Vehicle scratch early warning system based on dynamic risk assessment and damage positioning method

Abstract

The invention relates to the technical field of automobile active safety, and provides a vehicle scratch early warning system based on dynamic risk assessment and a damage positioning method.The method comprises the steps that the contact distance between an object and a vehicle body is detected in real time through a capacitive sensing vehicle cover, and a three-level risk identification model is started according to the contact distance; acquiring relative speed of an object based on a millimeter-wave radar, and acquiring approaching time by an ultrasonic sensor; calculating a threat level according to a real-time risk scoring algorithm; when the threat level is larger than a threshold value, multi-sensor fusion analysis is triggered, and a visual report containing the three-dimensional coordinates of the damaged part is generated. According to the invention, a vehicle-surrounding deployment scheme of the capacitive sensing vehicle cover and the 4D millimeter-wave radar is adopted, so that hardware innovation is realized; according to the method, a dynamic risk scoring model and a damage 3D mapping engine are adopted, a damage 3D mapping algorithm based on physical parameters is initiated, and the risk is dynamically evaluated; according to the method, trigger type directional monitoring is adopted, so that the storage space is saved, and the privacy can be improved.

Description

Technical Field

[0001] The present invention relates to the field of automobile active safety technology, and in particular to a vehicle scratch warning system and damage location method based on dynamic risk assessment. Background Art

[0002] In the existing technology, sentry mode is often used to automatically monitor the surrounding environment when the vehicle is parked. However, sentry mode (such as Tesla) relies on continuous recording and moving object detection.

[0003] In the existing technology, patent US20200198581A uses a pure visual solution, which makes it difficult to accurately quantify contact parameters. Patent CN113442834A's simple motion detection cannot distinguish threat levels. Vehicle environment monitoring technology cannot achieve centimeter-level damage positioning under low power consumption, and lacks a risk dynamic assessment engine based on physical parameters. Summary of the Invention

[0004] The purpose of the present invention is to provide a vehicle scratch warning system and damage location method based on dynamic risk assessment, which adopts dynamic risk assessment to achieve damage location.

[0005] The present invention is implemented as follows: a vehicle scratch warning system and damage location method based on dynamic risk assessment, comprising:

[0006] The capacitive sensing system detects the contact distance between the object and the car body in real time and activates a three-level risk identification model based on the contact distance;

[0007] The relative speed of the object is obtained based on the millimeter-wave radar, and the approach time is obtained by the ultrasonic sensor;

[0008] Calculate threat levels based on a real-time risk scoring algorithm;

[0009] When the threat level is greater than the threshold, multi-sensor fusion analysis is triggered and a visual report containing the three-dimensional coordinates of the damage site is generated.

[0010] Preferably, the three-level risk identification model is activated according to the contact distance, specifically including:

[0011] When the contact distance is greater than 0.5m, it is judged as low risk and data recording is not activated;

[0012] When the contact distance is 0.3-0.5m, it is judged as medium risk and the ultrasonic sensor scan is activated;

[0013] When the contact distance is less than 0.3m, it is judged as high risk and triggers multi-sensor fusion analysis.

[0014] Preferably, the calculation formula of the real-time risk scoring algorithm is:

[0015] Risk=dv×t×k object

[0016] Among them, Risk is the risk score value, v is the relative speed of the object, t is the object's approach time, d is the minimum contact distance between the object and the vehicle body, k object is the risk factor of the object.

[0017] Preferably, the multi-sensor fusion analysis specifically includes:

[0018] Capacitive car cover electrode positioning: Determine the center point of the contact area (x c ,y c );

[0019] Millimeter-wave radar data fusion: Obtain the object's azimuth angle θ and the distance d from the object to the vehicle r , calculate the spatial offset:

[0020] Δx=d r ×cosθ

[0021] Δy=d r ×sinθ

[0022] Wherein, Δx is the X-axis projection compensation of the millimeter radar wave, and Δy is the Y-axis projection compensation of the millimeter radar wave;

[0023] Ultrasonic time domain calibration: Correct the distance data d based on the duration of the approach t u :

[0024]

[0025] Among them, d u ′ is the sliding average filter value of ultrasonic ranging, n is the size of the sliding window, d u,i is the distance between the object and the vehicle measured at the i-th sampling moment, v i is the instantaneous velocity of the object at the i-th sampling moment, Δt i is the time interval between adjacent sampling points;

[0026] The calculation formula for the coordinates of the damaged part in the three-dimensional coordinate visualization report is:

[0027] Three-dimensional coordinate mapping: The weighted least squares method is used to generate the coordinates (x, y, z) of the injury site;

[0028]

[0029] b=[x c +Δx,y c +Δy,du ′] T

[0030] Among them, A is the sensor position matrix, A T is the transpose of A, W is the weight matrix, and b is the observation vector.

[0031] Preferably, the visualization report includes: a three-dimensional reconstruction of the contact object's motion trajectory, a damage probability prediction value, and characteristic information of the suspected object.

[0032] A vehicle scratch warning system based on dynamic risk assessment, characterized by comprising:

[0033] The perception layer uses multiple sensors to detect the contact distance between objects and the vehicle body in real time and establish a three-level risk contact identification model;

[0034] The edge computing layer uses a real-time risk scoring algorithm to evaluate and calculate the risk of scratches;

[0035] The interactive layer is used to push reports containing three-dimensional positioning maps and suspect information.

[0036] Preferably, the sensing layer includes a capacitive sensing car cover, a millimeter wave radar and an ultrasonic sensor;

[0037] Capacitive sensing car cover is arranged around the car, and the capacitive sensing car cover is composed of a distributed electrode array for real-time detection of the contact distance between the object and the car body;

[0038] 4D millimeter-wave radar array distribution, used to obtain the relative speed of objects;

[0039] Ultrasonic sensor, used to detect when an object is close to you.

[0040] Preferably, the edge computing layer uses a real-time risk scoring algorithm to evaluate and calculate the scratch risk, specifically including:

[0041] The dynamic risk assessment engine running on the Renesas R-Car V3H chip;

[0042] Real-time execution risk formula: Risk = dv × t × k object ;

[0043] Dynamic configuration of object risk coefficient, risk coefficient k for metal objects metal =1.5, the risk factor k of the organism biology =0.8.

[0044] Preferably, the interactive layer pushes a report containing a three-dimensional positioning map and information about the suspect, specifically including:

[0045] One-click generation of insurance reporting materials function module;

[0046] Intelligent estimation prompt of repair costs;

[0047] Directionally activate the image of the suspect captured by the camera.

[0048] Compared with the prior art, the present invention has the following beneficial effects:

[0049] 1. This invention adopts a capacitive sensing car cover and 4D millimeter wave radar around the car deployment solution to achieve hardware innovation;

[0050] 2. This invention uses a dynamic risk scoring model and a 3D damage mapping engine, pioneering a 3D damage mapping algorithm based on physical parameters to dynamically assess risk;

[0051] 3. The present invention adopts trigger-type directional monitoring, only identifies potential threat behaviors, and only saves risk event feature data, replacing the storage structure of the original video with feature data, which not only saves storage space but also improves privacy. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0053] Figure 1 is a flow chart of the method of embodiment 1 of the present invention;

[0054] Figure 2 This is a flow chart of the three-level risk identification model in Example 1 of the present invention;

[0055] Figure 3 This is a technical chain flow chart of Example 2 of the present invention;

[0056] Figure 4 This is a workflow diagram of Example 2 of the present invention. DETAILED DESCRIPTION

[0057] In order to more fully understand the technical content of the present invention, the technical solution of the present invention is further introduced and illustrated in conjunction with specific embodiments below, but is not limited thereto. The technical solution in the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is only a part of the embodiment of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0058] Example 1

[0059] refer to Figure 1 and Figure 2 The vehicle scratch damage location method based on dynamic risk assessment includes:

[0060] S1: Capacitive sensing is used to detect the contact distance between the object and the car body in real time, and a three-level risk identification model is activated based on the contact distance;

[0061] The three-level risk identification model is activated based on the contact distance, specifically including:

[0062] When the contact distance is greater than 0.5m, it is judged as low risk and data recording is not activated;

[0063] When the contact distance is 0.3-0.5m, it is judged as medium risk and the ultrasonic sensor scan is activated;

[0064] When the contact distance is less than 0.3m, it is judged as high risk and triggers multi-sensor fusion analysis.

[0065] S2: The relative speed of the object is obtained based on the millimeter-wave radar, and the approach time is obtained by the ultrasonic sensor;

[0066] S3: Calculates threat level based on real-time risk scoring algorithm;

[0067] The calculation formula of the real-time risk scoring algorithm is:

[0068] Risk=dv×t×k object

[0069] Among them, Risk is the risk score value, v is the relative speed of the object, t is the object's approach time, d is the minimum contact distance between the object and the vehicle body, k object The risk factor of the object

[0070] S4: When the threat level is greater than the threshold, multi-sensor fusion analysis is triggered and a visual report containing the three-dimensional coordinates of the damage site is generated.

[0071] Multi-sensor fusion analysis specifically includes:

[0072] Capacitive car cover electrode positioning: Determine the center point of the contact area (x c ,y c );

[0073] Millimeter-wave radar data fusion: Obtain the object's azimuth angle θ and the distance d from the object to the vehicle r , calculate the spatial offset:

[0074] Δx=d r ×cosθ

[0075] Δy=d r ×sinθ

[0076] Wherein, Δx is the X-axis projection compensation of the millimeter radar wave, and Δy is the Y-axis projection compensation of the millimeter radar wave;

[0077] Ultrasonic time domain calibration: Correct the distance data d based on the duration of the approach t u :

[0078]

[0079] Among them, d u ′ is the sliding average filter value of ultrasonic ranging, n is the size of the sliding window, d u,i is the distance between the object and the vehicle measured at the i-th sampling moment, v i is the instantaneous velocity of the object at the i-th sampling moment, Δt i is the time interval between adjacent sampling points.

[0080] The visualization report includes: a three-dimensional reconstruction of the contact object's motion trajectory, a damage probability prediction value, and characteristic information of the suspected object.

[0081] The calculation formula for the coordinates of the damaged part in the three-dimensional coordinate visualization report is:

[0082] Three-dimensional coordinate mapping: The weighted least squares method is used to generate the coordinates (x, y, z) of the injury site;

[0083]

[0084] b=[x c +Δx,y c +Δy,d u ′] T

[0085] Among them, A is the sensor position matrix, A T is the transpose of A, W is the weight matrix, and b is the observation vector.

[0086] A vehicle scratch warning system based on dynamic risk assessment, characterized by comprising:

[0087] The perception layer uses multiple sensors to detect the contact distance between objects and the vehicle body in real time and establish a three-level risk contact identification model;

[0088] The sensing layer includes capacitive sensing car cover, millimeter wave radar and ultrasonic sensor;

[0089] Capacitive sensing car cover is arranged around the car, and the capacitive sensing car cover is composed of a distributed electrode array for real-time detection of the contact distance between the object and the car body;

[0090] 4D millimeter-wave radar array distribution, used to obtain the relative speed of objects;

[0091] Ultrasonic sensor, used to detect when an object is close to you.

[0092] Multimodal threat behavior analysis using multiple sensors:

[0093] sensor Detection parameters Threat determination criteria millimeter-wave radar Relative speed>5km / h Fast approach means high risk Ultrasonic sensors Continuous proximity time> 2s Suspicious detention behavior Capacitive car cover <![CDATA[Contact area > 10 cm 2 > Eliminate interference from small objects such as flying insects

[0094] The edge computing layer uses a real-time risk scoring algorithm to evaluate and calculate the risk of scratches;

[0095] The edge computing layer uses a real-time risk scoring algorithm to evaluate and calculate the risk of scratches, specifically including:

[0096] The dynamic risk assessment engine running on the Renesas R-Car V3H chip;

[0097] Real-time execution risk formula: Risk = dv × t × k object ;

[0098] Dynamic configuration of object risk coefficient, risk coefficient k for metal objects metal =1.5, the risk factor k of the organism biology =0.8.

[0099] The interactive layer is used to push reports containing 3D positioning maps and suspect information;

[0100] The interactive layer pushes a report containing a three-dimensional positioning map and suspect information, specifically including:

[0101] One-click generation of insurance reporting materials function module;

[0102] Intelligent estimation prompt of repair costs;

[0103] Directionally activate the image of the suspect captured by the camera.

[0104] The essential differences between this invention and the sentinel mode are:

[0105] Technical Dimension Tesla Sentry Mode This program Innovation and Breakthrough Working Logic Continuous recording + motion object trigger saving Dynamic risk assessment triggers targeted monitoring 92% reduction in power consumption Identify the target All moving objects Only identify potentially threatening behaviors False positives dropped by 87% Output Video clips Injury site diagram + risk event report Information density increased by 300% Privacy Protection Full video recording can easily violate privacy Only save risk event characteristic data Comply with strict GDPR standards

[0106] Example 2

[0107] refer to Figure 3 and Figure 4 , roadside parking scene:

[0108] 1. A courier tricycle passes close to the right side of the vehicle at a speed of 8 km / h (distance < 0.3 m);

[0109] 2. System trigger:

[0110] Directionally activated right-view camera captures three-wheeled license plates;

[0111] The sensor performs multimodal threat behavior analysis, including capacitive vehicle cover proximity detection, 1.8 seconds of contact → dynamic risk assessment engine → millimeter-wave radar behavior analysis. The millimeter-wave radar determines that the speed is greater than the risk threshold. For example, the sensor detects an object approaching rapidly at a distance of 0.2 meters.

[0112] The processor then calculates the risk value, for example: a risk value of 86 is high risk;

[0113] If a high risk is detected, the high-definition camera is activated to take directional photos of the approaching object. For example, the right-view camera is activated to capture the license plate of a three-wheeled vehicle, and a vector diagram of the damage location is generated through 3D mapping of the damage area.

[0114] 3. Mobile phone push notification when the user returns:

[0115] Send visual directional reports to the user terminal, such as "Please check the right rear door."

[0116]

[0117] The embodiments described above are only part of the embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any ordinary technician in the field, within the technical scope disclosed by the present invention, can make equivalent replacements or changes based on the technical solutions and inventive concepts of the present invention, which should be covered by the scope of protection of the present invention.

Claims

1. A vehicle scratch damage location method based on dynamic risk assessment, characterized in that: Includes the following: The capacitive sensing system detects the contact distance between the object and the car body in real time and activates a three-level risk identification model based on the contact distance; The relative speed of the object is obtained based on the millimeter-wave radar, and the approach time is obtained by the ultrasonic sensor; Calculate threat levels based on a real-time risk scoring algorithm; When the threat level is greater than the threshold, multi-sensor fusion analysis is triggered and a visual report containing the three-dimensional coordinates of the damage site is generated.

2. The vehicle scratch damage location method based on dynamic risk assessment according to claim 1, characterized in that: The three-level risk identification model is activated based on the contact distance, specifically including: When the contact distance is greater than 0.5m, it is judged as low risk and data recording is not activated; When the contact distance is 0.3-0.5m, it is judged as medium risk and the ultrasonic sensor scan is activated; When the contact distance is less than 0.3m, it is judged as high risk and triggers multi-sensor fusion analysis.

3. The vehicle scratch damage location method based on dynamic risk assessment according to claim 1, characterized in that: The calculation formula of the real-time risk scoring algorithm is: Risk=dv×t×k object Among them, Risk is the risk score value, v is the relative speed of the object, t is the object's approach time, d is the minimum contact distance between the object and the vehicle body, k object is the risk factor of the object.

4. The vehicle scratch damage location method based on dynamic risk assessment according to claim 1, characterized in that: The multi-sensor fusion analysis specifically includes: Capacitive car cover electrode positioning: Determine the center point of the contact area (x c ,y c ); Millimeter-wave radar data fusion: Obtain the object's azimuth angle θ and the distance d from the object to the vehicle r , calculate the spatial offset: Δx=d r ×cosθ Δy=d r ×sinθ Wherein, Δx is the X-axis projection compensation of the millimeter radar wave, and Δy is the Y-axis projection compensation of the millimeter radar wave; Ultrasonic time domain calibration: Correct the distance data d based on the duration of the approach t u : Among them, d u ′ is the sliding average filter value of ultrasonic ranging, n is the size of the sliding window, d u,i is the distance between the object and the vehicle measured at the i-th sampling moment, v i is the instantaneous velocity of the object at the i-th sampling moment, Δt i is the time interval between adjacent sampling points.

5. The vehicle scratch damage location method based on dynamic risk assessment according to claim 1, characterized in that: The calculation formula for the coordinates of the damaged part in the three-dimensional coordinate visualization report is: Three-dimensional coordinate mapping: The weighted least squares method is used to generate the coordinates (x, y, z) of the injury site; b=[x c +Δx,y c +Δy,d u ′] T Among them, A is the sensor position matrix, A T is the transpose of A, W is the weight matrix, and b is the observation vector.

6. The vehicle scratch damage location method based on dynamic risk assessment according to claim 1, characterized in that: The visualization report includes: a three-dimensional reconstruction of the contact object's motion trajectory, a damage probability prediction value, and characteristic information of the suspected object.

7. A vehicle scratch warning system based on dynamic risk assessment, characterized in that: include: The perception layer uses multiple sensors to detect the contact distance between objects and the vehicle body in real time and establish a three-level risk contact identification model; The edge computing layer uses a real-time risk scoring algorithm to evaluate and calculate the risk of scratches; The interactive layer is used to push reports containing three-dimensional positioning maps and suspect information.

8. The vehicle scratch warning system based on dynamic risk assessment according to claim 7, characterized in that: The sensing layer includes capacitive sensing car cover, millimeter wave radar and ultrasonic sensor; Capacitive sensing car cover is arranged around the car, and the capacitive sensing car cover is composed of a distributed electrode array for real-time detection of the contact distance between the object and the car body; 4D millimeter-wave radar array distribution, used to obtain the relative speed of objects; Ultrasonic sensor, used to detect when an object is close to you.

9. The vehicle scratch warning system based on dynamic risk assessment according to claim 7, characterized in that: The edge computing layer uses a real-time risk scoring algorithm to evaluate and calculate the risk of scratches, specifically including: The dynamic risk assessment engine running on the Renesas R-Car V3H chip; Real-time execution risk formula: Risk = dv × t × k object ; Dynamic configuration of object risk coefficient, risk coefficient k for metal objects metal =1.5, the risk factor k of the organism biology =0.

8.

10. The vehicle scratch warning system based on dynamic risk assessment according to claim 7, characterized in that: The interactive layer pushes a report containing a three-dimensional positioning map and suspect information, specifically including: One-click generation of insurance reporting materials function module; Intelligent estimation prompt of repair costs; Directionally activate the image of the suspect captured by the camera.

Images