of the brake pedal, the state of the steering force, etc.) to confirm that the driver is ready to resume autonomous driving. For example, if the system detects that the driver's both hands have grasped the steering wheel and the steering force meets the requirements of the driving state, and the driver's line of sight has resumed observing the road conditions for more than a preset duration (such as 1 second), then the system determines that the driver has resumed the safe driving state.

[0077] After the judgment of the above-mentioned driving state recovery is established, the vehicle assisted driving system then automatically starts to smoothly exit the custody mode, actively and gradually releases the custody of the control of the vehicle's steering wheel, accelerator, and brake, and clearly prompts the trainee through the display terminal. For example, a text prompt such as "Attention has returned to normal, the assisted driving function is about to exit, please take over the vehicle" appears on the screen and is accompanied by a voice prompt to clearly hand over the vehicle driving control to the driving trainee, enabling the vehicle to smoothly break away from the assisted driving custody, ensuring the safety and smoothness of the entire control handover process and not affecting the normal driving training experience of the trainee.

[0078] In one embodiment, it further includes:

[0079] When the vehicle is under the custody of assisted driving, generate an attention warning message.

[0080] In one embodiment, it further includes: real-time identifying the types of surrounding vehicles, determining the corresponding theoretical structural parameters of the vehicles; automatically calculating the possible blind spots of vision of the environmental vehicles based on the theoretical structural parameters; continuously monitoring and dynamically analyzing the positional relationship between the driver's vehicle and the environmental vehicles; when the own vehicle approaches the theoretical blind spot area of the environmental vehicle, actively implement intelligent warning to remind the driving trainee to pay attention to driving safety.

[0081] It is understandable that during the driving teaching and training process, especially for trainees with less driving experience, it is common for the attention to be distracted due to frequently checking the prompt information on the display terminal of the AI intelligent instructor, and it is extremely easy to ignore the theoretical blind spots of the road environmental vehicles, increasing the accident risk. Therefore, this embodiment proposes a more perfect driving teaching safety management method, that is, during the driving training process, through real-time image data collection and deep learning vehicle type recognition technology, actively determine the types of surrounding vehicles and automatically determine their theoretical blind spot ranges. When the training vehicle of the driving trainee approaches the theoretical blind spot area of the environmental vehicle, immediately perform dynamic positional relationship calculation in real time and issue a warning to remind the driving trainee to take evasive measures in time, significantly improving the safety of driving training.

[0082] Exemplarily, first, multi-angle high-definition wide-angle cameras are installed outside the driver training vehicle (such as in front of the vehicle head, on the sides of the left and right rearview mirrors, at the rear of the vehicle, etc.) to obtain high-definition image data of the vehicle's surrounding environment in real time. The in-vehicle AI processing terminal receives and processes the image data in real time, and applies deep learning object recognition technology to automatically detect, classify, and identify various vehicles appearing on the road around the vehicle, and accurately identify the specific types of vehicles, including but not limited to sedans, SUVs, passenger cars, trucks, buses, motorcycles, bicycles, etc. For example, when the training vehicle approaches a large truck, the system quickly identifies that the vehicle is a large truck type.

[0083] Subsequently, according to the identified vehicle type, the system automatically extracts the theoretical structure parameters of the corresponding type of vehicle from the pre-set vehicle parameter database, including vehicle size parameters (length, width, height), theoretical driver's field of view range parameters, and blind spot range. For example, for the identified truck, the corresponding theoretical blind spot range is automatically matched: the area directly below the front, the position behind the left cab, the area in front of the right, and the blind spot information at a certain distance behind. These blind spot information have been pre-stored in the system database and are automatically matched by the specific vehicle type.

[0084] On this basis, the in-vehicle system calculates the real-time position relationship between the vehicle driven by the trainee and the surrounding vehicles in real time. Specifically, the system uses in-vehicle millimeter-wave radar or lidar to monitor the precise distance, real-time position, relative speed, and trajectory information between the vehicle and the surrounding vehicles in real time, and generates a dynamic trajectory map of the position relationship between the vehicles. For example, when the relative distance between the trainee vehicle and the large truck identified in the front right is gradually approaching, and the position relationship shows that the vehicle is about to enter the theoretical blind spot range on the right front side of the truck, the system immediately performs blind spot risk calculation and determination.

[0085] If, according to the real-time dynamic position calculation result, the vehicle is about to enter or has entered the theoretical blind spot range of the surrounding vehicle, the system will actively generate and immediately send visual and voice warning prompts to the driver to avoid potential dangers. Specifically, a clear and eye-catching red warning sign will be automatically displayed on the screen of the in-vehicle AI coach terminal, reminding the driver: "Blind spot on the right side of the vehicle ahead (large truck), please stay away in time!" and other warning messages, supplemented by voice warning prompts to ensure that the driver clearly realizes the danger of the current driving state. For example, when the trainee is performing overtaking training and gradually approaches a large truck, if the vehicle approaches the theoretical blind spot of the truck, the system will actively display the blind spot position of the truck dynamically on the in-vehicle display terminal and issue a voice warning: "Attention, currently entering the blind spot of the vehicle ahead, please drive carefully!"

[0086] Furthermore, when the driver's line of sight information indicates that the driver's attention has returned from the AI driving instructor display terminal to the normal driving field of view, the system will automatically and smoothly exit the previously enabled assisted driving function. Specifically, the in-vehicle AI system monitors the driver's line of sight position and the state of attention returning to the road in real time, and determines whether the driver's hands have re-grasped the steering wheel (judged by the steering wheel grip sensor data) and whether the feet have re-stepped on the brake or accelerator pedal. If the driver's attention and driving actions return to normal for more than a preset duration (for example, 1 second), the system immediately issues a prompt to the driver to return control, such as "Attention has been restored, and the assisted driving system will exit soon", and then gradually and smoothly releases the custody intervention of the vehicle control right. The driver regains the vehicle control right, realizing an intelligent closed-loop of active safety risk management. Thus, by automatically identifying the vehicle type and dynamically calculating the theoretical blind area, the risk of the trainee approaching the blind area is timely warned, greatly reducing the risk of blind area collision accidents during driving training. Through the real-time dynamic display and warning of the theoretical blind area, the intuitive understanding and attention of driving trainees to the vehicle blind area risk are enhanced, and the safe driving awareness is improved. A complete technical closed-loop is achieved from driver attention monitoring, assisted driving active custody, to vehicle type recognition, real-time analysis and warning of theoretical blind area risk, collision risk simulation, and smooth handover of vehicle control right, effectively reducing the probability of accidental accidents during driving teaching and ensuring the overall safety of driving training. By integrating multiple AI technologies (face and line of sight monitoring, vehicle type recognition, dynamic position relationship calculation, automatic warning of theoretical blind area), an efficient and intelligent driving safety management system is formed, improving the overall teaching quality and safety guarantee ability of the AI intelligent driving instructor system.

[0087] Please refer to Figure 2 , an embodiment of the AI-assisted teaching driving practice safety management device in the embodiment of the present application may include:

[0088] An acquisition unit 201, configured to acquire image data of the current driving trainee to determine the line of sight information of the driving trainee based on the image data;

[0089] An acquisition unit 202, configured to acquire the environmental image data of the current vehicle when the line of sight information indicates that the driving trainee is watching the display terminal of the AI driving instructor;

[0090] A display unit 203, configured to display the environmental image data on the display terminal together.

[0091] In summary, the AI-assisted teaching driving practice safety management device provided by the above embodiments collects the image data of the current driving trainee to determine the line-of-sight information of the driving trainee based on the image data; when the line-of-sight information indicates that the driving trainee is watching the display terminal of the AI driving instructor, it acquires the environmental image data of the current vehicle; and displays the environmental image data on the display terminal together. Thus, the fusion of real-time environmental images can prevent trainees from ignoring the external situation due to excessive attention to the screen, effectively reducing the probability of accidents caused by distracted attention. Even if the trainees focus on the screen information, they can intuitively and real-time see the external environmental status, realizing the synchronization of driving prompts and environmental perception, and effectively improving the driver's intuitive understanding and judgment ability of the surrounding traffic environment. This method relies on real-time face recognition and line-of-sight detection technologies to achieve the synchronous update of trainee behavior monitoring and environmental image fusion, with a natural and smooth interaction process, enhancing the training experience of trainees. Trainees can see both AI prompts and real-time external conditions on the screen without frequently shifting their line of sight back and forth, significantly reducing the cognitive burden during driving and reducing tension and anxiety emotions.

[0092] Above Figure 2 The AI-assisted teaching driving practice safety management device in the embodiments of the present application has been described from the perspective of modular functional entities. Next, the AI-assisted teaching driving practice safety management device in the embodiments of the present application will be described in detail from the perspective of hardware processing. Please refer to Figure 3 , an embodiment of the AI-assisted teaching driving practice safety management device 300 in the embodiments of the present application includes:

[0093] An input device 301, an output device 302, a processor 303, and a memory 304. Among them, the number of processors 303 can be one or more, Figure 3 Taking one processor 303 as an example. In some embodiments of the present application, the input device 301, the output device 302, the processor 303, and the memory 304 can be connected by a bus or other means. Among them, Figure 3 Taking the connection by a bus as an example.

[0094] Among them, by calling the operation instructions stored in the memory 304, the processor 303 is used to execute the above steps.

[0095] By calling the operation instructions stored in the memory 304, the processor 303 is also used to execute Figure 1 Any one of the corresponding embodiments.

[0096] Please refer to Figure 4 , Figure 4 which is a schematic diagram of an embodiment of the electronic system provided by the embodiments of the present application.

[0097] As Figure 4As shown in the figure, an embodiment of the present application provides an electronic system, including a memory 410, a processor 420, and a computer program 411 stored in the memory 420 and executable on the processor 420. When the processor 420 executes the computer program 411, the above steps are implemented.

[0098] In the specific implementation process, when the processor 420 executes the computer program 411, it can implement Figure 1 any implementation manner in the corresponding embodiment.

[0099] Since the electronic system introduced in this embodiment is the device adopted for implementing an AI-assisted teaching driving practice safety management device in the embodiment of the present application, based on the method introduced in the embodiment of the present application, those skilled in the art can understand the specific implementation manner and various variations of the electronic system in this embodiment. Therefore, the specific implementation of how this electronic system implements the method in the embodiment of the present application will not be described in detail here. As long as the device adopted by those skilled in the art to implement the method in the embodiment of the present application belongs to the scope protected by the present application.

[0100] Please refer to Figure 5 , Figure 5 which is a schematic diagram of an embodiment of a computer-readable storage medium provided by an embodiment of the present application.

[0101] As Figure 5 shown in the figure, this embodiment provides a computer-readable storage medium 500, on which a computer program 511 is stored. When the computer program 511 is executed by a processor, the above steps are implemented.

[0102] In the specific implementation process, when the computer program 511 is executed by a processor, it can implement Figure 1 any implementation manner in the corresponding embodiment.

[0103] It should be noted that in the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0104] Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0105] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded computers, or other programmable data processing devices to generate a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices generate means for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0106] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0107] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operational steps are performed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0108] Embodiments of the present application also provide a computer program product, which includes computer software instructions that, when running on a processing device, cause the processing device to execute the processes in the AI-assisted teaching driving practice safety management method in the corresponding embodiments Figure 1 as described.

[0109] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium may be any available medium that can be stored by a computer or a data storage device such as a server or a data center that includes one or more integrated available media. The available medium may be a magnetic medium (such as a floppy disk, a hard disk, or a magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (SSD)).

[0110] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be described herein again.

[0111] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods may be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Another point is that the couplings, direct couplings, or communication connections shown or discussed with each other may be indirect couplings or communication connections through some interfaces, devices, or units, and may be in electrical, mechanical, or other forms.

[0112] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0113] In addition, in each embodiment of the present application, each functional unit may be integrated into one processing unit, may exist physically alone for each unit, or two or more units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of a software functional unit.

[0114] If the above integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it may be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, may be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs that can store program codes.

[0115] As described above, the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of each embodiment of the present application.

Claims

1. An AI-assisted teaching driving practice safety management method, characterized in that, Including: Collecting image data of the current driving trainee to determine the line-of-sight information of the driving trainee based on the image data; When the line-of-sight information indicates that the driving trainee is viewing the display terminal of the AI driving instructor, acquiring the environmental image data of the current vehicle; Displaying the environmental image data on the display terminal together.

2. The method according to claim 1, characterized in that, Further including: When the line-of-sight information indicates that the driving trainee is viewing the display terminal of the AI driving instructor, waking up the assisted driving function to perform assisted driving custody on the vehicle.

3. The method according to claim 2, characterized in that, Further including: Collecting the driving state data of the vehicle, the driving state data of surrounding environmental vehicles, and the position relationship at the starting moment of taking over assisted driving custody; When it can be determined based on the driving state data of the vehicle, the driving state data of surrounding environmental vehicles, and the position relationship at the moment of taking over assisted driving custody that there is a possibility of collision if assisted driving custody is not performed, simulating the theoretical collision picture of the vehicle and surrounding environmental vehicles according to the driving state data of the vehicle, the driving state data of surrounding environmental vehicles, and the position relationship at the moment of taking over custody; Displaying the theoretical collision picture on the display terminal.

4. The method according to claim 2, wherein Further including: Determining the type of the surrounding environmental vehicle based on the environmental image data; Determining the theoretical structural parameters of the surrounding environmental vehicle according to the type of the surrounding environmental vehicle; When it can be determined based on the driving state data of the vehicle, the driving state data of surrounding environmental vehicles, and the position relationship at the moment of taking over assisted driving custody that there is a possibility of collision if assisted driving custody is not performed, simulating the theoretical collision picture of the vehicle and surrounding environmental vehicles according to the driving state data of the vehicle, the driving state data of surrounding environmental vehicles, the position relationship, and the theoretical structural parameters of the surrounding environmental vehicle, where the theoretical collision picture includes the theoretical damage degree after the vehicle and the surrounding environmental vehicles collide; Displaying the theoretical collision picture on the display terminal.

5. The method according to claim 2, characterized in that, Further including: When the line-of-sight information indicates that the driving trainee returns from viewing the display terminal of the AI driving instructor to the normal driving vision, exiting the assisted driving function to release the vehicle from assisted driving custody.

6. The method according to claim 2, characterized in that, Further including: When the line-of-sight information indicates that the driving trainee returns from viewing the display terminal of the AI driving instructor to the normal driving vision, acquiring the driving control information of the driving trainee; When the driving control information matches the current driving environment, exiting the assisted driving function to release the vehicle from assisted driving custody.

7. The method according to claim 2, wherein Further including: Generating an attention warning message when performing assisted driving custody on the vehicle.

8. An AI-assisted teaching driving practice safety management device, characterized in that, Including: A collecting unit for collecting image data of the current driving trainee to determine the line-of-sight information of the driving trainee based on the image data; An acquiring unit for acquiring the environmental image data of the current vehicle when the line-of-sight information indicates that the driving trainee is viewing the display terminal of the AI driving instructor; A display unit for displaying the environmental image data on the display terminal together.

9. An electronic system, comprising a memory and a processor, characterized in that, When the processor is used to execute the computer program stored in the memory, it implements the steps of the AI-assisted teaching driving practice safety management method described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the AI-assisted teaching driving practice safety management method described in any one of claims 1 to 7.

Images