An intelligent traffic auxiliary correction method and system based on remote sensing monitoring

Abstract

The application discloses an intelligent traffic auxiliary correction method and system based on remote sensing monitoring, relates to the technical field of intelligent traffic assistance, and comprises effective point feedback and point driving correction. First, user historical eye movement and vehicle dynamic data are collected through the effective point feedback step, driving behavior concentration duration is calculated, behavior types are matched, and eye movement data occurrence rates are counted, current driving data and image recognition are combined, and the first preferred point of the user is determined; voice prompts are given and steering auxiliary force is adjusted. Secondly, historical remote sensing data are collected through the point driving correction step, invalid data are analyzed, and eye movement focus side view angle correction values are summarized; driving directions are prompted, and turning compensation force is applied. The application realizes precise driving assistance through multi-source data fusion, improves safety, adapts to different driving habits through the linkage of historical and real-time data, and improves the precision of the auxiliary system.

Description

A method and system for intelligent traffic assistance correction based on remote sensing monitoring Technical Field

[0001] This invention relates to the field of intelligent traffic assistance technology, and specifically to an intelligent traffic assistance correction method and system based on remote sensing monitoring. Background Technology

[0002] With the increasing penetration rate of intelligent vehicles, users have higher and higher demands for the accuracy and adaptability of intelligent traffic assistance systems. Intelligent traffic assistance systems have been widely used in vehicle driving, so there is a need for an intelligent traffic assistance correction method and system based on remote sensing monitoring.

[0003] Most mainstream driver assistance systems focus on the vehicle's own dynamic parameters or basic environmental perception, neglecting the core variable in the interaction between people, vehicles, and the environment—driver behavior characteristics. In high-speed cruising scenarios, the system cannot determine whether the driver is paying attention to key observation points such as road signs and rearview mirrors through eye-tracking data, and only pushes assistance information according to fixed logic. This leads to some drivers becoming distracted due to information redundancy or lack of key prompts. In congested urban traffic, without taking into account the driver's historical eye-tracking focus patterns when changing lanes, the assisted steering force does not match the user's driving habits, thus increasing the burden of operation.

[0004] Existing correction strategies have several problems: First, the judgment of the effectiveness of observation points lacks remote sensing data support. When rain or fog causes the roadside information collected by the camera to be blurry, the system still triggers correction according to the normal road condition threshold, which can easily lead to false corrections, such as incorrect application of steering compensation force. Second, the perception of the driver's state relies solely on real-time data and does not utilize historical eye-tracking and vehicle dynamic data to build a user driving behavior model. This makes it impossible to predict the driver's possible operational tendencies, such as the habit of checking the rearview mirror in advance on specific road sections, resulting in a correction response that lags behind actual driving needs. Summary of the Invention

[0005] In view of the above-mentioned technical shortcomings, the purpose of this invention is to provide an intelligent traffic auxiliary correction method and system based on remote sensing monitoring.

[0006] To solve the above technical problems, the present invention adopts the following technical solution: The present invention provides an intelligent traffic assistance correction method based on remote sensing monitoring, including the following steps: Step 1, effective location feedback: collect eye-tracking data and vehicle dynamic data of the user at each historical time stamp, analyze the eye-tracking data and vehicle dynamic data of the user at each historical time stamp, obtain the user's preferred location setting scheme, and at the same time set the user's traffic assistance correction scheme.

[0007] Step 2, Point-of-Site Driving Correction: Collect remote sensing data for each historical time stamp of the user, analyze the remote sensing data for each historical time stamp of the user, and set up a driving point-of-site correction scheme.

[0008] Preferably, the driving point correction scheme is set as follows: Driving point correction scheme: Collect the user's current effectiveness index. If the user's current effectiveness index is less than the preset standard effectiveness index, drive point correction is performed. Based on the current preset observation point, obtain the corrected focal lateral viewing angle corresponding to the focal lateral viewing angle of the current preset observation point. At the same time, obtain the focal lateral viewing angle corresponding to the observation point in the current eye movement data. If the focal lateral viewing angle corresponding to the current observation point is less than the corrected focal lateral viewing angle corresponding to the focal lateral viewing angle of the current preset observation point, prompt the user to drive laterally to the right, and apply a preset size of steering compensation force for right turn lateral deviation. If the focal lateral viewing angle corresponding to the current observation point is greater than the corrected focal lateral viewing angle corresponding to the focal lateral viewing angle of the current preset observation point, prompt the user to drive laterally to the left, and apply a preset size of steering compensation force for left turn lateral deviation.

[0009] On the other hand, the present invention provides an intelligent traffic assistance correction system based on remote sensing monitoring, including the following modules: an effective location feedback module, used to collect eye-tracking data and vehicle dynamic data of the user at each historical time stamp, analyze the eye-tracking data and vehicle dynamic data of the user at each historical time stamp, obtain the user's preferred location setting scheme, and set the user's traffic assistance correction scheme at the same time.

[0010] The point-of-use driving correction module is used to collect remote sensing data of the user's various historical timestamps, analyze the remote sensing data of the user's various historical timestamps, and set driving point-of-use correction schemes.

[0011] The beneficial effects of this invention are as follows: 1. This invention first collects historical eye movement and vehicle dynamic data of the user through an effective point feedback step. It first calculates the attention duration of driving behavior, then matches the behavior type and statistically analyzes the occurrence rate of eye movement data. Combining current driving data and image recognition, it determines the user's preferred point of view; provides voice prompts and adjusts the steering wheel assist force. Secondly, it collects historical remote sensing data through a point-based driving correction step, analyzes invalid data, and summarizes the correction value for the eye movement focus side-view angle; it prompts the driving direction and applies steering compensation force. This invention achieves precise driving assistance through multi-source data fusion, improving safety. By linking historical and real-time data, it adapts to different driving habits and improves the accuracy of the assistance system.

[0012] 2. This invention provides personalized driving adaptation. Based on historical data mining, it mines the eye movement focus patterns of users under various driving behaviors, which can accurately determine the user's preferred observation point. In addition, the steering wheel assist force adjustment can be matched with personal operating preferences, avoiding the problem of conflict between traditional universal solutions and user habits, and improving driving comfort and acceptability.

[0013] 3. This invention calculates the effectiveness index of observation points using remote sensing data, which can promptly identify information distortion scenarios and call correction parameters; at the same time, relying on the linkage analysis of historical and real-time data, it realizes the upgrade from passive to active correction, reduces correction lag and misjudgment, effectively reduces driving risks in complex road conditions, and increases the safety of traffic assistance.

[0014] 4. This invention integrates three types of data: eye-tracking, vehicle dynamics, and remote sensing. It not only grasps the driver's subjective state of focus but also understands the objective driving conditions and environmental effectiveness. This solves the problem that traditional systems rely on only a single data source and ignore the coordination between people, vehicles, and the environment. In abnormal weather conditions such as rainy days, it can simultaneously combine remote sensing data to judge the clarity of road signs and eye-tracking data to judge the driver's attention, making the assistance more comprehensive and reducing the safety of the assistance correction capability in abnormal weather. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] Figure 1 is a schematic diagram of the implementation steps of the method of the present invention.

[0017] Figure 2 is a schematic diagram of the system structure connection of the present invention. Detailed Implementation

[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] As shown in Figure 1, the present invention provides an intelligent traffic assistance correction method based on remote sensing monitoring, including the following steps: Step 1, effective location feedback: collect eye-tracking data and vehicle dynamic data of users at each historical time stamp, analyze the eye-tracking data and vehicle dynamic data of users at each historical time stamp, obtain the user's preferred location setting scheme, and set the user's traffic assistance correction scheme at the same time.

[0020] In one specific embodiment, the collection of eye-tracking data and vehicle dynamic data at each historical time stamp of the user is carried out in the following specific process: the eye-tracking data at each historical time stamp of the user includes, but is not limited to, the user's focus duration, focus distance, center-line focus distance, and focus item type at each historical time stamp.

[0021] The vehicle dynamic data for each historical time stamp of the user includes, but is not limited to, the lateral offset, vehicle speed, and steering angle for each historical time stamp of the user.

[0022] An infrared eye-tracking camera captures eye movement trajectories and records the duration of continuous focus on the same area, which is recorded as focus duration. The eye-tracking camera, combined with a laser ranging module, calculates the straight-line distance from the driver's visual focus to the vehicle by combining the eye's gaze direction with the laser ranging. Image recognition is performed by the camera to identify the position of the lane centerline and calculate the lateral distance between the visual focus and the centerline. At the same time, by capturing images of the road conditions ahead, image recognition technology is used to determine the category of the object corresponding to the driver's visual focus.

[0023] By using lane line recognition cameras and GPS positioning modules, and combining the position of the lane centerline with the vehicle's GPS coordinates, the lateral distance between the vehicle's center and the centerline is calculated. The vehicle's speed and steering angle are then read via the vehicle's CAN bus.

[0024] It should be noted that the centerline focal distance is: draw a perpendicular line from the focal point to the centerline to obtain the point where the focal point is perpendicular to the centerline, and record the distance from the user vehicle to the hammer point as the centerline focal distance.

[0025] For example, eye-tracking data: the driver's gaze duration was 2.3 seconds, the visual focus distance was 120 meters, the vertical point of the visual focus was at the center line point 100 meters ahead, and the observed object was a fence.

[0026] For example, vehicle dynamic data: lateral offset +3cm, vehicle speed 55km / h, steering angle 0°, "+" indicates the positive or negative value of the offset relative to the lane centerline, "+" indicates the right side, and "-" indicates the left side.

[0027] In one specific embodiment, the analysis of eye-tracking data and vehicle dynamic data of the user at each historical time stamp is carried out as follows: basic analysis is performed on the eye-tracking data and vehicle dynamic data of the user at each historical time stamp to obtain the user's driving behavior focus duration; based on the user's driving behavior focus duration, in-depth analysis is performed on the vehicle dynamic data of the user at each historical time stamp to obtain the historical time stamps corresponding to various types of user driving behaviors; then, eye-tracking data of the historical time stamps corresponding to various types of user driving behaviors is obtained; and in-depth analysis is performed on the eye-tracking data of the historical time stamps corresponding to various types of user driving behaviors to obtain the user's preferred location setting scheme.

[0028] In one specific embodiment, the basic analysis of the user's eye-tracking data and vehicle dynamic data at each historical time point is performed as follows: the user's attention duration at each historical time point is obtained from the eye-tracking data at each historical time point, and the lateral offset of the user's historical time point is obtained from the vehicle dynamic data at each historical time point. Each historical time point within the user's historical attention duration is recorded as a historical attention time point, thereby obtaining the lateral offset of each historical attention time point corresponding to the user's historical time point.

[0029] Based on the horizontal offset of each user's historical timestamp and corresponding focus historical timestamp, the horizontal offset change rate of each user's historical timestamp is calculated. The user's historical timestamps with a horizontal offset change rate greater than the preset standard horizontal offset change rate are recorded as change historical timestamps. Thus, the user's focus duration for each change historical timestamp is obtained. The user's focus duration for driving behavior is calculated by averaging the focus durations of each change historical timestamp.

[0030] It should be noted that the preset standard lateral offset change rate was obtained by staff through internet searches or experience, and the specific value was set by the staff.

[0031] In one specific embodiment, the in-depth analysis of the vehicle dynamic data of each historical time stamp of the user is carried out as follows: each historical time stamp within the user's focused driving behavior duration is recorded as a valid time stamp, thereby obtaining each valid time stamp corresponding to each historical time stamp of the user, and thus obtaining the vehicle dynamic data corresponding to each valid time stamp of each historical time stamp of the user. Lateral offset data, vehicle speed data, and steering angle data are obtained from the vehicle dynamic data corresponding to each valid time stamp of the user's historical time stamp, and the lateral offset rate, vehicle speed change rate, and steering angle change rate corresponding to each historical time stamp of the user are calculated.

[0032] The standard lateral offset rate, standard speed change rate, and standard steering angle change rate corresponding to various user driving behaviors are obtained from the database. Similarity is calculated to obtain the similarity of various driving behaviors corresponding to each historical timestamp of the user. The driving behavior type corresponding to the maximum similarity is recorded as the driving behavior type of each historical timestamp of the user. In this way, the historical timestamps corresponding to various user driving behaviors are statistically obtained.

[0033] It should be noted that the similarity calculation is based on existing technology, such as cosine similarity calculation.

[0034] Various driving behaviors include, but are not limited to, straight driving, lane changing, turning, and U-turns. The standard lateral deviation rate, standard speed change rate, and standard steering angle change rate corresponding to various user driving behaviors are obtained by staff based on experimental statistics. For example, the lateral deviation rate, speed change rate, and steering angle change rate corresponding to each experiment for various user driving behaviors are obtained through experiments, and the average values ​​are calculated to obtain the standard lateral deviation rate, standard speed change rate, and standard steering angle change rate corresponding to various user driving behaviors. The specific values ​​are set by staff.

[0035] In one specific embodiment, the deep analysis of eye-tracking data corresponding to various historical timestamps of user driving behaviors is carried out as follows: focal distance data, centerline focal distance data, and focal item type are obtained from the eye-tracking data corresponding to various historical timestamps of user driving behaviors, and the occurrence rates of various focal distances, centerline focal distances, and focal item types corresponding to various user driving behaviors are statistically obtained.

[0036] It should be noted that the centerline focal distance is as follows: draw a perpendicular line from the focal point to the centerline, and obtain the point where the focal point is perpendicular to the centerline. This point is denoted as the centerline focal point. The distance from the user's vehicle to the hammer point is denoted as the centerline focal distance.

[0037] It should be noted that the statistics are obtained by counting the number of times each focal distance, each centerline focal distance, and each focal item type appear for each type of user driving behavior. The total number of counts for each type of user driving behavior is then obtained. The number of times each focal distance, each centerline focal distance, and each focal item type appear for each type of user driving behavior is divided by the total number of counts for each type of user driving behavior to obtain the occurrence rate of each focal distance, the occurrence rate of each centerline focal distance, and the occurrence rate of each focal item type for each type of user driving behavior.

[0038] In one specific embodiment, the process of obtaining the user's preferred location setting scheme is as follows: User preferred location setting scheme: Collect the user's current vehicle driving data. Based on the user's current vehicle driving data, obtain the user's current driving behavior type. Use image recognition technology to obtain the focal distance, centerline focal distance, and focal item type of each preset observation point. Based on the occurrence rate of each focal distance, centerline focal distance, and focal item type corresponding to the user's current driving behavior type, record the occurrence rate as a probability to obtain the current behavior's focal distance probability, centerline focal distance probability, and focal item type probability of each preset observation point. Perform weighted calculation to obtain the user's selection probability of each preset observation point. Record the preset observation point with the highest selection probability as the user's preferred location.

[0039] It should be noted that the weighted calculation is as follows: the staff presets the weight factor corresponding to the focal distance, the weight factor corresponding to the center line focal distance, and the weight factor corresponding to the focal item type. The user's current behavior's current focal distance probability, center line focal distance probability, and focal item type probability at each preset observation point are multiplied by their corresponding weight factors, and then added together to obtain the user's current selection probability at each preset observation point.

[0040] In one specific embodiment, the process of setting the user traffic assistance correction scheme is as follows: collect the user's current eye movement data to obtain the occurrence rate of the current focal distance corresponding to various driving behaviors of the user, and record it as the probability of various driving behaviors at the current focal distance of the user. This yields the probability of various driving behaviors at the current centerline focal distance of the user and the probability of various driving behaviors for each focal item type. These are then weighted and calculated to obtain the probability of various driving behaviors of the user at the current time. The driving behavior type with the highest probability is recorded as the user's current preset driving behavior type.

[0041] It should be noted that the weighted calculation is as follows: the weight factors corresponding to the driving behavior probability, the weight factors corresponding to the driving behavior, and the weight factors corresponding to the focus item are obtained from the database. The probabilities of various driving behaviors at the current focus distance of the user, the probabilities of various driving behaviors at the current centerline focus distance, and the probabilities of various driving behaviors for each focus item type are multiplied by their corresponding weight factors, and then added together to obtain the probability of the user's current driving behaviors.

[0042] User traffic assistance correction scheme: If the user's current preset driving behavior type is the same as the current driving behavior type, no user traffic assistance correction will be performed. If the user's current preset driving behavior type is different from the current driving behavior type, traffic assistance correction will be performed: the text of the user's current preset driving behavior type will be enhanced by voice, and the steering wheel assist force will be changed according to the preset gradient to the assist force corresponding to the user's current preset driving behavior type.

[0043] It should be noted that when a user applies a force in a certain direction to the steering wheel, the force in the corresponding direction is increased through a mechanical structure, which is called the auxiliary force. The preset gradient change of the auxiliary force is set by the staff.

[0044] Step 2, Point-of-Site Driving Correction: Collect remote sensing data for each historical time stamp of the user, analyze the remote sensing data for each historical time stamp of the user, and set up a driving point-of-site correction scheme.

[0045] In one specific embodiment, the remote sensing data for each historical time stamp of the user is collected in the following specific process: The remote sensing data for each historical time stamp of the user includes, but is not limited to, the integrity, clarity, and interference inverse coefficient of the preset observation point under each historical time stamp. The integrity of the road lane centerline is collected by a low-altitude UAV. Image recognition technology is used to determine whether the road lane centerline is broken, thereby obtaining the complete length and broken length of the road lane centerline in the area corresponding to the preset observation point. The integrity is obtained by dividing the complete length of the road lane centerline at the preset observation point under each historical time stamp by the sum of the complete length and broken length of the area corresponding to the preset observation point.

[0046] It should be noted that a circle is drawn with the center of the centerline focal point corresponding to the predicted observation point as the center and the distance from the centerline focal point as the radius. The area inside the circle is the area corresponding to the preset observation point.

[0047] Sharpness is obtained by quantizing grayscale values. The grayscale values ​​of each pixel in the area corresponding to the preset observation point are collected by the camera. The mean grayscale value of the area corresponding to the preset observation point is obtained. The mean grayscale value is subtracted from the preset grayscale threshold and then divided by the grayscale threshold to obtain the sharpness.

[0048] It should be noted that the grayscale threshold is set by the staff, and the specific value is set by the staff.

[0049] Images of the corresponding areas of the predicted observation points are collected by cameras. The image area of ​​the obstructing object and the total image area are obtained through image recognition technology. The image area of ​​the obstructing object is subtracted from the total image area, and then divided by the total image area to obtain the interference inverse coefficient.

[0050] In one specific embodiment, the analysis of remote sensing data for each historical time stamp of the user is carried out as follows: The completeness, clarity, and interference inverse coefficient of the preset observation point at each historical time stamp are obtained from the remote sensing data of each historical time stamp of the user. After normalization, a weighted calculation is performed to obtain the validity index of each historical time stamp of the user. Historical time stamps with validity indices less than the preset standard validity index are recorded as invalid historical time stamps. Eye-tracking data for each invalid historical time stamp of the user is obtained. Then, the focal lateral viewing angle and the focal lateral viewing angle of the preset observation point are obtained from the eye-tracking data of each invalid historical time stamp of the user. The invalid corrected focal lateral viewing angles corresponding to each focal lateral viewing angle of the preset observation point are summarized, and the corrected focal lateral viewing angles corresponding to each focal lateral viewing angle of the preset observation point are calculated from the mean.

[0051] It should be noted that the weighted calculation after normalization is as follows: the completeness, sharpness, and inverse interference coefficient are divided by preset thresholds to perform normalization, resulting in the normalized values ​​of completeness, sharpness, and inverse interference coefficient for each preset observation point under each historical timestamp. Based on the weight factors corresponding to completeness, sharpness, and inverse interference coefficient preset by the staff, the normalized values ​​of completeness, sharpness, and inverse interference coefficient for each preset observation point under each historical timestamp are multiplied by their respective weight factors, and then summed to obtain the validity index of each historical timestamp for the user.

[0052] The standard effectiveness index is a threshold set by the staff. The setting always revolves around the core objective of distinguishing between the environmental conditions of normal driving observation points and the environmental conditions of driving without observation points. The specific value is set by the staff.

[0053] In one specific embodiment, the driving point correction scheme is set up as follows: Driving point correction scheme: Collect the user's current effectiveness index. If the user's current effectiveness index is less than the preset standard effectiveness index, driving point correction is performed. Based on the current preset observation point, obtain the corrected focal lateral viewing angle corresponding to the focal lateral viewing angle of the current preset observation point. At the same time, obtain the focal lateral viewing angle corresponding to the observation point in the current eye movement data. If the focal lateral viewing angle corresponding to the current observation point is less than the corrected focal lateral viewing angle corresponding to the focal lateral viewing angle of the current preset observation point, prompt the user to drive laterally to the right, and apply a preset size of steering compensation force for right turn lateral deviation. If the focal lateral viewing angle corresponding to the current observation point is greater than the corrected focal lateral viewing angle corresponding to the focal lateral viewing angle of the current preset observation point, prompt the user to drive laterally to the left, and apply a preset size of steering compensation force for left turn lateral deviation.

[0054] It should be noted that normalization, through the principle of dimensional consistency and mathematical standardization, can translate physical quantities with different properties into unitless standard values ​​or superimposed parameters of the same dimension. This eliminates the interference of different dimensions on the computational logic, allowing the formula to retain the original data distribution characteristics while possessing mathematical rationality and adaptability to objective laws. The descriptions are merely exemplary embodiments of the present invention and should not be construed as limiting the scope of the invention.

[0055] As shown in Figure 2, the present invention provides an intelligent traffic assistance correction system based on remote sensing monitoring, comprising the following modules: an effective location feedback module, a location driving correction module, and a database.

[0056] The effective point feedback module is connected to the point driving correction module, and both the effective point feedback module and the point driving correction module are connected to the database.

[0057] The effective location feedback module is used to collect eye-tracking data and vehicle dynamic data of users at various historical time stamps, analyze the eye-tracking data and vehicle dynamic data of users at various historical time stamps to obtain the user's preferred location setting scheme, and at the same time set the user's traffic assistance correction scheme.

[0058] The point-of-use driving correction module is used to collect remote sensing data of the user's various historical timestamps, analyze the remote sensing data of the user's various historical timestamps, and set driving point-of-use correction schemes.

[0059] The database stores the standard lateral offset rate, standard speed change rate, standard steering angle change rate, weighting factor corresponding to the probability of various user driving behaviors, weighting factor corresponding to various user driving behaviors, and weighting factor corresponding to the focus item.

[0060] The examples described in this invention are not limited to the specific embodiments listed above. The examples are merely illustrative to facilitate understanding of the invention and do not constitute a limitation on the scope of protection of this invention. Any modifications, equivalent substitutions, etc., made within the spirit and principles of this invention should be included within the scope of protection.

[0061] The above description is merely an example and illustration of the concept of the present invention. Those skilled in the art can make various modifications or additions to the specific embodiments described or use similar methods to replace them, as long as they do not deviate from the concept of the invention or exceed the scope defined in this specification, they should all fall within the protection scope of the present invention.

Claims

1. A method for intelligent traffic auxiliary correction based on remote sensing monitoring, characterized in that, The process includes the following steps: Step 1, Effective Location Feedback: Collect eye-tracking data and vehicle dynamic data from each historical time stamp of the user, analyze the eye-tracking data and vehicle dynamic data from each historical time stamp of the user to obtain the user's preferred location setting scheme, and simultaneously set the user's traffic assistance correction scheme; Step 2, Location Driving Correction: Collect remote sensing data from each historical time stamp of the user, analyze the remote sensing data from each historical time stamp of the user, and set the driving location correction scheme; The specific analysis process of the eye-tracking data and vehicle dynamic data from each historical time stamp of the user is as follows: [The text abruptly ends here, likely due to an incomplete sentence or a formatting error.] Eye-tracking data and vehicle dynamic data are used for basic analysis to obtain the user's attention span during driving. Based on the user's attention span, in-depth analysis is performed on the vehicle dynamic data at each historical time stamp to obtain the historical time stamps corresponding to various types of driving behaviors. This leads to the eye-tracking data for each historical time stamp corresponding to various types of driving behaviors. In-depth analysis is then performed on the eye-tracking data for each historical time stamp corresponding to various types of driving behaviors to obtain the user's preferred location setting scheme. The specific process for obtaining the user's preferred location setting scheme is as follows: User preferred location setting scheme: Data collection... Based on the user's current vehicle driving data, the user's current driving behavior type is obtained. Image recognition technology is used to obtain the focal distance, centerline focal distance, and focal item type of each preset observation point. Based on the occurrence rates of each focal distance, centerline focal distance, and focal item type corresponding to the user's current driving behavior type, these occurrence rates are recorded as probabilities. The probabilities of the current behavior's focal distance, centerline focal distance, and focal item type at each preset observation point are then obtained. These probabilities are weighted to obtain the selection probability of each preset observation point. The preset observation point with the highest selection probability is recorded as the user's preferred point. The specific process for setting the user's traffic assistance correction scheme is as follows: The user's current eye-tracking data is collected to obtain the current focal distance occurrence rate corresponding to each type of driving behavior, which is recorded as the probability of each type of driving behavior at the user's current focal distance. This is used to obtain the probability of each type of driving behavior at the user's current centerline focal distance and the probability of each type of driving behavior at each focal item type. These probabilities are weighted to obtain the probability of each type of driving behavior at the user's current driving behavior. The driving behavior type with the highest probability is recorded as the user's current preset driving behavior type. User traffic assistance correction scheme: If the user's current preset driving behavior type is the same as the current driving behavior type, no user traffic assistance correction is performed; if the user's current preset driving behavior type is different from the current driving behavior type, traffic assistance correction is performed: the text of the user's current preset driving behavior type is enhanced with speech, and the steering wheel assist force is changed according to a preset gradient to the assist force corresponding to the user's current preset driving behavior type; the remote sensing data of each historical time stamp of the user is analyzed, and the specific analysis process is as follows: the preset observation points at each historical time stamp are obtained from the remote sensing data of each historical time stamp of the user. The integrity, clarity, and interference coefficients of each position are normalized and then weighted to obtain the validity index of each historical timestamp of the user. Historical timestamps with validity indices less than the preset standard validity index are recorded as invalid historical timestamps. Eye-tracking data of each invalid historical timestamp of the user is obtained in this way. Then, the focal lateral gaze angle and the focal lateral gaze angle of the preset observation point are obtained from the eye-tracking data of each invalid historical timestamp of the user. The invalid corrected focal lateral gaze angles corresponding to each focal lateral gaze angle of the preset observation point are summarized, and the corrected focal lateral gaze angles corresponding to each focal lateral gaze angle of the preset observation point are calculated by averaging.

2. The intelligent traffic auxiliary correction method based on remote sensing monitoring according to claim 1, characterized in that, The basic analysis of eye-tracking data and vehicle dynamic data at each historical timepoint of the user is performed as follows: The user's attention duration at each historical timepoint is obtained from the eye-tracking data, and the lateral offset at each historical timepoint is obtained from the vehicle dynamic data. Each historical timepoint within the user's historical attention duration is recorded as a specific attention historical timepoint, thus obtaining the lateral offset of each user's historical timepoint corresponding to each attention historical timepoint. Based on the lateral offset of each user's historical timepoint corresponding to each attention historical timepoint, the lateral offset change rate corresponding to each user's historical timepoint is calculated. Historical timepoints with a lateral offset change rate greater than a preset standard lateral offset change rate are recorded as specific change historical timepoints, thus obtaining each user's specific change historical timepoint, and subsequently, the user's attention duration at each specific change historical timepoint. The average of the user's attention duration at each specific change historical timepoint is calculated to obtain the user's attention duration for driving behavior.

3. The intelligent traffic auxiliary correction method based on remote sensing monitoring according to claim 1, characterized in that, The in-depth analysis of vehicle dynamic data for each historical time stamp of the user is performed as follows: Each historical time stamp within the user's focused driving behavior duration is recorded as a valid time stamp, thus obtaining the vehicle dynamic data corresponding to each historical time stamp. Lateral offset data, vehicle speed data, and steering angle data are obtained from the vehicle dynamic data of each valid time stamp, and the lateral offset rate, vehicle speed change rate, and steering angle change rate corresponding to each historical time stamp are calculated. Standard lateral offset rate, standard vehicle speed change rate, and standard steering angle change rate corresponding to various types of user driving behavior are obtained from the database, and similarity calculations are performed to obtain the similarity of various types of user driving behavior for each historical time stamp. The driving behavior type with the highest similarity is recorded as the driving behavior type for each historical time stamp, and the historical time stamps corresponding to various types of user driving behavior are statistically obtained.

4. The intelligent traffic auxiliary correction method based on remote sensing monitoring according to claim 3, characterized in that, The deep analysis of eye-tracking data corresponding to various historical timestamps of user driving behaviors is carried out as follows: focal distance data, centerline focal distance data and focal item type are obtained from the eye-tracking data corresponding to various historical timestamps of user driving behaviors, and the occurrence rate of each focal distance, the occurrence rate of each centerline focal distance and the occurrence rate of each focal item type corresponding to various user driving behaviors are statistically obtained.

5. The intelligent traffic auxiliary correction method based on remote sensing monitoring according to claim 1, characterized in that, The driving point correction scheme is set up as follows: Driving point correction scheme: Collect the user's current effectiveness index. If the user's current effectiveness index is less than the preset standard effectiveness index, driving point correction is performed. Based on the current preset observation point, obtain the corrected focal lateral gaze angle corresponding to the focal lateral gaze angle of the current preset observation point. At the same time, obtain the focal lateral gaze angle corresponding to the observation point in the current eye movement data. If the focal lateral gaze angle corresponding to the current observation point is less than the corrected focal lateral gaze angle corresponding to the focal lateral gaze angle of the current preset observation point, prompt the user to drive laterally to the right, and apply a preset size of steering compensation force for right turn lateral deviation. If the focal lateral gaze angle corresponding to the current observation point is greater than the corrected focal lateral gaze angle corresponding to the focal lateral gaze angle of the current preset observation point, prompt the user to drive laterally to the left, and apply a preset size of steering compensation force for left turn lateral deviation.

6. An auxiliary correction system applying the intelligent traffic auxiliary correction method based on remote sensing monitoring as described in any one of claims 1-5, characterized in that, It includes the following modules: Effective location feedback module, which is used to collect eye-tracking data and vehicle dynamic data of users at each historical time stamp, analyze the eye-tracking data and vehicle dynamic data of users at each historical time stamp, obtain the user's preferred location setting scheme, and set the user's traffic assistance correction scheme at the same time. The point-of-use driving correction module is used to collect remote sensing data of the user's various historical timestamps, analyze the remote sensing data of the user's various historical timestamps, and set driving point-of-use correction schemes.

Images