Intelligent anti-collision pad monitoring and alarming method, intelligent anti-collision pad and intelligent anti-collision pad monitoring and alarming system

Abstract

The invention relates to the technical field of traffic anti-collision devices, and provides an intelligent anti-collision pad monitoring and alarming method, an intelligent anti-collision pad and a system. The monitoring and alarming method comprises the steps of determining collision time corresponding to the collision and collision detection data corresponding to the collision time under the condition that the collision of the intelligent anti-collision pad is monitored based on detection data collected by a collision detection module; determining collision video data corresponding to the collision from monitoring data acquired by a video monitoring module based on the collision time; determining a collision degree corresponding to the collision in combination with the collision video data and the collision detection data; and generating collision alarm information based on the collision degree, and controlling an alarm module to output the collision alarm information. By adopting the technical scheme, the collision degree can be comprehensively determined based on the collision video data and the collision detection data corresponding to the collision process, so that the accuracy of the determined collision degree can be improved, and the accuracy of collision alarm of the crash pad can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of traffic anti-collision alarm devices, and in particular to an intelligent anti-collision pad monitoring alarm method, an intelligent anti-collision pad and a system. Background Art

[0002] The crash cushion used on the road is an important safety facility. Its main function is to absorb the energy of a vehicle collision, so that the vehicle can be stopped safely to avoid serious injuries to the occupants.

[0003] In the related art, there is provided an intrusion detection and warning system including a traffic cone and an alarm terminal, which can output an alarm message through the alarm terminal in the event of a collision with the traffic cone.

[0004] However, unlike traffic cones, since crash pads will deform during a collision, it is impossible to accurately judge the degree of collision using traditional methods, which in turn affects the accuracy of the collision alarm information. Summary of the invention

[0005] In order to help improve the accuracy of anti-collision pad collision alarm information, the present application provides an intelligent anti-collision pad monitoring alarm method, system and intelligent anti-collision pad.

[0006] In the first aspect, the present application provides an intelligent anti-collision pad monitoring and alarm method, which adopts the following technical solution: A smart anti-collision pad monitoring and alarm method is used in a controller of a smart anti-collision pad, wherein the smart anti-collision pad further comprises a collision detection module, a video monitoring module and an alarm module connected to the controller signal, wherein the collision detection module is used to detect the deformation of the anti-collision pad body, and the video monitoring module is used to collect image information of the area where the anti-collision pad body is located. The method comprises: Monitoring the detection data collected by the collision detection module; In the case where a collision occurs with the smart anti-collision pad based on the detection data, determining a collision time corresponding to the collision and collision detection data corresponding to the collision time; Determining collision video data corresponding to the collision from the monitoring data collected by the video monitoring module based on the collision time; Determine the collision degree corresponding to the current collision by combining the collision video data and the collision detection data; Collision alarm information is generated based on the collision degree, and the alarm module is controlled to output the collision alarm information.

[0007] By adopting the above technical solution, when a collision of the smart anti-collision pad is detected based on the detection data, the degree of collision can be comprehensively determined based on the collision video data and collision detection data corresponding to the collision process. This can help improve the accuracy of the determined collision degree, and further help improve the accuracy of the anti-collision pad collision alarm.

[0008] Optionally, the collision video data includes post-collision video data, and the collision video data and the detection data determine the collision degree corresponding to the current collision, including: Determining a vehicle deformation degree of the target vehicle corresponding to the collision based on the post-collision video data; Determining the degree of deformation of the crash pad corresponding to the collision based on the collision detection data; The collision degree is determined based on the vehicle deformation degree and the crash pad deformation degree.

[0009] By adopting the above technical solution, the vehicle deformation and anti-collision pad deformation caused by the collision can be fully considered in the process of collision degree analysis, which can help avoid the problem of inaccurate collision degree determined by a single deformation, and further help improve the accuracy of the final determined collision degree.

[0010] Optionally, the collision video data further includes pre-collision video data, and before determining the collision degree based on the deformation degree of the vehicle and the deformation degree of the crash pad, further includes: determining vehicle parameters of the target vehicle based on the pre-collision video data; Determining a prediction confidence corresponding to the degree of deformation of the vehicle based on the vehicle parameters; Predicting the deformation degree of the crash pad body based on the prediction confidence and the deformation degree of the vehicle to obtain a predicted deformation range; Determining whether the deformation degree of the crash pad is within the predicted deformation range; In a case where the degree of deformation of the crash pad is within the predicted deformation range, the degree of collision is determined based on the degree of deformation of the vehicle and the degree of deformation of the crash pad.

[0011] By adopting the above technical solution, before determining the collision degree in combination with the deformation degree of the crash pad, the deformation degree of the vehicle can be used to verify whether there is an error in the deformation degree of the crash pad, which can help ensure the accuracy of the degree finally determined.

[0012] Optionally, after determining whether the collision deformation degree is within the predicted deformation range, the method further includes: When the degree of deformation of the crash pad is not within the predicted deformation range, determining a predicted deformation amount of the crash pad body based on the post-collision video data; determining whether the degree of deformation of the crash pad matches the predicted deformation amount; In a case where the crash pad deformation degree matches the predicted deformation amount, the collision degree is determined based on the crash pad deformation degree and the predicted deformation amount.

[0013] By adopting the above technical scheme, the reason why the deformation degree of the anti-collision pad does not match the deformation degree of the vehicle can be determined by verifying the deformation degree of the anti-collision pad, which can help avoid the influence of unreasonable vehicle deformation on the determination of the collision degree, and further help ensure the accuracy of the final determined collision degree.

[0014] Optionally, after determining whether the deformation degree of the crash pad matches the predicted deformation amount, the method further includes: In the case where the collision deformation degree does not match the predicted deformation amount, determining whether the predicted deformation amount matches the predicted deformation range; In a case where the predicted deformation amount matches the predicted deformation range, the collision degree is determined based on the predicted deformation amount and the vehicle deformation degree.

[0015] By adopting the above technical scheme, when the collision deformation degree does not match the predicted deformation amount, the collision degree can be further determined based on the matching relationship between the predicted deformation amount and the predicted deformation range to determine whether the collision degree needs to be determined in combination with the vehicle deformation degree. This can help to accurately judge the reference value of the vehicle deformation degree in the collision degree determination process, and further help to further improve the accuracy of the collision degree while ensuring that the determined collision degree has a high accuracy.

[0016] Optionally, the monitoring of the detection data collected by the collision detection module includes: Monitoring whether the detection data fluctuates; In the case where the detection data is monitored to fluctuate, determining waveform features corresponding to the detection data; Determine whether the smart crash pad collides based on the waveform feature.

[0017] By adopting the above technical solution, when the detection data is monitored to fluctuate, it is possible to further determine whether the smart collision pad has collided based on the fluctuation of the detection data for verification, thereby helping to reduce the impact of factors other than collision on collision monitoring, and further helping to improve the accuracy of collision monitoring.

[0018] Optionally, the collision detection module includes a first detection component and a second detection component, the detection data includes first detection data corresponding to the first detection component and second detection data corresponding to the second detection component, and when fluctuations in the detection data are detected, determining waveform characteristics of the detection data includes: When the first detection data and / or the second detection data are detected to fluctuate, respectively determining a first waveform feature of the first detection data and a second waveform feature of the second detection data; The step of determining whether the smart anti-collision pad collides based on the waveform characteristics comprises: determining a collision probability based on the first waveform feature; determining whether the collision probability is less than a collision probability threshold; When the collision probability is less than the collision probability threshold, determining whether there is a risk feature in the second waveform feature; In the case where the risk feature exists, performing an abnormality check on the risk feature based on the first waveform feature; When the abnormality check result indicates that the risk feature is abnormal, it is determined that the smart anti-collision pad has not collided.

[0019] By adopting the above technical solution, it is possible to comprehensively determine whether a collision occurs with the smart anti-collision pad by combining the waveform characteristics corresponding to the detection data collected by different detection components, which can help improve the accuracy of collision monitoring.

[0020] Optionally, after performing abnormality check on the risk feature based on the first waveform feature, the method further includes: When the abnormality check result indicates that the risk feature does not have an abnormality, determining a collision feature in the first waveform feature; Performing a collision check on the collision feature based on the second waveform feature; When the collision check result indicates that the collision feature has a collision, it is determined that the smart anti-collision pad has a collision.

[0021] By adopting the above-mentioned technical scheme, when the abnormality check result indicates that there is no abnormality in the risk feature, the collision feature in the first waveform feature can be further checked based on the second waveform feature to finally determine whether the smart anti-collision pad has collided. In this way, the first waveform feature and the second waveform feature can be fully combined to judge whether the smart anti-collision pad has collided, which can help to further improve the accuracy of collision monitoring.

[0022] In the second aspect, the present application provides a smart anti-collision pad, which adopts the following technical solution: An intelligent crash pad, comprising a controller and a collision detection module, a video monitoring module and an alarm module connected to the controller by signal, wherein the collision monitoring module is used to detect the deformation of the crash pad body, and the video monitoring module is used to collect image information of the area where the crash pad body is located; The controller is used to execute any one of the intelligent anti-collision pad monitoring and alarm methods provided in the first aspect, and send collision alarm information to the background server.

[0023] In a third aspect, the present application provides an intelligent anti-collision pad alarm system, which adopts the following technical solutions: An intelligent anti-collision pad alarm system, the system comprising the intelligent anti-collision pad provided by the second aspect, and a background server connected to the intelligent anti-collision pad in communication; The intelligent anti-collision pad is used to generate collision warning information and send it to the background server; The background server is used to process the collision warning information.

[0024] In summary, the present application includes at least one of the following beneficial technical effects: 1. When the smart anti-collision pad is detected to have collided based on the detection data, the degree of collision can be determined based on the collision video data and collision detection data corresponding to the collision process, which can help improve the accuracy of the determined collision degree, and further help improve the accuracy of the anti-collision pad collision alarm; 2. In the process of collision degree analysis, the deformation of the vehicle and the anti-collision pad caused by the collision can be fully considered, which can help avoid the problem of inaccurate collision degree determined by using a single deformation, and further help improve the accuracy of the final determined collision degree. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a structural schematic diagram of a smart anti-collision pad provided in an embodiment of the present application; Figure 2a and Figure 2b is a schematic diagram of a crash pad deployment method provided in an embodiment of the present application; Figure 3 It is a structural schematic diagram of an intelligent anti-collision pad system provided in an embodiment of the present application; Figure 4 It is a flowchart of a smart anti-collision pad monitoring and alarm method provided in an embodiment of the present application; Figure 5 It is a flowchart of a method for determining the degree of collision provided in an embodiment of the present application; Figure 6 It is a flow chart of a method for verifying the deformation degree of a crash pad provided in an embodiment of the present application; Figure 7 It is a flow chart of a detection data monitoring method provided in an embodiment of the present application; Figure 8 It is a flow chart of another intelligent anti-collision pad monitoring and alarm method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figures 1 to 8 It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0027] The present application embodiment discloses a smart anti-collision pad, referring to Figure 1 The smart anti-collision pad 110 includes a controller 111 , and a collision detection module 112 , a video monitoring module 113 and an alarm module 114 which are connected to the controller 111 by signals.

[0028] The collision detection module 112 is disposed on the anti-collision pad body, and the controller 111, the video monitoring module 113 and the alarm component 140 can be disposed on the anti-collision pad body, or can be disposed separately from the anti-collision pad body. Figure 2a and Figure 2b The anti-collision pad body is connected to the rear of the transport vehicle so as to be moved to a predetermined position driven by the transport vehicle. At this time, the controller 111, the video monitoring module 113 and the alarm component 140 can be set on the transport vehicle, which can help avoid the impact of the anti-collision pad body on it during the collision process, and further help ensure that the controller 111, the video monitoring module 113 and the alarm module 114 can work normally during the collision process, thereby ensuring an effective collision alarm.

[0029] The collision detection module 112 is used to detect the deformation of the anti-collision pad body. Specifically, the collision detection module 112 includes at least one detection component, and different detection components have different types and / or installation positions. In actual implementation, the detection component can be implemented based on a sensor that can sense deformation, such as a piezoelectric sensor, a Hall sensor, a linear variable differential transformer (LVDT), etc.

[0030] The video monitoring module 113 is used to collect image information of the area where the crash pad body is located. Specifically, the video monitoring module 113 includes at least one camera, and different cameras are installed in different positions. In one example, the video monitoring module 113 includes two wide-angle cameras, which are respectively located on both sides of the crash pad body and are used to collect image information of the crash pad body and its surrounding environment. In another example, the video monitoring module includes cameras in four directions: front, back, left, and right, which are respectively used to collect information in the four directions.

[0031] The alarm module 114 is used to output collision alarm information. In one example, the alarm module 114 includes an alarm device, such as an alarm light, a horn, a display screen, etc. In another example, the alarm module 114 includes a signal transmitting device, such as a network communication module (4G module), and the alarm device 140 can send an alarm message to the background server.

[0032] The controller 111 is used to obtain the detection data collected by the collision detection module 112 and the video data collected by the video monitoring module 113, and judge the collision and the degree of collision based on the detection data and video data, and then output the collision prompt information through the alarm module 114 to prompt the collision.

[0033] In actual implementation, the smart crash pad 110 may also include a positioning module to locate the location where the collision occurs, which can help improve the efficiency of accident rescue.

[0034] The present application also provides an intelligent anti-collision pad alarm system, referring to Figure 3 , the intelligent anti-collision pad alarm system includes the intelligent anti-collision pad 110 provided in the above embodiment and a background server 120 which is in communication with the intelligent anti-collision pad; The smart anti-collision pad 110 is used to generate collision warning information and send it to the background server 120. In one example, the smart anti-collision pad 110 includes a communication module (such as a 4G communication module) connected to the controller signal to establish a communication connection with the background server 120.

[0035] The backend server 120 is used to process the collision warning information sent by the smart crash pad 110. Specifically, the backend server 120 can automatically push prompts to the crash pad manufacturer, local emergency management department and other relevant departments (for example, through text messages, in-app messages), so as to quickly respond to the collision warning information. In addition, the backend server 120 can also save the content of the collision warning information for subsequent investigation.

[0036] The present application also discloses a smart anti-collision pad monitoring and alarm method, which is used in the controller of the smart anti-collision pad provided in the above embodiment. Figure 4 , the intelligent anti-collision pad monitoring alarm method includes the following steps: Step 201 , monitoring the detection data collected by the collision detection module.

[0037] Specifically, since the collision monitoring module can collect the deformation of the anti-collision pad body, and after a collision with a vehicle, the anti-collision pad body will be deformed and cause the detection data to change, it is possible to monitor whether the smart anti-collision pad has collided by monitoring the value of the detection data and / or the change in the value of the detection data.

[0038] In one example, the collision detection module includes a detection component arranged at the rear of the anti-collision pad body. In a natural state, the detection data collected by the detection component is constant. When a vehicle hits a detection area where the detection component is located, the detection area will be deformed, thereby causing the detection data corresponding to the detection component to change.

[0039] Step 202 : when a collision of the smart anti-collision pad is detected based on the detection data, a collision time corresponding to the collision and collision detection data corresponding to the collision time are determined.

[0040] The collision time refers to the time when the collision occurs. Specifically, the collision time can be the time when the collision starts and / or ends, or can also be the time period during which the collision lasts.

[0041] The collision detection data refers to the detection data collected during the collision. Specifically, the collision detection data may be detection data within a preset time period from the start time of the collision, or detection data within a preset time period before the end time of the collision, or detection data within a duration period of the collision.

[0042] In one example, when the value of the monitored detection data is greater than a preset threshold, it is determined that the smart anti-collision pad has collided.

[0043] In another example, it is determined that the smart crash pad has collided when the fluctuation amplitude of the monitored data is greater than a preset amplitude threshold.

[0044] Step 203: Determine collision video data corresponding to the collision from monitoring data collected by the video monitoring module based on the collision time.

[0045] In one example, the video monitoring module may continuously collect images. At this time, after the collision time is determined, the collision video data corresponding to the collision may be intercepted from the monitoring data based on the collision time.

[0046] In another example, the video monitoring module maintains a low power consumption mode under normal circumstances, and starts recording to collect monitoring data once receiving an activation instruction sent by the controller. The activation instruction may be generated by the controller when the smart anti-collision pad is detected to have collided based on the detection data, or may be generated by the controller when it is determined that there is a risk of collision, for example, when the controller is determined to be approaching a fast vehicle.

[0047] Step 204 , determining the collision degree corresponding to the current collision by combining the collision video data and the collision detection data.

[0048] The collision degree is used to indicate the damage of the smart crash pad and the vehicle during the collision. In actual implementation, the collision degree can be expressed in a quantifiable manner such as a value or a grade.

[0049] Specifically, during the research process, it was found that the judgment of the degree of collision includes many influencing factors, such as: collision speed, collision duration, collision intensity, vehicle deformation degree, anti-collision pad deformation degree, etc., and for some influencing factors, it is impossible to accurately judge only by collision detection data. Based on this, in this embodiment, the collision degree is comprehensively judged by combining collision video data and collision detection data to accurately judge different influencing factors.

[0050] In one example, the collision degree corresponding to the current collision is determined in combination with the collision video data and the collision detection data, including: determining the collision speed based on the collision video data; determining the collision intensity based on the collision detection data; and determining the collision degree based on the collision speed and the collision intensity.

[0051] In another example, the collision degree corresponding to the current collision is determined in combination with the collision video data and the collision detection data, including: comprehensively determining the collision duration based on the collision video data and the collision detection data; and determining the collision degree in combination with the collision duration.

[0052] In actual implementation, the collision degree may be synthesized by combining the methods in the above two examples, or the collision degree may be determined based on other methods.

[0053] Step 205: generate collision alarm information based on the collision degree, and control the alarm module to output the collision alarm information.

[0054] Optionally, the collision warning information includes the degree of collision, which can facilitate determining the urgency of collision handling based on the warning information, thereby helping to improve the efficiency of collision handling.

[0055] In one example, the alarm module includes an alarm component. Accordingly, controlling the alarm module to output collision alarm information includes: controlling the alarm light to flash, controlling the speaker to broadcast an alarm tone, and controlling the display component to display an alarm logo.

[0056] In another example, the alarm module includes a signal transmitting device, and accordingly, controlling the alarm module to output collision alarm information includes: controlling the signal transmitting device to send the collision alarm information to a background server, so that the background server processes the collision alarm.

[0057] In actual implementation, the collision warning information may also include the location information of the smart anti-collision pad, which can help improve the efficiency of accident rescue.

[0058] The implementation principle of a smart crash pad monitoring and alarm method in an embodiment of the present application is as follows: monitoring the detection data collected by the collision detection module; in the case where a collision of the smart crash pad is detected based on the detection data, determining the collision time corresponding to this collision and the collision detection data corresponding to the collision time; determining the collision video data corresponding to this collision from the monitoring data collected by the video monitoring module based on the collision time; determining the collision degree corresponding to this collision in combination with the collision video data and the collision detection data; generating collision alarm information based on the collision degree, and controlling the alarm module to output the collision alarm information. By adopting the above technical solution, in the case where a collision of the smart crash pad is detected based on the detection data, the collision degree can be comprehensively determined based on the collision video data corresponding to this collision process and the collision detection data, which can help improve the accuracy of the determined collision degree, and further help improve the accuracy of the crash pad collision alarm.

[0059] In some embodiments, the collision video data includes post-collision video data, referring to Figure 5 The above step 204, combining the collision video data and the collision detection data to determine the collision degree corresponding to the current collision, specifically includes the following steps: Step 301 : determining the vehicle deformation degree of the target vehicle corresponding to the collision based on the post-collision video data.

[0060] The degree of vehicle deformation is used to indicate the damage to the target vehicle during the collision. Specifically, although the crash pad can offset most of the impact generated during the collision, some of the impact will still act on the target vehicle and may cause the target vehicle to deform. Therefore, it is necessary to consider the degree of vehicle deformation during the collision degree analysis.

[0061] Optionally, determining the degree of deformation of the target vehicle corresponding to the collision based on the post-collision video data includes: extracting a post-collision image from the post-collision video data; intercepting a vehicle image corresponding to the target vehicle from the post-collision image; and determining the degree of deformation of the vehicle based on the vehicle image. Specifically, feature recognition can be performed on the vehicle image to determine the missing features in the vehicle image, such as the front bumper, the hood, the front windshield, etc., and then determining the degree of deformation of the vehicle based on the missing features in the vehicle image.

[0062] In one example, when the vehicle image indicates that the hood of the target vehicle is mostly intact, it can be determined that the vehicle is slightly deformed; when the vehicle image indicates that the hood of the target vehicle is mostly damaged and the front windshield is intact or not seriously damaged, it can be determined that the vehicle is moderately deformed; when the vehicle image indicates that the hood of the target vehicle is missing and the front windshield is also damaged or the front windshield is severely damaged, it can be determined that the vehicle is severely deformed.

[0063] Furthermore, when the deformation level of the vehicle is preliminarily determined, the deformation degree of the vehicle can be further scored within the scoring range corresponding to the deformation level based on the scoring method corresponding to the deformation level, and finally the deformation score can be obtained. Specifically, the value range of the deformation score can be divided based on the preset damage level to obtain the scoring ranges corresponding to different deformation levels, for example: the value range of the deformation score is 1-10, and the score is positively correlated with the deformation level. At this time, the scoring range corresponding to mild deformation is 1-3, the scoring range corresponding to moderate deformation is 4-7, and the scoring range corresponding to severe deformation is 8-10.

[0064] Among them, there are differences in the scoring methods corresponding to different deformation levels. Specifically, the reference features of the scoring methods corresponding to different scoring levels are different. In one example, the reference features corresponding to mild deformation include: the damage to the hood; the reference features corresponding to moderate deformation include: the damage to the hood and whether the front windshield is damaged; the reference features corresponding to severe deformation include: the damage to the front windshield.

[0065] In actual implementation, the pre-collision image can be further extracted from the post-collision video data, so that the pre-collision image and the post-collision image can be compared and analyzed in the process of judging the degree of vehicle deformation, which can help improve the accuracy of the final determination of the degree of vehicle damage.

[0066] Step 302: determining the degree of deformation of the crash pad corresponding to the current collision based on the collision detection data.

[0067] Optionally, the degree of deformation of the crash pad corresponding to the collision is determined based on the collision detection data, including: determining the degree of deformation of the crash pad based on the peak value of the collision detection data. Specifically, since the crash pad is elastic, it will generally deform significantly first and then rebound slightly during the collision process. Accordingly, the collision detection data will also fluctuate. The peak position of the collision detection data generally corresponds to the position where the degree of deformation of the crash pad is the largest. Therefore, the maximum degree of deformation of the crash pad body can be determined based on the peak position, which can help ensure the accuracy of the determined degree of deformation of the crash pad.

[0068] The peak value is greater than the corresponding relationship with the deformation degree of the anti-collision pad, which is preset. In one example, the value range of the collision detection data is divided according to the deformation level to obtain the numerical intervals corresponding to different deformation levels. At this time, the deformation level can be determined according to the numerical interval to which the peak value belongs. In another example, the corresponding relationship between the collision detection data and the deformation score can be obtained based on experiments. Generally speaking, the collision detection data is positively correlated with the deformation score.

[0069] Furthermore, in the case where the collision detection module includes at least two detection components, the collision detection data includes collision detection data collected by different detection components. At this time, the deformation degree of the anti-collision pad corresponding to the anti-collision pad is determined based on the peak value of the collision detection data, including: superimposing the collision detection data collected by different detection components according to a preset superposition rule to obtain superimposed detection data; determining the deformation degree of the anti-collision pad based on the peak value of the superimposed detection data.

[0070] Among them, the superposition rules can be set based on the relative positions of each detection component. Specifically, since the deformation detection ranges corresponding to the detection components set at different positions are different, and the deformation detection ranges corresponding to different detection components may overlap, and the deformations at different positions may also affect each other, it is necessary to pre-set the superposition rules based on the relative positions of each sensing component to achieve accurate superposition of the detection data collected by different detection components.

[0071] In other embodiments, when the collision detection module includes at least two detection components, the peaks corresponding to the collision detection data collected by different detection components can also be determined separately, and the peaks corresponding to the detection data collected by different detection components can be comprehensively analyzed to finally obtain the degree of deformation of the crash pad. For example, the collision process can be simulated based on the timing of the peaks corresponding to the detection data collected by different detection components, so that the deformation process of the crash pad can be predicted based on the collision process, and then the deformation degree of the crash pad can be predicted based on the deformation process of the crash pad.

[0072] In actual implementation, the deformation degree of the crash pad may be further determined in combination with factors such as the collision duration indicated by the collision detection data to improve the accuracy of the deformation degree of the crash pad that is finally determined.

[0073] Step 303: determining the degree of collision based on the degree of deformation of the vehicle and the degree of deformation of the crash pad.

[0074] In one example, the deformation degree is represented by a deformation grade, and accordingly, the collision degree is determined based on the deformation degree of the vehicle and the deformation degree of the crash pad, including: determining the more serious one of the deformation degree of the vehicle and the deformation degree of the crash pad as the collision degree. For example: if the deformation degree of the vehicle is moderate and the deformation degree of the crash pad is severe, then the collision degree is severe.

[0075] In another example, the deformation degree is represented by a deformation score, and the collision degree is determined based on the deformation degree of the vehicle and the deformation degree of the crash pad, including: determining whether the difference between the deformation degree of the vehicle and the deformation degree of the crash pad is greater than a preset score difference threshold; if so, determining the larger one of the deformation degree of the vehicle and the deformation degree of the crash pad as the collision degree; if not, determining the average of the deformation degree of the vehicle and the deformation degree of the crash pad as the collision degree. For example: if the score difference threshold is 3, the deformation degree of the vehicle is 5, and the deformation degree of the crash pad is 6, then the collision degree is the average of the deformation degree of the vehicle and the deformation degree of the crash pad, that is, 5.5.

[0076] In the above-mentioned embodiment, in the process of determining the degree of collision, the degree of deformation of the vehicle can be first determined based on the post-collision video data, and the degree of deformation of the anti-collision pad can be determined based on the collision detection data. The degree of collision is then comprehensively analyzed based on the degree of deformation of the vehicle and the degree of deformation of the anti-collision pad. In this way, the deformation of the vehicle and the deformation of the anti-collision pad caused by the collision can be fully considered in the process of analyzing the degree of collision, which can help avoid the problem of inaccurate collision degree determined by a single deformation, and further help improve the accuracy of the final determined collision degree.

[0077] At the same time, since the degree of vehicle deformation is obtained based on the analysis of post-collision video data without relying on the sensor equipment installed on the vehicle, this can reduce the difficulty and cost of analyzing the degree of vehicle deformation, making it easier to put this method into practical application.

[0078] Based on the above implementation, further, the collision video data also includes pre-collision video data, referring to Figure 6 Before determining the collision degree based on the deformation degree of the vehicle and the deformation degree of the crash pad in step 303, the above-mentioned steps further include: Step 401 : determining vehicle parameters of the target vehicle based on pre-collision video data.

[0079] The pre-collision video data is collected before the target vehicle is deformed. Specifically, the pre-collision video data can be collected before the target vehicle contacts the crash pad body, such as data 30 seconds before the collision, or collected at the moment of contact with the crash pad body.

[0080] In one example, the vehicle parameters include vehicle models, which may include sedans, SUVs, vans, trucks, etc. Specifically, since the structures of vehicles of different models may be different, and the structure of the vehicle will affect the deformation after the collision, that is, for the same collision degree, the deformation effects produced by different types of vehicles may be different, for example, the deformation degree of a sedan may be greater than that of an SUV, so in the process of determining the collision degree based on the vehicle deformation degree, it is necessary to consider the vehicle model.

[0081] In another example, the vehicle parameters include the vehicle speed. Specifically, it was found during the test that the vehicle speed will affect the collision process, which may cause different deformations of the vehicle, that is, different factors may cause the vehicle to deform at different vehicle speeds. Therefore, in the process of determining the degree of collision based on the degree of vehicle deformation, it is necessary to consider the vehicle speed.

[0082] In actual implementation, vehicle parameters may also include parameters related to the actual condition of the vehicle, such as the length of the vehicle's hood, the width of the vehicle, whether the vehicle is equipped with a front bumper, etc. This can help to accurately judge the actual condition of the vehicle.

[0083] Step 402: Determine the prediction confidence corresponding to the degree of vehicle deformation based on vehicle parameters.

[0084] Among them, the prediction confidence is used to indicate the correlation between the degree of vehicle deformation and the degree of collision. Specifically, since the correspondence between the degree of vehicle deformation and the degree of collision may change dynamically with factors such as vehicle type and vehicle speed, in order to help improve the accuracy of the final predicted degree of collision, it is necessary to determine the prediction confidence based on vehicle parameters and take it into account in the subsequent collision degree prediction process.

[0085] Step 403 , predicting the deformation degree of the crash pad body based on the prediction confidence and the deformation degree of the vehicle to obtain a predicted deformation range.

[0086] Specifically, unlike the deformation degree of the vehicle, the structure and material of the anti-collision pad body are predetermined, so the correspondence between the deformation degree of the anti-collision pad and the collision degree is roughly fixed. The range of the collision degree can be roughly determined based on the prediction confidence and the deformation degree of the vehicle. Therefore, the deformation degree of the anti-collision pad body can be predicted based on the prediction confidence and the deformation degree of the vehicle.

[0087] Step 404 , determining whether the deformation degree of the crash pad is within the predicted deformation range.

[0088] Specifically, since the deformation degree of the anti-collision pad is determined based on the collision detection data collected by the collision detection module, and the collision module may be displaced and deformed unreasonably due to the force during the collision, which may cause errors in the collision detection data, thereby affecting the accuracy of the deformation degree of the anti-collision pad. Based on this, in this embodiment, the accuracy of the deformation degree of the anti-collision pad is verified by predicting the deformation range to make an error judgment on the deformation degree of the anti-collision pad.

[0089] Step 405 , when the deformation degree of the crash pad is within the predicted deformation range, determine the collision degree based on the deformation degree of the vehicle and the deformation degree of the crash pad, that is, execute step 303 .

[0090] Specifically, when the deformation degree of the crash pad is within the predicted deformation range, it means that the probability of error in the deformation degree of the crash pad is low. At this time, determining the collision degree based on the deformation degree of the vehicle and the deformation degree of the crash pad can help ensure the accuracy of the collision degree.

[0091] Optionally, when the deformation degree of the crash pad is not within the predicted deformation range, it means that the probability of an error in the deformation degree of the crash pad is high. At this time, it can be directly determined not to execute the step of determining the collision degree based on the deformation degree of the vehicle and the deformation degree of the crash pad, that is, step 303 is not executed, and the collision degree is determined based on other methods, such as: directly determining the collision degree based on the deformation degree of the vehicle; or it can also be further verified based on other methods to see if there is an error in the deformation degree of the crash pad.

[0092] In the above technical scheme, before determining the degree of collision based on the deformation degree of the vehicle and the deformation degree of the crash pad, the predicted deformation range of the crash pad body can be predicted based on the vehicle deformation degree of the target vehicle and its corresponding prediction confidence, and the degree of collision can be determined based on the deformation degree of the vehicle and the deformation degree of the crash pad only when the deformation degree of the crash pad is within the predicted deformation range. In this way, before determining the degree of collision based on the deformation degree of the crash pad, the deformation degree of the vehicle can be used to verify whether there is any error in the deformation degree of the crash pad, which can help ensure the accuracy of the degree finally determined.

[0093] At the same time, since the prediction confidence corresponding to the degree of vehicle deformation is calculated based on vehicle parameters, and the vehicle parameters are determined based on pre-collision video data, there is no need to rely on sensor equipment installed on the vehicle. This can reduce the difficulty and cost of anti-collision pad deformation verification and facilitate the practical application of this method.

[0094] Based on the above implementation mode, further referring to Figure 6, step 404, after determining whether the deformation degree of the crash pad is within the predicted deformation range, further comprising: Step 406 : when the deformation degree of the crash pad is not within the predicted deformation range, determine the predicted deformation amount of the crash pad body based on the post-collision video data.

[0095] The predicted deformation refers to the deformation of the crash pad body identified based on the post-collision video data. In one example, the predicted deformation may be the maximum deformation of the crash pad body during the collision. In this case, it is necessary to extract multiple frames of images from the post-collision video at intervals to analyze the deformation of the crash pad body to determine the maximum deformation. In actual implementation, the predicted deformation may also be the deformation of the crash pad body after the collision.

[0096] In one example, the predicted deformation variable is represented by the difference between the area of ​​the crash pad body after the collision and the area in the natural state. The area of ​​the crash pad body after the collision is estimated by the post-collision video data. In one instance, based on the number of pixels occupied by the crash pad in the post-collision video data and the area of ​​the cavity crash pad body after the collision, the corresponding relationship between the number of pixels and the area is preset. For example: based on the video data collected by the video monitoring module during the non-collision process, the number of pixels occupied by the crash pad body in the natural state is determined, and then the corresponding relationship between the number of pixels and the area is determined based on the area of ​​the crash pad body and the number of pixels occupied by it in the natural state.

[0097] Step 407 , determining whether the deformation degree of the crash pad matches the predicted deformation amount.

[0098] The corresponding relationship between the deformation amount and the deformation degree is preset.

[0099] In one example, the deformation degree of the crash pad is represented by a deformation grade. Accordingly, determining whether the deformation degree of the crash pad matches the predicted deformation amount includes: determining whether the predicted deformation amount belongs to a deformation amount interval corresponding to the deformation grade of the crash pad; if so, determining that the deformation degree of the crash pad matches the predicted deformation amount; if not, determining that the deformation degree of the crash pad does not match the predicted deformation amount.

[0100] In another example, the deformation degree of the crash pad is represented by a deformation score. Accordingly, determining whether the deformation degree of the crash pad matches the predicted deformation variable includes: determining whether the difference between the deformation score corresponding to the predicted deformation variable and the deformation degree of the crash pad is less than a preset difference threshold; if so, determining that the deformation degree of the crash pad matches the predicted deformation variable; if not, determining that the deformation degree of the crash pad does not match the predicted deformation variable.

[0101] Step 408 : When the deformation degree of the crash pad matches the predicted deformation amount, determine the collision degree based on the deformation degree of the crash pad and the predicted deformation amount.

[0102] Specifically, the manner of determining the collision degree based on the deformation degree of the crash pad and the expected deformation amount can be similar to the manner of determining the collision degree based on the deformation degree of the vehicle and the deformation degree of the crash pad in step 303, which will not be described in detail here.

[0103] In another embodiment, when the deformation degree of the anti-collision pad matches the predicted deformation amount, the reference time point can be first determined based on the collision detection data, and a reference image corresponding to the reference time point can be captured from the post-collision image data; the reference deformation amount of the anti-collision pad body can be determined based on the reference image corresponding to the reference time point, so as to finally determine the collision degree based on the reference deformation amount.

[0104] The reference time point may be a time point corresponding to a peak value of the collision detection data. When the detection data includes detection data collected by multiple detection components, the time point corresponding to the detection data with the highest peak value may be determined as the reference time point, or the detection data collected by different detection components may be superimposed, and the time point corresponding to the peak value of the superimposed detection data may be determined as the reference time point.

[0105] In one example, the method of determining the corrected deformation amount of the crash pad body based on the reference image corresponding to the reference time point may be similar to the method of determining the predicted deformation amount of the crash pad body based on the post-collision video data in step 406 .

[0106] In another example, the deformation amount is represented by a deformation score. At this time, the collision degree is finally determined based on the reference deformation amount, including: determining the modified deformation amount based on the numerical value corresponding to the reference time point in the collision detection data; and determining the collision degree based on the modified deformation amount and the reference deformation amount.

[0107] The method of determining the collision degree based on the correction deformation amount and the reference deformation amount can be analogous to the specific implementation method corresponding to the above step 303 of determining the collision degree based on the deformation degree of the vehicle and the deformation degree of the anti-collision pad, which will not be repeated here.

[0108] Optionally, when the collision deformation degree does not match the predicted deformation amount, the collision degree may be determined directly based on the predicted deformation amount and the vehicle deformation degree.

[0109] Specifically, since the correspondence between the deformation amount and the deformation degree is pre-set, the method of determining the collision degree based on the predicted deformation amount and the vehicle deformation degree can be analogous to the specific implementation method of determining the collision degree based on the vehicle deformation degree and the anti-collision pad deformation degree in the above step 303, which will not be repeated here.

[0110] In the above-mentioned embodiment, when the deformation degree of the crash pad is not within the predicted deformation range, the predicted deformation amount of the crash pad body will be determined based on the post-collision video data, and when the deformation degree of the crash pad matches the predicted deformation amount, the collision degree will be determined directly based on the deformation degree of the crash pad and the predicted deformation amount. In this way, the reason for the mismatch between the deformation degree of the crash pad and the deformation degree of the vehicle can be determined by verifying the deformation degree of the crash pad, which can help avoid the influence of unreasonable vehicle deformation on the determination of the collision degree, and further help ensure the accuracy of the finally determined collision degree.

[0111] Based on the above implementation mode, further referring to Figure 6 , step 407, after determining whether the deformation degree of the crash pad matches the predicted deformation amount, the following steps are also included: Step 409: When the collision deformation degree does not match the predicted deformation amount, determine whether the predicted deformation amount matches the predicted deformation range.

[0112] Specifically, since the correspondence between the deformation amount and the deformation degree is preset, the method of determining whether the predicted deformation amount matches the predicted deformation range can be similar to the specific implementation method of determining whether the deformation degree of the anti-collision pad is within the predicted deformation range in step 404, which will not be repeated here.

[0113] Step 410 : When the predicted deformation amount matches the predicted deformation range, the collision degree is determined based on the predicted deformation amount and the deformation degree of the vehicle.

[0114] Optionally, when the predicted deformation amount does not match the predicted deformation range, the collision degree may be determined directly based on the predicted deformation amount.

[0115] In the above-mentioned implementation manner, since the collision degree can be further determined based on the matching relationship between the predicted deformation value and the predicted deformation range when the collision deformation degree does not match the predicted deformation value, whether the collision degree needs to be determined in combination with the vehicle deformation degree can be determined. This can help to accurately judge the reference value of the vehicle deformation degree in the collision degree determination process, and further help to further improve the accuracy of the collision degree while ensuring that the determined collision degree has a high accuracy.

[0116] In some embodiments, reference Figure 7 , step 201, monitoring the detection data collected by the collision detection module, including the following steps: Step 501, monitoring whether the detection data fluctuates.

[0117] Specifically, when other objects collide with the smart crash pad, the crash pad body may be deformed, and the detection data can reflect the deformation of the crash pad body, so collision detection can be performed based on whether the detection data fluctuates.

[0118] Step 502: when fluctuations in the detection data are detected, determine waveform features corresponding to the detection data.

[0119] The waveform characteristics may include factors related to the actual waveform, such as fluctuation amplitude, peak size, duration of fluctuation, and difference in detection data before and after fluctuation.

[0120] Step 503: determining whether the smart anti-collision pad has collided based on the waveform characteristics.

[0121] Specifically, it is found in actual use that in addition to collisions with vehicles, some special situations (such as flying objects hitting the anti-collision pad body) may also cause fluctuations in the detection data. In order to improve the accuracy of collision judgment, it is necessary to adopt a suitable method to distinguish the fluctuations caused by vehicle collisions from the fluctuations caused by other situations. In this embodiment, considering that the waveform characteristics of fluctuations caused by different situations may be different, in order to more accurately judge the collision of the smart anti-collision pad, the waveform characteristics are used to determine whether the anti-collision pad has collided.

[0122] In one example, determining whether a smart anti-collision pad has collided based on waveform characteristics includes: determining whether the fluctuation amplitude is greater than a preset amplitude threshold; if the fluctuation amplitude is greater than the preset amplitude threshold, further determining whether the fluctuation duration is greater than a preset duration threshold; if the fluctuation duration is greater than the preset fluctuation duration threshold, determining that the smart anti-collision pad has collided.

[0123] Furthermore, when the duration of the fluctuation is less than the preset fluctuation duration threshold, it is further determined whether the difference in the detection data before and after the fluctuation is greater than the preset difference threshold; when the difference in the detection data before and after the fluctuation is greater than the preset difference threshold, it is determined that the smart anti-collision pad has collided.

[0124] In the above implementation, since it is possible to further determine whether the smart collision pad has collided based on the fluctuation of the detection data when fluctuations in the detection data are monitored, this can help reduce the impact of factors other than collision on collision monitoring, and thus help improve the accuracy of collision monitoring.

[0125] Based on the above technical solution, further, the collision detection module includes a first detection component and a second detection component, and correspondingly, the detection data includes first detection data corresponding to the first detection component and second detection data corresponding to the second detection component.

[0126] In one example, the first detection component and the second detection component are divided based on the installation direction of the detection component, for example: the first detection component is installed on the front side of the anti-collision pad body, that is, the side facing the driving direction of the vehicle during use; the second detection component is installed on the side of the anti-collision pad body, that is, the side perpendicular to the driving direction of the vehicle during use. At this time, the first detection component is the main detection component, and the second detection component is used to assist the first detection component to work.

[0127] In actual implementation, the number of first detection components and second detection components can be one or more. For example, when the number of first detection components is more than two, the detection data collected by each first detection component is fused to obtain first detection data.

[0128] refer to Figure 8 Step 502, when the detected data fluctuates, determining the waveform characteristics of the detected data includes the following steps: Step 601 : when it is detected that the first detection data and / or the second detection data fluctuate, a first waveform feature of the first detection data and a second waveform feature of the second detection data are determined respectively.

[0129] Specifically, when it is monitored that any of the first detection data and the second detection data fluctuates, the first waveform feature and the second waveform feature are respectively determined.

[0130] Accordingly, step 503, determining whether the smart anti-collision pad has collided based on the waveform characteristics, includes the following steps: Step 602: determine the collision probability based on the first waveform feature.

[0131] Optionally, determining the collision probability based on the first waveform feature includes: determining the collision probability based on a difference between the first waveform feature and a corresponding preset feature threshold.

[0132] The corresponding relationship between the difference and the collision probability is preset. Generally speaking, the collision probability is positively correlated with the difference. For example, the first fluctuation feature includes the fluctuation amplitude. At this time, the greater the difference between the fluctuation amplitude and the fluctuation amplitude threshold, the more serious the deformation of the anti-collision pad body is, and the greater the corresponding collision probability is.

[0133] Furthermore, in the case where the first waveform features include more than two, the collision probabilities corresponding to different first waveform features can be calculated respectively, and the collision probability can be determined comprehensively based on the collision probabilities corresponding to the various first waveform features. For example, the collision probabilities corresponding to the various first waveform features can be weighted and summed according to the weights of the various first waveform features to obtain the collision probability. Among them, the weights corresponding to the first waveform features can be pre-set based on factors such as the priority of the first waveform features in the judgment process and the importance of the first waveform features.

[0134] It should be noted that, when the first waveform features include more than two, some of the first waveform features may be smaller than the corresponding preset feature threshold, that is, the difference between the feature value and the preset feature threshold is a negative number. In this case, a benchmark probability (for example, 50%) can be set in the collision probability calculation process. For the first waveform feature whose feature value is greater than the corresponding preset feature threshold, it is necessary to determine the collision probability corresponding to the first waveform feature above the benchmark probability, that is, to increase the corresponding probability on the basis of the benchmark value; for the first waveform feature whose feature value is smaller than the corresponding preset feature threshold, it is necessary to determine the collision probability corresponding to the first waveform feature below the benchmark probability, that is, to reduce the corresponding probability on the basis of the benchmark value. This can improve the accuracy of abnormality judgment.

[0135] Step 603: Determine whether the collision probability is less than a collision probability threshold.

[0136] In one example, the collision probability threshold is 70%.

[0137] Step 604: When the collision probability is less than the collision probability threshold, determine whether there is a risk feature in the second waveform feature.

[0138] The risk feature refers to a feature in the waveform that indicates that there may be no collision.

[0139] Optionally, determining whether there is a risk feature in the second waveform feature includes: determining whether the second waveform feature is less than a corresponding preset feature threshold; if so, determining that there is a risk feature; if not, determining that there is no risk feature.

[0140] Furthermore, in the case where the second waveform features include more than two, different second waveform features can be compared with corresponding preset feature thresholds, and the second fluctuation feature whose feature value is less than the corresponding preset feature threshold is determined as a risk. For example, if the waveform feature includes fluctuation amplitude, it can be determined whether the fluctuation amplitude indicated by the second waveform feature is greater than the preset fluctuation amplitude threshold; if so, it is determined that the abnormal factor includes the fluctuation amplitude.

[0141] Step 605: When there is a risk feature, perform an abnormality check on the risk feature based on the first waveform feature.

[0142] In one example, an abnormality check is performed on the risk feature based on the first waveform feature, including: determining whether the risk feature in the first waveform feature is less than the corresponding feature check value; if so, generating an abnormality check result indicating that the risk feature has an abnormality; if not, generating an abnormality check result indicating that the risk feature does not have an abnormality.

[0143] The feature verification value is less than the preset feature threshold corresponding to the risk feature. For example, the feature verification value is 60% of the preset feature value.

[0144] The principle of the above-mentioned abnormality check on the risk feature is: when the risk feature is too small, it means that the probability of the risk feature indicating the existence of a collision is low, and at this time it can be determined that the risk feature is abnormal.

[0145] Optionally, in the absence of risk features, it can be directly determined that the smart crash pad has collided.

[0146] Step 606: When the verification result indicates that the risk feature is abnormal, determine that the smart anti-collision pad has not collided.

[0147] Optionally, when the verification result indicates that there is no abnormality in the risk feature, it can be directly determined that the smart crash pad has collided.

[0148] In the above implementation, the waveform characteristics corresponding to the detection data collected by different detection components can be combined to comprehensively determine whether the smart anti-collision pad has collided, which can help improve the accuracy of collision monitoring.

[0149] At the same time, when the collision probability is determined to be low based on the first waveform feature, the risk feature will be determined based on the second waveform feature, and the risk feature will be verified using the first waveform feature. Only when the verification result indicates that there is an abnormality in the risk feature will it be determined that the smart anti-collision pad has not collided. This can reduce the impact of factors other than collision on collision monitoring, which can help improve the accuracy of collision monitoring.

[0150] Based on the above implementation, further, continue to refer to Figure 8 , step 604, performing an abnormality check on the risk feature based on the first waveform feature, comprising the following steps: Step 607: When the verification result indicates that there is no abnormality in the risk feature, determine the collision feature in the first waveform feature.

[0151] The collision feature refers to a feature in the waveform that indicates the possible existence of a collision.

[0152] Optionally, determining the collision feature in the first waveform feature includes: determining the first waveform feature that is greater than the corresponding preset feature threshold in the first waveform feature as the collision feature. For example, if the fluctuation amplitude in the first waveform feature is greater than the preset change amplitude threshold, the fluctuation amplitude is the collision feature; if the fluctuation duration is less than the preset fluctuation duration threshold, the fluctuation duration is not the collision feature.

[0153] Step 608: perform collision check on the collision feature based on the second waveform feature.

[0154] In one example, the collision feature is verified based on the second waveform feature, including: determining whether the risk feature in the second waveform feature is greater than the corresponding verification value; if so, generating a collision verification result indicating that the collision feature has a collision; if not, generating a verification result indicating that the collision feature has no collision.

[0155] The feature check value is less than a preset feature threshold corresponding to the collision feature. For example, the feature check value is 60% of the preset feature value.

[0156] Step 609: When the collision check result indicates that the collision feature has a collision, it is determined that the smart anti-collision pad has a collision.

[0157] Optionally, when the collision check result indicates that the collision feature does not have a collision, it is determined that the smart collision pad has a collision.

[0158] In the above implementation, since when the abnormality check result indicates that there is no abnormality in the risk feature, the collision feature in the first waveform feature will be further checked based on the second waveform feature to finally determine whether the smart anti-collision pad has collided, the first waveform feature and the second waveform feature can be fully combined to judge whether the smart anti-collision pad has collided, which can help to further improve the accuracy of collision monitoring.

[0159] It should be understood that although the steps in the flowchart of the accompanying drawings are shown in sequence according to the instructions of the arrows, these steps are not necessarily performed in the order indicated by the arrows. Unless otherwise clearly stated in this document, the execution of these steps is not strictly limited in order and can be performed in other orders.

[0160] The above are only some implementation methods of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A smart anti-collision pad monitoring and alarm method, characterized in that: In a controller for an intelligent crash pad, the intelligent crash pad further comprises a collision detection module, a video monitoring module and an alarm module connected to the controller signal, the collision detection module is used to detect the deformation of the crash pad body, the video monitoring module is used to collect image information of the area where the crash pad body is located, and the method comprises: Monitoring the detection data collected by the collision detection module; In the case where a collision occurs with the smart anti-collision pad based on the detection data, determining a collision time corresponding to the collision and collision detection data corresponding to the collision time; Determining collision video data corresponding to the collision from the monitoring data collected by the video monitoring module based on the collision time; Determine the collision degree corresponding to the current collision by combining the collision video data and the collision detection data; Collision alarm information is generated based on the collision degree, and the alarm module is controlled to output the collision alarm information.

2. The method according to claim 1, characterized in that The collision video data includes post-collision video data, and the collision video data and the detection data determine the collision degree corresponding to the current collision, including: Determining a vehicle deformation degree of the target vehicle corresponding to the collision based on the post-collision video data; Determining the degree of deformation of the crash pad corresponding to the collision based on the collision detection data; The collision degree is determined based on the vehicle deformation degree and the crash pad deformation degree.

3. The method according to claim 2, characterized in that The collision video data further includes pre-collision video data, and before determining the collision degree based on the deformation degree of the vehicle and the deformation degree of the crash pad, further includes: determining vehicle parameters of the target vehicle based on the pre-collision video data; Determining a prediction confidence corresponding to the degree of deformation of the vehicle based on the vehicle parameters; Predicting the deformation degree of the crash pad body based on the prediction confidence and the deformation degree of the vehicle to obtain a predicted deformation range; determining whether the deformation degree of the crash pad is within the predicted deformation range; In a case where the degree of deformation of the crash pad is within the predicted deformation range, the degree of collision is determined based on the degree of deformation of the vehicle and the degree of deformation of the crash pad.

4. The method according to claim 3, characterized in that After determining whether the collision deformation degree is within the predicted deformation range, the method further includes: When the degree of deformation of the crash pad is not within the predicted deformation range, determining a predicted deformation amount of the crash pad body based on the post-collision video data; determining whether the degree of deformation of the crash pad matches the predicted deformation amount; In a case where the crash pad deformation degree matches the predicted deformation amount, the collision degree is determined based on the crash pad deformation degree and the predicted deformation amount.

5. The method according to claim 4, characterized in that After determining whether the deformation degree of the crash pad matches the predicted deformation amount, the method further includes: In the case where the collision deformation degree does not match the predicted deformation amount, determining whether the predicted deformation amount matches the predicted deformation range; In a case where the predicted deformation amount matches the predicted deformation range, the collision degree is determined based on the predicted deformation amount and the vehicle deformation degree.

6. The method according to claim 1, characterized in that The monitoring of the detection data collected by the collision detection module includes: Monitoring whether the detection data fluctuates; In the case where the detection data is monitored to fluctuate, determining waveform features corresponding to the detection data; Determine whether the smart crash pad collides based on the waveform feature.

7. The method according to claim 6, characterized in that The collision detection module includes a first detection component and a second detection component, the detection data includes first detection data corresponding to the first detection component and second detection data corresponding to the second detection component, and when the detection data is detected to fluctuate, determining the waveform characteristics of the detection data includes: When the first detection data and / or the second detection data are detected to fluctuate, respectively determining a first waveform feature of the first detection data and a second waveform feature of the second detection data; The step of determining whether the smart anti-collision pad collides based on the waveform characteristics comprises: determining a collision probability based on the first waveform feature; determining whether the collision probability is less than a collision probability threshold; When the collision probability is less than the collision probability threshold, determining whether there is a risk feature in the second waveform feature; In the case where the risk feature exists, performing an abnormality check on the risk feature based on the first waveform feature; When the abnormality check result indicates that the risk feature is abnormal, it is determined that the smart anti-collision pad has not collided.

8. The method according to claim 7, characterized in that After performing abnormality check on the risk feature based on the first waveform feature, the method further includes: When the abnormality check result indicates that the risk feature does not have an abnormality, determining a collision feature in the first waveform feature; Performing a collision check on the collision feature based on the second waveform feature; When the collision check result indicates that the collision feature has a collision, it is determined that the smart anti-collision pad has a collision.

9. An intelligent anti-collision pad, characterized in that: The intelligent anti-collision pad includes a controller and a collision detection module, a video monitoring module and an alarm module connected to the controller by signal, wherein the collision monitoring module is used to detect the deformation of the anti-collision pad body, and the video monitoring module is used to collect image information of the area where the anti-collision pad body is located; The controller is used to execute the intelligent anti-collision pad monitoring and alarm method according to any one of claims 1 to 8, and send collision alarm information to the background server.

10. An intelligent anti-collision pad alarm system, characterized in that: The system comprises the smart anti-collision pad according to claim 9, and a background server communicatively connected to the smart anti-collision pad; The intelligent anti-collision pad is used to generate collision warning information and send it to the background server; The background server is used to process the collision warning information.

Images