Intelligent driving training method and device

Abstract

The invention provides an intelligent driving training method and device, and relates to the technical field of artificial intelligence, in particular to the technical field of intelligent traffic, intelligent driving and deep learning. A specific embodiment of the method comprises the steps of obtaining driving data generated during a test driving period of a user; determining the driving ability of the user based on the driving data; a training path is generated based on the driving ability.

Description

Technical Field

[0001] The present disclosure relates to the field of artificial intelligence technology, and in particular to the fields of smart transportation, intelligent driving, and deep learning. Background Art

[0002] As the automotive industry continues to flourish, the number of cars on the road is steadily increasing. However, novice drivers (those with less than one year of driving experience) and those who have not driven for more than three years face significant challenges. Due to their lack of driving experience, these drivers lack confidence and are unfamiliar with road conditions. This makes driving training difficult, especially in complex urban environments. Furthermore, traditional driving school training grounds cannot fully simulate actual road conditions, limiting training scenarios.

[0003] To improve their driving skills, novice drivers and those who haven't driven for a long time typically seek out offline training. These trainers typically take novice drivers and those who haven't driven for a long time to specific, low-traffic areas for training, based on their experience and road conditions. Summary of the Invention

[0004] The embodiments of the present disclosure provide an intelligent driving training method, apparatus, device, storage medium, and program product.

[0005] In a first aspect, an embodiment of the present disclosure proposes an intelligent driving training method, comprising: obtaining driving data generated by a user during a vehicle test drive; determining the user's driving ability based on the driving data; and generating a training path based on the driving ability.

[0006] In a second aspect, an embodiment of the present disclosure proposes an intelligent driving training device, comprising: a driving data acquisition module, configured to acquire driving data generated by a user during a test drive of a vehicle; a driving ability determination module, configured to determine the user's driving ability based on the driving data; and a training path generation module, configured to generate a training path based on the driving ability.

[0007] In a third aspect, an embodiment of the present disclosure proposes an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method described in the first aspect.

[0008] In a fourth aspect, an embodiment of the present disclosure proposes a non-transitory computer-readable storage medium storing computer instructions, where the computer instructions are used to enable a computer to execute the method described in the first aspect.

[0009] In a fifth aspect, an embodiment of the present disclosure proposes a computer program product, including a computer program, which implements the method described in the first aspect when executed by a processor.

[0010] The key or important features of the embodiments of the present disclosure are not intended to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Other features, objects, and advantages of the present disclosure will become more apparent upon reading the detailed description of the non-limiting embodiments made with reference to the following drawings. The drawings are provided for a better understanding of the present disclosure and do not constitute a limitation of the present disclosure. Among them: Figure 1 is a flow chart of an embodiment of the intelligent driving training method according to the present disclosure; Figure 2 is a flow chart of another embodiment of the intelligent driving training method according to the present disclosure; Figure 3 is a flow chart of another embodiment of the intelligent driving training method according to the present disclosure; Figure 4 is a structural diagram of an embodiment of an intelligent driving training device according to the present disclosure; Figure 5 is a block diagram of an electronic device used to implement the intelligent driving training method according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0012] The following description of exemplary embodiments of the present disclosure is made in conjunction with the accompanying drawings, including various details of the embodiments of the present disclosure to facilitate understanding. These details should be considered as merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for the sake of clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0013] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in the present disclosure may be combined with each other. The present disclosure will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0014] Figure 1 A process 100 of an embodiment of an intelligent driving training method according to the present disclosure is shown. The intelligent driving training method includes the following steps: Step 101: Acquire driving data generated by a user during a vehicle test drive.

[0015] In this embodiment, the execution entity of the intelligent driving training method can obtain driving data generated by the user during a test drive of the vehicle.

[0016] The intelligent driving training method is typically implemented by a server. The server can be either hardware or software. If the server is hardware, it can be implemented as a distributed server cluster consisting of multiple servers or as a single server. If the server is software, it can be implemented as multiple software programs or software modules (for example, to provide distributed services) or as a single software program or software module. This is not specifically limited here.

[0017] For novice drivers and those who haven't driven for a long time, test drives can be conducted. During a test drive, driving data related to the user or vehicle can be collected. This driving data may include, but is not limited to, facial image data, steering wheel grip strength data, steering wheel rotation data, brake pedal data, accelerator pedal data, vehicle angle change data, and driving trajectory data.

[0018] Step 102: Determine the user's driving ability based on the driving data.

[0019] In this embodiment, the execution entity may determine the user's driving ability based on the driving data.

[0020] In the case where the driving data includes facial image data collected by the vehicle-mounted camera, the user's blinking frequency and gaze frequency can be first determined based on the facial image data; then the user's tension score can be determined based on the blinking frequency and gaze frequency; finally, the user's driving ability can be determined based on the tension score.

[0021] In the case where the driving data includes steering wheel grip force data measured by a steering wheel torque sensor, the user's steering wheel grip force score can be first determined based on the steering wheel grip force data; and then the user's driving ability can be determined based on the steering wheel grip force score.

[0022] When the driving data includes facial image data collected by the vehicle-mounted camera, the frequency of the user's head angle change can be first determined based on the facial image data; then, based on the frequency of the head angle change, the user's head posture stability score can be determined; finally, based on the head posture stability score, the user's driving ability can be determined.

[0023] When the driving data includes both facial image data and steering wheel grip strength data, the user's psychological state score can also be determined based on the tension score, steering wheel grip strength score and head posture stability score; and then the user's driving ability can be determined based on the psychological state score.

[0024] In the case where the driving data includes steering wheel rotation data collected by an electronic power steering system, the user's steering wheel steering smoothness score can be first determined based on the steering wheel rotation data; and then the user's driving ability can be determined based on the steering wheel steering smoothness score.

[0025] In the case where the driving data includes brake pedaling data collected by a brake system pressure sensor, the user's brake control smoothness score can be first determined based on the brake pedaling data; and then the user's driving ability can be determined based on the brake control smoothness score.

[0026] In the case where the driving data includes throttle pedaling data collected by a throttle system pressure sensor, the user's throttle control smoothness score can be first determined based on the throttle pedaling data; and then the user's driving ability can be determined based on the throttle control smoothness score.

[0027] When the driving data includes steering wheel rotation data, brake pedaling data and throttle pedaling data at the same time, the user's control stability score can be determined based on the steering wheel steering smoothness score, the brake control smoothness score and the throttle control smoothness score; then, based on the control stability score, the user's driving ability can be determined.

[0028] In the case where the driving data includes vehicle angle change data collected by an inertial measurement unit, the user's lane change smoothness score can be first determined based on the vehicle angle change data; and then the user's driving ability can be determined based on the lane change smoothness score.

[0029] In the case where the driving data includes driving trajectory data collected by a positioning system, the user's traffic light reaction time score may be first determined based on the driving trajectory data; and then the user's driving ability may be determined based on the traffic light reaction time score.

[0030] When the driving data includes both vehicle angle change data and driving trajectory data, the user's road adaptability score can also be determined based on the lane change smoothness score and the traffic light reaction time score; and then the user's driving ability can be determined based on the road adaptability score.

[0031] In the case where the driving data includes driving trajectory data collected by the positioning system, the number of sudden lane changes, sudden brakes and sudden accelerations of the user can be determined first based on the driving trajectory data; then the user's risk awareness score can be determined based on the number of sudden lane changes, sudden brakes and sudden accelerations; finally, the user's driving ability can be determined based on the risk awareness score.

[0032] When the driving data also includes facial image data, steering wheel grip strength data, steering wheel rotation data, brake pedaling data, accelerator pedaling data, vehicle angle change data, and driving trajectory data, the user's driving ability score can also be determined based on the control stability score, road adaptability score, risk awareness score, and psychological state score.

[0033] Step 103: Generate a training path based on the driving ability.

[0034] In this embodiment, the execution entity may generate a training route based on driving ability.

[0035] Different driving ability scores correspond to different driving levels. For example, a driving ability score between 0-40 corresponds to the novice level, where you need to start with basic operations and choose simple roads; a driving ability score between 41-60 corresponds to the elementary level, where you can practice in low-traffic urban areas; a driving ability score between 61-80 corresponds to the intermediate level, where you can also practice on highways and complex intersections; and a driving ability score between 81-100 corresponds to the advanced level, where you can conduct advanced training on real, complex roads.

[0036] The embodiments of the present disclosure provide an intelligent driving training method that determines a user's driving ability based on driving data generated during a test drive of a vehicle, and automatically generates a personalized training path based on the user's driving ability, allowing the user to perform targeted driving training according to the personalized training path, thereby reducing the labor cost of intelligent driving training and improving the efficiency of intelligent driving training.

[0037] Continue to refer Figure 2 , which shows a process 200 of another embodiment of the intelligent driving training method according to the present disclosure. The intelligent driving training method includes the following steps: Step 201: Obtain personal driving information input by the user.

[0038] In this embodiment, the execution entity of the intelligent driving training method can obtain personal driving information input by the user.

[0039] A driving skills module can be added to the in-car navigation personal center. Clicking on the module will open a form page where users can manually enter their driving information. This information may include, but is not limited to, driving experience, last driving date, and driving fears. These fears may include, but are not limited to, fear of changing lanes, fear of parking, fear of overtaking, fear of driving at night, fear of merging lanes, fear of U-turns, fear of turning, fear of reversing, fear of highways, and fear of traffic jams.

[0040] Step 202: Generate a guidance route based on road condition information.

[0041] In this embodiment, the execution entity may generate a guidance route based on road condition information.

[0042] For novice drivers and those who haven't driven for a long time, a test drive can be provided. Because novice and inexperienced drivers lack driving experience, guidance routes are typically generated based on roads with minimal traffic and short distances. These guidance routes can be used to guide users along pre-selected roads.

[0043] Step 203 : Acquire driving data generated when the user test drives the vehicle along the guided route.

[0044] In this embodiment, the execution entity may obtain driving data generated when the user test drives the vehicle along the guided route.

[0045] During a test drive, users can collect driving data related to the user or vehicle along the guided route. This driving data may include, but is not limited to, facial image data, steering wheel grip strength data, steering wheel rotation data, brake pedaling data, accelerator pedaling data, vehicle angle change data, and driving trajectory data.

[0046] Step 204 : Determine the user's driving ability based on the driving data.

[0047] In this embodiment, the specific operation of step 204 has been Figure 1 In the illustrated embodiment, step 102 is described in detail and will not be repeated here.

[0048] Step 205 : Generate a training route based on the personal driving information and driving ability.

[0049] In this embodiment, the execution entity may generate a training route based on personal driving information and driving ability.

[0050] Different driving ability scores correspond to different driving levels. For example, a driving ability score between 0-40 corresponds to the novice level, where you need to start with basic operations and choose simple roads; a driving ability score between 41-60 corresponds to the elementary level, where you can practice in low-traffic urban areas; a driving ability score between 61-80 corresponds to the intermediate level, where you can also practice on highways and complex intersections; and a driving ability score between 81-100 corresponds to the advanced level, where you can conduct advanced training on real, complex roads.

[0051] Incorporating human driving data, training routes for different levels can also be adjusted. For novice or beginner drivers, route planning can avoid roads that are often driving fear points. For example, if a user is afraid of merging lanes, route planning will prioritize avoiding multi-lane, heavily trafficked roads, focusing training on single-lane, low-traffic routes to gradually build confidence. For intermediate or advanced drivers, routes with pre-set weights (e.g., 30%-50%) can be added to the route planning to address fear points, providing specialized training and helping overcome these fears.

[0052] from Figure 2 It can be seen that Figure 1 Compared to the corresponding embodiment, process 200 of the intelligent driving training method in this embodiment adds test drive guidance and training path adjustment steps. Thus, the solution described in this embodiment proactively guides users on test drives, improving the convenience of acquiring driving data. Training paths are adjusted for different levels based on user driving information. For novice or beginner users, fear-prone roads are avoided to enhance driving confidence. For intermediate and advanced users, fear-prone roads are proactively introduced to provide specialized training and help overcome fears.

[0053] Further references Figure 3 , which shows a process 300 of another embodiment of the intelligent driving training method according to the present disclosure. The intelligent driving training method includes the following steps: Step 301: Obtain facial image data of a user during a vehicle test drive captured by a vehicle-mounted camera.

[0054] In this embodiment, the execution subject of the intelligent driving training method can obtain facial image data of the user during the test drive of the vehicle captured by the vehicle camera.

[0055] Vehicle-mounted cameras can capture facial image data. For example, an on-board infrared camera (850nm / 940nm wavelength) can be aimed at the user's face, with a fixed focal length of 50-70cm, a sampling frequency of 60Hz-120Hz, and an infrared fill light (intensified to 10mW in low-light environments) simultaneously activated. This allows the on-board infrared camera to capture a grayscale image of the user's face (resolution ≥640×480) in real time, i.e., facial image data.

[0056] Step 302: Determine the user's blinking frequency and looking frequency based on the facial image data.

[0057] In this embodiment, the execution entity may determine the user's blinking frequency and looking frequency based on the facial image data.

[0058] The eye position can be located from facial image data. If the image area at the eye position contains blinking features, it can be considered that the user blinked. The blink frequency can be calculated by counting the number of blinks within a preset time period (e.g., 1 minute) and dividing the number of blinks by the preset time period.

[0059] In some embodiments, the user's blink frequency can be determined by the following steps: The first step is to locate the eye positions from the facial image data.

[0060] In the second step, edge detection is performed at the eye position using an edge detection operator to obtain the eyelid edge.

[0061] The edge detection operator may be, for example, a Canny operator, which may be an edge detection optimization operator derived under certain constraints.

[0062] The third step is to determine the degree of eyelid opening based on the eyelid margin.

[0063] The eyelid opening degree may be equal to the ratio E of the vertical eyelid distance to the average open eyelid distance (E=eyelid distance / average open eyelid distance). The average open eyelid distance may be the average of all open eyelid distances in an initial preset number (e.g., 30) of facial image frames.

[0064] The fourth step is to count the number of times the eyelid opening and closing degrees meet the first preset condition within a preset time period to obtain the blinking frequency.

[0065] The first preset condition may include the eyelid opening being less than a preset threshold (e.g., E < 0.3) and lasting longer than a preset duration (e.g., 80ms). For example, if E < 0.3 and lasts for more than 80ms, it can be considered a blink. Counting the number of blinks within a minute can yield the blink frequency.

[0066] In some embodiments, the user's viewing frequency can be determined by the following steps: The first step is to locate the eye positions from the facial image data.

[0067] In the second step, the corner detection algorithm is used to extract feature points at the eye position to obtain stable feature points.

[0068] The corner detection algorithm can be, for example, FAST (Features from Accelerated Segment Test) corner detection. FAST corner detection can determine whether a point of interest is a corner by judging the 16 pixels on the circle around the point of interest. If the current center pixel is dark or bright, it is determined to be a corner.

[0069] The FAST corner detector can be used to extract feature points at the eye position, delete duplicate feature points that are too close, and retain 50-100 high-quality feature points, that is, stable feature points.

[0070] In the third step, the sparse optical flow algorithm is used to solve the optical flow equation for the stable feature points and obtain the displacement vector of the stable feature points.

[0071] The sparse optical flow algorithm may be, for example, Lucas-Kanade. Lucas-Kanade is a widely used differential method for optical flow estimation, which assumes that the optical flow is constant in the neighborhood of a pixel point and then uses the least squares method to solve the basic optical flow equation for all pixels in the neighborhood.

[0072] The optical flow equation can be: .

[0073] in, and It is the gradient of the image in the x and y directions, which can reflect how fast the pixel brightness changes. It is the brightness change in the time dimension, which can reflect the brightness difference of pixels at the same position in adjacent frames. is the displacement vector of the feature point, that is, the speed of the point movement caused by eye movement.

[0074] Calculate the displacement vector between adjacent frames , such as shifting right by 2 pixels per frame, and using RANSAC (Random Sample Consensus) to remove abnormal displacement vectors. Abnormal displacement vectors typically exceed 10 pixels. If a feature point's displacement exceeds 10 pixels per frame, it far exceeds the normal eye movement speed and can be identified as eyelash occlusion or noise, and discarded.

[0075] The fourth step is to use the intrinsic parameter matrix of the vehicle-mounted camera to convert the displacement vector into a direction vector in the camera coordinate system.

[0076] The intrinsic parameter matrix K is a 3×3 matrix of the internal optical characteristics of the vehicle camera. It is only related to the hardware parameters of the vehicle camera itself and does not change with the external position of the vehicle camera.

[0077] The fifth step is to use the external parameter matrix to map the direction vector to the world coordinate system of the driving scene to obtain the line of sight.

[0078] Among them, the external parameter matrix [R t] is the position and orientation of the vehicle camera in the world coordinate system. For example, if the line of sight is at an angle of 30° to the front, then the line of sight is pointing to the left rearview mirror.

[0079] Step 6: Connect the end points of the sight lines of each frame of the facial image data in chronological order to generate a continuous sight line trajectory curve.

[0080] Step 7: Count the number of gaze trajectories that meet the second preset condition in the continuous gaze trajectory curve to obtain the gaze frequency.

[0081] The second preset condition may include cross-zone gaze shifts. The driving field of view can be divided into five areas: the road ahead, the left and right rearview mirrors, the instrument panel, and the center console. Frequent gaze shifts occur when the gaze shifts across these areas more than six times within one minute (e.g., shifting from the front to the left rearview mirror to the front to the right rearview mirror three times). Normal driving requires less than two shifts per minute.

[0082] In some embodiments, when determining blink frequency and gaze frequency, it is necessary to first locate the eye position from the facial image data. The eye position can be located by the following steps: In the first step, the pupil area is segmented from the facial image data using the maximum inter-class variance method.

[0083] Because the pupil has a high infrared absorption rate, the pupil region can be located using the maximum inter-class variance method. This method is an automatic thresholding method suitable for bimodal situations. It separates the image into two components: background and target based on its grayscale characteristics. The larger the inter-class variance between the background and target, the greater the difference between the two components of the image.

[0084] The second step is to calculate the pupil center coordinates in the pupil area using the least squares method.

[0085] The least squares method is a mathematical optimization technique that finds the best function matching data by minimizing the sum of squared errors. Least squares can be used to easily find unknown data and minimize the sum of squared errors between the found data and the actual data.

[0086] The third step is to locate the eye position in the facial image data based on the pupil center coordinates.

[0087] For example, the area of a circle with the pupil center coordinate as the center and a preset length as the radius is positioned as the eye position.

[0088] Step 303: Determine the user's stress level score based on the blinking frequency and the looking frequency.

[0089] In this embodiment, the execution entity may determine the user's stress level score based on the blinking frequency and the looking frequency.

[0090] Different blink rates correspond to different scores. Different gaze rates correspond to different scores. First, the blink rate and gaze rate are assigned component values, and then a weighted sum is taken to obtain the user's stress score.

[0091] For example, the blink frequency score is based on a 10-point scale. A blink frequency of 15-20 times / minute indicates normal blinking, and is therefore scored 10 points. A blink frequency of 10-15 times / minute or 20-25 times / minute is scored 7 points. A blink frequency of 5-10 times / minute or 25-30 times / minute is scored 5 points. A blink frequency below 5 times / minute or above 30 times / minute indicates a high level of nervousness, and is therefore scored 2 points.

[0092] For example, the gaze frequency score is based on a 10-point scale. A gaze frequency of less than 2 times / minute indicates that the user is focused and is scored 10 points; a gaze frequency of 2-5 times / minute is scored 8 points; a gaze frequency of 6-10 times / minute is scored 5 points; and a gaze frequency of more than 10 times / minute indicates that the user is highly nervous and is scored 2 points.

[0093] For example, a user's stress score can be calculated as follows: 0.6 × Gazing Frequency Score + 0.4 × Blinking Frequency Score. The Gazing Frequency Score has a weight of 0.6, and the Blinking Frequency Score has a weight of 0.4. The specific weight values are for example purposes only and can be adjusted based on actual circumstances.

[0094] Step 304: Determine the frequency of changes in the user's head angle based on the facial image data.

[0095] In this embodiment, the execution entity may determine the frequency of changes in the user's head angle based on facial image data.

[0096] The head position can be determined from facial image data. If the image area at the head position contains a head-turning feature, it can be considered that the user's head angle has changed. The frequency of head angle changes can be calculated by counting the number of head angle changes within a preset time period (e.g., 1 minute) and dividing the number of head angle changes by the preset time period.

[0097] Step 305: Determine the user's head posture stability score based on the frequency of head angle changes.

[0098] The execution entity may determine the user's head posture stability score based on the frequency of head angle changes.

[0099] Different head angle change frequencies can correspond to different head posture stability scores. For example, frequent head turning (e.g., head angle changes greater than 5 times / minute) can lower the head posture stability score. For example, if the user frequently turns their head (e.g., changes their head angle more than 5 times / minute), 2 points will be deducted for each additional head angle change, based on a baseline of 5 times / minute.

[0100] Step 306 : Obtain steering wheel grip force data measured by the steering wheel torque sensor during the user's vehicle test drive.

[0101] In this embodiment, the execution entity may obtain steering wheel grip force data of a user during a vehicle test drive measured by a steering wheel torque sensor, wherein the steering wheel torque sensor may measure the steering wheel grip force data of the user.

[0102] Step 307: Determine the user's steering wheel grip strength score based on the steering wheel grip strength data.

[0103] In this embodiment, the execution entity may determine the user's steering wheel grip strength score based on the steering wheel grip strength data.

[0104] Different steering wheel grip strengths can correspond to different steering wheel grip strength scores. For example, the steering wheel grip strength score is based on a 10-point scale. A steering wheel grip strength of 5N-15N indicates a normal steering wheel grip, which is worth 10 points. A steering wheel grip strength of less than 5N indicates a loose grip, which is worth 5 points. A steering wheel grip strength of more than 15N indicates a tight grip, which is worth 3 points.

[0105] Step 308 : Determine the user's psychological state score based on the tension level score, the steering wheel grip strength score, and the head posture stability score.

[0106] In this embodiment, the execution entity may determine the user's psychological state score based on the tension level score, the steering wheel grip strength score, and the head posture stability score.

[0107] The user's psychological state score is calculated by weighting the stress score, steering wheel grip strength score, and head posture stability score. For example, the user's psychological state score can be calculated as follows: stress score × 0.6 + steering wheel grip strength score × 0.2 + head posture stability score × 0.2. The stress score is weighted 0.6, the steering wheel grip strength score is weighted 0.2, and the head posture stability score is weighted 0.2. The specific weight values are for example purposes only and can be adjusted based on actual conditions.

[0108] Step 309 : Acquire steering wheel rotation data collected by the electronic power steering system during the user's vehicle test drive.

[0109] In this embodiment, the execution subject may obtain the steering wheel rotation data collected by the electronic power steering system during the user's vehicle test drive.

[0110] Step 310 : Determine the user's steering wheel smoothness score based on the steering wheel rotation data.

[0111] In this embodiment, the execution entity may determine the user's steering wheel smoothness score based on the steering wheel rotation data.

[0112] Different steering wheel rotation data can correspond to different steering smoothness scores. For example, the steering smoothness score is based on a 10-point scale. A steering wheel rotation rate of change greater than 50° / s indicates a sharp steering turn, which is assigned a score of 3 points. A steering wheel rotation rate of change of at least 10° / s and no greater than 50° / s indicates smooth steering, which is assigned a score of 7-10 points. A steering wheel rotation rate of change less than 10° / s indicates insufficient steering, which is assigned a score of 5 points.

[0113] Step 311: Obtain brake pedaling data collected by the brake system pressure sensor during the user's test drive of the vehicle.

[0114] In this embodiment, the execution subject may obtain the brake pedaling data collected by the brake system pressure sensor during the user's test drive of the vehicle.

[0115] Step 312: Determine the user's brake control smoothness score based on the brake pedaling data.

[0116] In this embodiment, the execution entity may determine the user's brake control smoothness score based on the brake pedaling data.

[0117] Different brake pedaling data can correspond to different brake control smoothness scores. For example, the brake control smoothness score is based on a 10-point scale. If the rate of change of the brake pedaling force is greater than 300N / s, it indicates that the vehicle is braking suddenly, and the corresponding score is 3 points. If the rate of change of the brake pedaling force is not less than 50N / s and not greater than 300N / s, it indicates that the vehicle is braking normally, and the corresponding score is 7-10 points. If the rate of change of the brake pedaling force is less than 50N / s, it indicates that the vehicle is braking insufficiently, and the corresponding score is 5 points.

[0118] Step 313: Acquire the accelerator pedaling data collected by the accelerator system pressure sensor during the user's test drive of the vehicle.

[0119] In this embodiment, the execution subject may obtain the accelerator pedaling data collected by the accelerator system pressure sensor during the user's test drive of the vehicle.

[0120] Step 314: Determine the user's throttle control smoothness score based on the throttle pedaling data.

[0121] In this embodiment, the execution entity may determine the user's throttle control smoothness score based on the throttle pedaling data.

[0122] Different throttle pedaling data can correspond to different throttle control smoothness scores. For example, the throttle control smoothness score is based on a 10-point scale. If the rate of change in throttle pedaling force is greater than 40% / s, it indicates that the vehicle is accelerating too hard, and the corresponding score is 3 points. If the rate of change in throttle pedaling force is not less than 10% / s and not greater than 40% / s, it indicates that the vehicle is accelerating smoothly, and the corresponding score is 7-10 points. If the rate of change in throttle pedaling force is less than 10% / s, it indicates that the vehicle is not accelerating enough, and the corresponding score is 5 points.

[0123] Step 315 : Determine the user's control stability score based on the steering wheel smoothness score, the brake control smoothness score, and the throttle control smoothness score.

[0124] In this embodiment, the execution entity may determine the user's control stability score based on the steering smoothness score, the brake control smoothness score, and the throttle control smoothness score.

[0125] The user's control stability score can be obtained by weighted summing the steering smoothness score, the brake control smoothness score, and the throttle control smoothness score. The user's control stability score can be, for example,: steering smoothness score × 0.4 + brake smoothness score × 0.3 + throttle control smoothness score × 0.3. Among them, the steering smoothness score has a weight of 0.4, the brake smoothness score has a weight of 0.3, and the throttle control smoothness score has a weight of 0.3. The specific values of the weights are for example only and can be adjusted according to actual conditions.

[0126] Step 316 : Obtain vehicle angle change data collected by the inertial measurement unit during the user's test drive of the vehicle.

[0127] In this embodiment, the execution subject may obtain the vehicle angle change data collected by the inertial measurement unit during the user's test drive of the vehicle.

[0128] Step 317 : Determine the user's lane change smoothness score based on the vehicle angle change data.

[0129] In this embodiment, the execution entity may determine the user's lane change smoothness score based on the vehicle angle change data.

[0130] Different vehicle angle change data can be associated with different lane change smoothness scores. For example, the lane change smoothness score is based on a 10-point scale. A vehicle angle change greater than 20° / s indicates a sharp lane change, which is assigned a score of 3 points. A vehicle angle change of at least 10° / s and no more than 20° / s indicates a smooth lane change, which is assigned a score of 7-10 points.

[0131] Step 318: Acquire the driving trajectory data of the user during the test drive of the vehicle collected by the positioning system.

[0132] In this embodiment, the execution entity may obtain the driving trajectory data collected by the positioning system during the user's test drive of the vehicle. The positioning system may collect the driving trajectory data.

[0133] Step 319 : Determine the user's traffic light reaction time score based on the driving trajectory data.

[0134] In this embodiment, the execution entity may determine the user's traffic light reaction time score based on the driving trajectory data.

[0135] In driving trajectory data, the time interval from the traffic light changing to the user starting the vehicle is the traffic light reaction time. Different traffic light reaction times can be associated with different traffic light reaction time scores. For example, a traffic light reaction time of less than 1 second corresponds to 10 points; a traffic light reaction time between 1-3 seconds corresponds to 7 points; a traffic light reaction time between 3-5 seconds corresponds to 5 points; and a traffic light reaction time greater than 5 seconds corresponds to 3 points.

[0136] Step 320 : Determine the user's road adaptability score based on the lane change fluency score and the traffic light reaction time score.

[0137] In this embodiment, the execution entity may determine the user's road adaptability score based on the lane change smoothness score and the traffic light reaction time score.

[0138] The user's Road Adaptability Score is calculated by weighting the Lane Change Smoothness Score and the Traffic Light Reaction Time Score. For example, the Road Adaptability Score can be calculated as: Lane Change Smoothness Score × 0.5 + Traffic Light Reaction Time Score × 0.5. The Lane Change Smoothness Score is weighted 0.5, and the Traffic Light Reaction Time Score is weighted 0.5. The specific weighting values are for example purposes only and may be adjusted based on actual circumstances.

[0139] Step 321: Acquire the driving trajectory data of the user during the test drive of the vehicle collected by the positioning system.

[0140] In this embodiment, the execution entity may obtain the driving trajectory data collected by the positioning system during the user's test drive of the vehicle. The positioning system may collect the driving trajectory data.

[0141] Step 322: Determine the number of sudden lane changes, sudden braking, and sudden accelerations of the user based on the driving trajectory data.

[0142] In this embodiment, the execution entity may determine the number of sudden lane changes, sudden braking, and sudden accelerations of the user based on the driving trajectory data.

[0143] In the driving trajectory data, the time interval from the start of lane change to the end of lane change is the lane change time. If the lane change time is less than the preset lane change time threshold, it means that the vehicle has changed lanes suddenly. The number of sudden lane changes within the preset time period is counted, and the number of sudden lane changes is divided by the preset time period to obtain the number of sudden lane changes. The time interval from the start of braking to the end of braking is the braking time. If the braking time is less than the preset braking time threshold, it means that the vehicle has braked suddenly. The number of sudden brakes within the preset time period is counted, and the number of sudden brakes is divided by the preset time period to obtain the number of sudden brakes. The time interval from the start of acceleration to the end of acceleration is the acceleration time. If the acceleration time is less than the preset acceleration time threshold, it means that the vehicle has accelerated suddenly. The number of sudden accelerations within the preset time period is counted, and the number of sudden accelerations is divided by the preset time period to obtain the number of sudden accelerations.

[0144] Step 323 : Determine the user's risk awareness score based on the number of sudden lane changes, sudden braking, and sudden acceleration.

[0145] In this embodiment, the execution entity may determine the user's risk awareness score based on the number of sudden lane changes, sudden braking, and sudden acceleration.

[0146] The user's risk awareness score can be obtained by taking a weighted sum of the number of sudden lane changes, sudden braking, and sudden acceleration. For example, the user's risk awareness score can be calculated as follows: number of sudden braking times (-5) + number of sudden acceleration times (-5) + number of sudden lane changes times (-10). Each sudden braking action deducts 5 points, each sudden acceleration action deducts 5 points, and each sudden lane change deducts 10 points. The specific deduction values are for example purposes only and can be adjusted based on actual circumstances.

[0147] Step 324 , determining the user's driving ability score based on the control stability score, the road adaptability score, the risk awareness score, and the psychological state score.

[0148] In this embodiment, the execution entity may determine the user's driving ability score based on the control stability score, the road adaptability score, the risk awareness score, and the psychological state score.

[0149] The user's driving ability score is calculated by weighting the control stability score, road adaptability score, risk awareness score, and psychological state score. For example, the user's driving ability score could be: 0.4 × control stability score + 0.3 × road adaptability score + 0.2 × risk awareness score + 0.1 × psychological state score. The control stability score has a weight of 0.4, the road adaptability score has a weight of 0.3, the risk awareness score has a weight of 0.2, and the psychological state score has a weight of 0.1. The specific weights are for example purposes only and can be adjusted based on actual circumstances.

[0150] In step 325 , the user's driving level is determined based on the driving ability score.

[0151] In this embodiment, the execution entity may determine the user's driving level based on the driving ability score.

[0152] Different driving ability scores correspond to different driving levels. For example, a driving ability score between 0-40 corresponds to the novice level, where you need to start with basic operations and choose simple roads; a driving ability score between 41-60 corresponds to the elementary level, where you can practice in low-traffic urban areas; a driving ability score between 61-80 corresponds to the intermediate level, where you can also practice on highways and complex intersections; and a driving ability score between 81-100 corresponds to the advanced level, where you can conduct advanced training on real, complex roads.

[0153] Step 326 : Select a road corresponding to the driving level and generate a training path.

[0154] In this embodiment, the execution entity may select a road corresponding to the driving level and generate a training route.

[0155] Different driving levels can correspond to different roads. Based on the driving level, you can select targeted roads to generate training paths.

[0156] For drivers at the novice level, indicating that their driving skills are still unstable, they can choose at least one of the following routes: roads with traffic below a first preset traffic threshold, roads with road complexity below a first preset complexity threshold, etc. Novice-level route planning selects low-traffic roads to avoid complex road conditions. Task types primarily focus on simple maneuvers, such as straight-line driving, slow-speed turns, and stop-and-go operations. Routes at this level will be wide and without complex intersections, avoiding highways and complex intersections.

[0157] For beginners, this means their driving control has gradually stabilized, but their ability to handle complex road conditions is still limited. They can choose at least one of the following types of roads: urban roads, traffic light roads, and lane-changing roads. The beginner route planning considers urban roads and introduces traffic light recognition and lane-changing exercises to moderately increase the challenge level.

[0158] For intermediate driving levels, users have relatively strong driving skills and can handle complex road conditions. They can choose at least one of the following routes: highways, roads with complex intersections, expressways, and elevated roads. Intermediate route planning can incorporate highways, complex intersections, expressways, and elevated roads to help users further improve their driving skills.

[0159] For advanced driving levels, indicating advanced driving skills, users can select at least one of the following routes: roads with traffic exceeding a second preset traffic threshold, highway ramps, roads with road complexity exceeding a second preset complexity threshold, emergency stop roads, and dynamic lane change roads. Advanced route planning can incorporate high-traffic roads, highway ramps, complex urban road conditions, as well as challenging driving tasks such as emergency stops and dynamic lane changes, challenging the driver's adaptability and on-the-spot reaction capabilities.

[0160] from Figure 3 It can be seen that Figure 1 Compared to the corresponding embodiment, process 200 of the intelligent driving training method in this embodiment emphasizes the driving data acquisition and driving ability assessment steps. By acquiring various driving data through various sensors and combining these data to assess driving ability, a comprehensive assessment of driving ability is achieved, improving the accuracy of the assessment. Furthermore, through intelligent path planning, sensor data collection, on-board camera recognition, and voice guidance, novice drivers can practice driving safely and effectively in a real-world road environment.

[0161] Further references Figure 4 As an implementation of the methods shown in the above figures, the present disclosure provides an embodiment of an intelligent driving training device. Figure 1 Corresponding to the method embodiment shown, the device can be specifically applied to various electronic devices.

[0162] like Figure 4As shown, the intelligent driving training device 400 of this embodiment may include: a driving data acquisition module 401, a driving ability determination module 402, and a training path generation module 403. The driving data acquisition module 401 is configured to acquire driving data generated by a user during a vehicle test drive; the driving ability determination module 402 is configured to determine the user's driving ability based on the driving data; and the training path generation module 403 is configured to generate a training path based on the driving ability.

[0163] In this embodiment, the specific processing of the driving data acquisition module 401, the driving ability determination module 402 and the training path generation module 403 and the technical effects thereof can be referred to in the respective embodiments. Figure 1 The relevant descriptions of steps 101-103 in the corresponding embodiment are not repeated here.

[0164] In some optional implementations of this embodiment, the driving data acquisition module 401 is further configured to: generate a guidance route based on road condition information; and acquire driving data generated when the user test drives the vehicle along the guidance route.

[0165] In some optional implementations of this embodiment, the driving data acquisition module 401 is further configured to: acquire facial image data of the user during a vehicle test drive captured by an on-board camera; and the driving ability determination module 402 is further configured to: determine the user's blinking frequency and looking frequency based on the facial image data; and determine the user's tension score based on the blinking frequency and looking frequency.

[0166] In some optional implementations of this embodiment, the driving ability determination module 402 is further configured to: locate the eye position from the facial image data; use an edge detection operator to perform edge detection at the eye position to obtain the eyelid edge; determine the eyelid opening and closing degree based on the eyelid edge; count the number of eyelid opening and closing degrees that meet the first preset condition within a preset time period to obtain the blinking frequency, wherein the first preset condition includes that the eyelid opening and closing degree is less than a preset opening and closing threshold and the duration exceeds a preset time length.

[0167] In some optional implementations of this embodiment, the driving ability determination module 402 is further configured to: use a corner detection algorithm to extract feature points at the eye position to obtain stable feature points; use a sparse optical flow algorithm to solve the optical flow equation for the stable feature points to obtain a displacement vector of the stable feature points; use the intrinsic parameter matrix of the vehicle-mounted camera to convert the displacement vector into a direction vector in the camera coordinate system; use the extrinsic parameter matrix to map the direction vector to the world coordinate system of the driving scene to obtain the line of sight; connect the end points of the line of sight of each frame of facial image in the facial image data in chronological order to generate a continuous line of sight trajectory curve; count the number of line of sight trajectories that meet the second preset condition in the continuous line of sight trajectory curve to obtain the looking frequency, wherein the second preset condition includes the cross-region transfer of the line of sight trajectory.

[0168] In some optional implementations of this embodiment, the driving ability determination module 402 is further configured to: segment the pupil area from the facial image data using the maximum inter-class variance method; calculate the pupil center coordinates in the pupil area using the least squares method; and locate the eye position in the facial image data based on the pupil center coordinates.

[0169] In some optional implementations of this embodiment, the driving data acquisition module 401 is further configured to: obtain steering wheel grip force data of the user during a vehicle test drive measured by a steering wheel torque sensor; and the driving ability determination module 402 is further configured to: determine the user's steering wheel grip force score based on the steering wheel grip force data.

[0170] In some optional implementations of this embodiment, the driving ability determination module 402 is further configured to: determine the frequency of changes in the user's head angle based on the facial image data; and determine the user's head posture stability score based on the frequency of changes in the head angle.

[0171] In some optional implementations of this embodiment, the driving ability determination module 402 is further configured to determine the user's psychological state score based on the tension level score, the steering wheel grip strength score, and the head posture stability score.

[0172] In some optional implementations of this embodiment, the driving data acquisition module 401 is further configured to: acquire steering wheel rotation data collected by the electronic power steering system during the user's test drive of the vehicle; and the driving ability determination module 402 is further configured to: determine the user's steering wheel steering smoothness score based on the steering wheel rotation data.

[0173] In some optional implementations of this embodiment, the driving data acquisition module 401 is further configured to: obtain brake pedaling data collected by the brake system pressure sensor during the user's test drive of the vehicle; and the driving ability determination module 402 is further configured to: determine the user's brake control smoothness score based on the brake pedaling data.

[0174] In some optional implementations of this embodiment, the driving data acquisition module 401 is further configured to: acquire throttle pedaling data collected by the throttle system pressure sensor during the user's test drive of the vehicle; and the driving ability determination module 402 is further configured to: determine the user's throttle control smoothness score based on the throttle pedaling data.

[0175] In some optional implementations of this embodiment, the driving ability determination module 402 is further configured to determine the user's control stability score based on the steering smoothness score, the brake control smoothness score, and the throttle control smoothness score.

[0176] In some optional implementations of this embodiment, the driving data acquisition module 401 is further configured to: obtain vehicle angle change data collected by the inertial measurement unit during the user's test drive of the vehicle; and the driving ability determination module 402 is further configured to: determine the user's lane change smoothness score based on the vehicle angle change data.

[0177] In some optional implementations of this embodiment, the driving data acquisition module 401 is further configured to: obtain driving trajectory data collected by the positioning system during the user's test drive of the vehicle; and the driving ability determination module 402 is further configured to: determine the user's traffic light reaction time score based on the driving trajectory data.

[0178] In some optional implementations of this embodiment, the driving ability determination module 402 is further configured to determine the user's road adaptability score based on the lane change fluency score and the traffic light reaction time score.

[0179] In some optional implementations of this embodiment, the driving data acquisition module 401 is further configured to: obtain driving trajectory data collected by the positioning system during the user's test drive of the vehicle; and the driving ability determination module 402 is further configured to: determine the number of sudden lane changes, sudden brakes and sudden accelerations of the user based on the driving trajectory data; and determine the user's risk awareness score based on the number of sudden lane changes, sudden brakes and sudden accelerations.

[0180] In some optional implementations of this embodiment, the driving ability determination module 402 is further configured to: determine the control stability score, road adaptability score, risk awareness score and psychological state score based on the driving data; determine the user's driving ability score based on the control stability score, road adaptability score, risk awareness score and psychological state score.

[0181] In some optional implementations of this embodiment, the training path generation module 403 is further configured to: determine the user's driving level based on the driving ability score; and select a road corresponding to the driving level to generate a training path.

[0182] In some optional implementations of this embodiment, the training path generation module 403 is further configured to: if the driving level is novice level, select at least one of the following roads: roads with traffic lower than a first preset traffic threshold, roads with road complexity lower than a first preset complexity threshold; if the driving level is elementary, select at least one of the following roads: urban roads, traffic light roads and lane-changing roads; if the driving level is intermediate, select at least one of the following roads: highways, complex intersection roads, expressways and elevated roads; if the driving level is advanced, select at least one of the following roads: roads with traffic higher than a second preset traffic threshold, highway ramps, roads with road complexity higher than a second preset complexity threshold, emergency stop roads, and dynamic lane-changing roads.

[0183] In some optional implementations of this embodiment, the intelligent driving training device 400 further includes: a driving information acquisition module configured to acquire personal driving information input by a user; and the training path generation module 403 is further configured to generate a training path based on the personal driving information and driving ability.

[0184] In some optional implementations of this embodiment, the training path generation module 403 is further configured to: if the driving level is novice or elementary, avoid driving fear points in the personal driving information.

[0185] In the technical solutions disclosed herein, the acquisition, storage, and application of user personal information involved comply with the provisions of relevant laws and regulations and do not violate public order and good morals.

[0186] According to an embodiment of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0187] Figure 5A schematic block diagram of an example electronic device 500 that can be used to implement embodiments of the present disclosure is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are provided as examples only and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0188] like Figure 5 As shown, device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. RAM 503 may also store various programs and data required for the operation of device 500. Computing unit 501, ROM 502, and RAM 503 are connected to each other via a bus 504. An input / output (I / O) interface 505 is also connected to bus 504.

[0189] Various components in device 500 are connected to I / O interface 505, including: an input unit 506, such as a keyboard, mouse, etc.; an output unit 507, such as various types of displays, speakers, etc.; a storage unit 508, such as a magnetic disk, optical disk, etc.; and a communication unit 509, such as a network card, modem, wireless communication transceiver, etc. The communication unit 509 allows device 500 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0190] The computing unit 501 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 501 performs the various methods and processes described above, such as the intelligent driving training method. For example, in some embodiments, the intelligent driving training method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 508. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 500 via the ROM 502 and / or the communication unit 509. When the computer program is loaded into the RAM 503 and executed by the computing unit 501, one or more steps of the intelligent driving training method described above can be performed. Alternatively, in other embodiments, the computing unit 501 can be configured to perform the intelligent driving training method by any other suitable means (e.g., via firmware).

[0191] Various embodiments of the systems and techniques described above can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0192] The program code for implementing the method of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device so that when the program code is executed by the processor or controller, the functions / operations specified in the flow chart and / or block diagram are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0193] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fibers, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0194] To provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0195] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or a web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0196] A computer system may include a client and a server. The client and server are generally remote from each other and typically interact through a communication network. The client-server relationship arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server in a distributed system, or a server integrated with a blockchain.

[0197] It should be understood that the various forms of processes shown above can be used to reorder, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solutions provided by this disclosure can be achieved. This is not limited herein.

[0198] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. An intelligent driving training method, comprising: Obtain driving data generated by users during their test drives of vehicles; determining the user's driving ability based on the driving data; Based on the driving ability, a training route is generated.

2. The method according to claim 1, wherein The obtaining of driving data generated by the user during the vehicle test drive includes: Generate a guidance path based on road condition information; Driving data generated by the user during a test drive of the vehicle along the guided route is acquired.

3. The method according to claim 1, wherein The obtaining of driving data generated by the user during the vehicle test drive includes: Obtaining facial image data of the user during a vehicle test drive captured by a vehicle-mounted camera; and The determining the user's driving ability based on the driving data includes: determining a blinking frequency and a looking frequency of the user based on the facial image data; A stress level score of the user is determined based on the blink frequency and the gaze frequency.

4. The method according to claim 3, wherein: The determining, based on the facial image data, the blinking frequency and the looking frequency of the user includes: locating eye positions from the facial image data; Using an edge detection operator, performing edge detection at the eye position to obtain an eyelid edge; determining an eyelid opening degree based on the eyelid margin; The blinking frequency is obtained by counting the number of eyelid opening and closing degrees that meet a first preset condition within a preset time period, wherein the first preset condition includes that the eyelid opening and closing degree is less than a preset opening and closing degree threshold and the duration exceeds a preset time length.

5. The method according to claim 4, wherein The determining of the user's blinking frequency and looking frequency based on the facial image data further includes: Using a corner detection algorithm, feature points are extracted at the eye positions to obtain stable feature points; Using a sparse optical flow algorithm, solving an optical flow equation for the stable feature point to obtain a displacement vector of the stable feature point; Using the intrinsic parameter matrix of the vehicle-mounted camera, the displacement vector is converted into a direction vector in the camera coordinate system; Using an extrinsic matrix, the direction vector is mapped to the world coordinate system of the driving scene to obtain the line of sight; Connecting the end points of the sight lines of each frame of facial image in the facial image data in chronological order to generate a continuous sight line trajectory curve; The number of gaze trajectories that meet a second preset condition in the continuous gaze trajectory curve is counted to obtain the looking frequency, wherein the second preset condition includes that the gaze trajectory shifts across regions.

6. The method according to claim 4, wherein: The locating eye positions from the facial image data comprises: Segmenting the pupil region from the facial image data using a maximum inter-class variance method; Calculating the pupil center coordinates in the pupil area using a least squares method; Based on the pupil center coordinates, eye positions in the facial image data are located.

7. The method according to claim 3, wherein: The obtaining of driving data generated by the user during the vehicle test drive includes: Obtaining steering wheel grip force data of the user during a vehicle test drive measured by a steering wheel torque sensor; and The determining the user's driving ability based on the driving data includes: Based on the steering wheel grip strength data, a steering wheel grip strength score of the user is determined.

8. The method according to claim 7, wherein: The determining the user's driving ability based on the driving data includes: determining a frequency of changes in the user's head angle based on the facial image data; Based on the frequency of the head angle change, a head posture stability score of the user is determined.

9. The method according to claim 8, wherein The determining the user's driving ability based on the driving data includes: The user's psychological state score is determined based on the tension level score, the steering wheel grip strength score, and the head posture stability score.

10. The method according to claim 1, wherein The obtaining of driving data generated by the user during the vehicle test drive includes: Obtaining steering wheel rotation data collected by the electronic power steering system during the user's test drive of the vehicle; and The determining the user's driving ability based on the driving data includes: A steering wheel steering smoothness score of the user is determined based on the steering wheel rotation data.

11. The method according to claim 10, wherein: The obtaining of driving data generated by the user during the vehicle test drive includes: Obtaining brake pedaling data collected by a brake system pressure sensor during the user's test drive of the vehicle; and The determining the user's driving ability based on the driving data includes: Based on the brake pedaling data, a brake control smoothness score of the user is determined.

12. The method according to claim 11, wherein The obtaining of driving data generated by the user during the vehicle test drive includes: Obtaining throttle pedaling data collected by a throttle system pressure sensor during the user's vehicle test drive; and The determining the user's driving ability based on the driving data includes: Based on the throttle pedaling data, a throttle control smoothness score of the user is determined.

13. The method according to claim 12, wherein: The determining the user's driving ability based on the driving data includes: A control stability score of the user is determined based on the steering wheel smoothness score, the brake control smoothness score, and the throttle control smoothness score.

14. The method according to claim 1, wherein The obtaining of driving data generated by the user during the vehicle test drive includes: Obtaining vehicle angle change data collected by an inertial measurement unit during the user's vehicle test drive; and The determining the user's driving ability based on the driving data includes: Determine a lane change smoothness score of the user based on the vehicle angle change data.

15. The method according to claim 14, wherein The obtaining of driving data generated by the user during the vehicle test drive includes: Obtaining driving trajectory data collected by a positioning system during the user's test drive of the vehicle; and The determining the user's driving ability based on the driving data includes: Based on the driving trajectory data, a traffic light reaction time score of the user is determined.

16. The method according to claim 15, wherein The determining the user's driving ability based on the driving data includes: The user's road adaptability score is determined based on the lane change fluency score and the traffic light reaction time score.

17. The method according to claim 1, wherein The obtaining of driving data generated by the user during the vehicle test drive includes: Obtaining driving trajectory data collected by a positioning system during the user's test drive of the vehicle; and The determining the user's driving ability based on the driving data includes: Determining the number of sudden lane changes, sudden braking, and sudden accelerations of the user based on the driving trajectory data; A risk awareness score of the user is determined based on the number of sudden lane changes, the number of sudden brakes, and the number of sudden accelerations.

18. The method according to claim 1, wherein The determining the user's driving ability based on the driving data includes: determining a control stability score, a road adaptability score, a risk awareness score, and a psychological state score based on the driving data; The user's driving ability score is determined based on the control stability score, the road adaptability score, the risk awareness score, and the psychological state score.

19. The method according to claim 18, wherein The generating of a training path based on the driving ability includes: determining a driving grade of the user based on the driving ability score; A road corresponding to the driving level is selected to generate the training path.

20. The method according to claim 19, wherein The selecting of the road corresponding to the driving level includes: If the driving level is a novice level, at least one of the following roads is selected: a road with a flow rate lower than a first preset flow rate threshold, or a road with a road complexity lower than a first preset complexity threshold; If the driving level is elementary, at least one of the following roads is selected: an urban road, a traffic light road, and a lane-changing road; If the driving level is intermediate, select at least one of the following roads: expressway, complex intersection road, expressway and elevated road; If the driving level is advanced, at least one of the following roads is selected: a road with a flow rate higher than a second preset flow rate threshold, a highway ramp, a road with a road complexity higher than a second preset complexity threshold, an emergency stop road, and a dynamic lane change road.

21. The method according to claim 20, wherein The method further comprises: Obtaining personal driving information input by the user; and The generating of a training path based on the driving ability includes: A training route is generated based on the personal driving information and the driving ability.

22. The method according to claim 21, wherein The selecting of the road corresponding to the driving level includes: If the driving level is novice or elementary, avoid the driving fear point road in the personal driving information; If the driving level is intermediate or advanced, the driving fear point roads in the personal driving information are increased according to a preset weight.

23. An intelligent driving training device, comprising: a driving data acquisition module configured to acquire driving data generated by a user during a test drive of a vehicle; a driving ability determination module configured to determine the user's driving ability based on the driving data; The training path generation module is configured to generate a training path based on the driving ability.

24. An electronic device comprising: at least one processor; as well as a memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 22.

25. A non-transitory computer-readable storage medium storing computer instructions, the computer instructions being configured to cause the computer to execute the method of any one of claims 1 to 22.

26. A computer program product comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 22.

Images