A system and method for impact force detection in a glazing

The system addresses the challenge of detecting impact forces on glazing and predicting breakage by employing piezoelectric sensors and edge computation with machine learning, providing real-time alerts and preventive maintenance to enhance safety and efficiency in insurance claims.

WO2025032618A1PCT designated stage expired Publication Date: 2025-02-13SAINT GOBAIN SEKURIT FRANCE +1

Patent Information

Application Number
PCT/IN2024/051479
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-08-10
Filing Date
2024-08-09
Publication Date
2025-02-13

AI Technical Summary

Technical Problem

Existing solutions fail to effectively detect impact forces on glazing, such as windshields, and do not provide timely alerts or predictive maintenance to prevent breakage, which is critical for safety and insurance cost estimation.

Method used

A system and method utilizing piezoelectric sensors and edge computation to detect impact forces on glazing, assess the health of the glazing, and predict potential breakage by comparing real-time data with historical data using machine learning approaches, triggering alerts and prompting preventive services.

Benefits of technology

The system provides real-time impact force detection and predictive maintenance, enabling immediate alerts and reducing the risk of windshield breakage, thereby enhancing safety and streamlining insurance claims processes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure IN2024051479_13022025_PF_FP_ABST
    Figure IN2024051479_13022025_PF_FP_ABST
Patent Text Reader

Abstract

Disclosed herein is a solution system (100) and a method for impact force detection in a glazing (110) and prompting of further actions The system includes impact detection sensing units (101) to detect a force impact on the glazing (110). It further includes a slave control unit (102) implementing edge computation to obtain input from said one or more impact detection sensing units (101) and to assess a force impact level on the glazing (110). It further includes a master control unit (103) being operably configured to a computation unit (104) to determine a risk level of the glazing (110) based on the state of the glazing (110) and to perform further actions. It provides an improved solution for tracing the impact force on the glazing for forecasting of breakage of the glazing that would help in preventive service prompting and informing the stakeholders.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] A SYSTEM AND METHOD FOR IMPACT FORCE DETECTION IN A GLAZING

[0002] TECHNICAL FIELD

[0003] The present disclosure relates to impact force detection in a glazing, it particularly relates to an internet of things (loT) based system involving edge computation for impact force detection in a glazing.

[0004] BACKGROUND

[0005] Background description includes information that may be useful in understanding the present disclosure. It is not an admission that any of the information provided herein is prior art or relevant to the presently claimed disclosure, or that any publication specifically or implicitly referenced is prior art.

[0006] When a small rock hits a windshield of a vehicle, the impact it creates may often not be alarming. But eventually over a period, it may be adversely affecting the health of the windshield. Such consequential impacts over a period on windshield from objects such as sharp stone or so could lead to immediate cracks or gradual breakage. Immediate cracks create a negative effect on visibility of a driver, structural integrity of the windshield and the vehicle. Thus, it is critical for safety if the impact of objects on windshield of a vehicle is not attended at right time. Additionally, the insurance industry relies heavily on cost estimation. The use of automated damage detection and cost estimation techniques will reduce both the amount of time, and for the purpose a real time-based data is required.

[0007] Reference is made to EP3766746 that discloses a solution for operating a vehicle comprises providing measurement data and determining condition data dependent on the measurement data, the condition data being representative of a condition of the windshield. However, it does not talk about any alert or apt prompt being conveyed to users for taking a corrective action.

[0008] Another reference is made to JP3802897 that discloses a solution that focuses on the phenomenon that when the tempered glass used in a vehicle is broken, the glass expands to the outside of the vehicle at the same time as cracks generated at a high speed, thereby abruptly occurring in the vehicle. In this solution, an alarm signal is generated only by detecting a specific pressure change that occurs only in an abnormal situation such as a glass breakage of a vehicle, so that a general pressure change is detected. However, it does not talk about alerting stakeholders such as vehicle owners or insurance stakeholders. Additionally, it is also silent about detecting the health of the glazing for prediction of breakage.

[0009] Yet another reference is made to W02020005554A1 that provides an improved glazing or glass laminate systems using laminates that are thin, light, and durable against certain impacts but that have optimized breakage for pedestrian safety in the event of a pedestrian impact. Provisions of detection of windshield health and prediction of breakage is not provided by said reference.

[0010] In view of the fore-going and the drawbacks discussed in the existing solution, there exists a dire need of an improved solution for tracing the impact force of an object on the glazing, obtaining the health of the glazing for forecasting of breakage of the glazing that would help in preventive service prompting and informing the stakeholders (owners, insurance stakeholders and vehicle manufacturers).

[0011] SUMMARY OF THE DISCLOSURE

[0012] An object of the present invention is to provide a solution for overcoming the drawbacks of the prior art solutions.

[0013] Another object of the present invention is to provide an improved solution for detecting an impact force of an object on the glazing.

[0014] Yet another object of the present invention is to provide a solution for obtaining the health of the glazing based instant data and historical data.

[0015] Still another object of the present invention is to provide a solution for forecasting breakage of the glazing informing the stakeholders and prompting preventive service.

[0016] These and other objects of the invention are achieved by the following aspects of the invention. The following disclosure presents a simplified summary of the invention in order to provide a basic understanding of some aspects of the invention. This presents some concept of the invention in a simplified form to a more detailed description of the invention presented later. It is a comprehensive summary of the disclosure and it is not an extensive overview of the present invention. The intend of this summary is to provide a fundamental understanding of some of the aspects of the present invention.

[0017] In an aspect of the present invention is provided a system for impact force detection in a glazing and prompting of further actions. Said system comprises one or more impact detection sensing units, disposed on the glazing, configured to detect a force impact on the glazing. It further comprises a slave control unit configured to obtain input from said one or more impact detection sensing units and to assess a force impact level on the glazing. The slave control unit is further configured to determine a deviation of the obtained input from a standard data and generate an impact data, said impact data being indicative of a state of the glazing and conditionally trigger a master control unit, said master control unit being operably configured to a computation unit to determine a risk level of the glazing based on the state of the glazing and to perform further actions. Said one or more impact detection sensing units are piezoelectric sensors disposed in the glazing and is adapted to obtain a signature of a signal sensed through the material of the glazing. Said signature of a signal is defined by at least the amplitude and frequency of the signal. The slave control unit is configured to obtain the force impact level by comparing a change in signature of the signal (amplitude, frequency) obtained from said one or more impact detection sensing units with the standard data, said standard data include pre-determined values of amplitude and frequency for measurement of impact level and trigger the master control unit, if obtained impact level is more than a set threshold value. The computation unit is wirelessly coupled with the master control unit and is configured to assess a change in state of the glazing, state of glazing being identified at least with the force impact level and one or more physical parameters of glazing. The computation unit is configured to apply machine learning approaches to compare the change in state of the glazing at different instances based on historical data, and further to assess a risk factor of the glazing. The computation unit is configured to identify deviation characteristics for identifying the change in state of the glazing. The computation unit is configured to classify the state of glazing into at least three risk factor-based categories including low risk, medium risk and high risk. The computation unit is configured to trigger a cautionary message via a communication unit to one or more users for medium risk state and high-risk states. The computation unit is configured to trigger the master control unit to display a cautionary message in a display unit for risk state. The slave control unit is configured to calibrate a data of a previous impact against a physical model as a base line for estimating data for a consequent impact on the glazing. The slave control unit is configured to compare the changes in obtained signature of signal with pre-determined values and the slave control unit is further configured to identify a first major change in obtained signature, said first major change is measured with high precision. The computational unit is configured to perform secondary analysis to compare the change in state of the glazing at different instances based on historical data, and further to assess a risk factor of the glazing.

[0018] In another aspect of the present invention is provided a method for impact force detection in a glazing and prompting of further actions. Said method comprising obtaining, by impact detecting sensing units, one or more parameters to determine a state of a glazing, detecting, by a slave control unit, for deviation in the state of the glazing for obtaining an impact level, conditionally triggering, by the slave control unit, a master control unit for obtaining a risk factor on the state of the glazing, wherein the master control unit is triggered, if an obtained impact level is more than a set threshold value and communicating, by the master control unit, a cautionary message indicating the risk of the glazing. The step of detecting for deviation in the state of the glazing comprises obtaining, by the slave control unit, signature of the signal (amplitude and frequency of signal) sensed through the material of the glazing and comparing, by the slave control unit, the changes in obtained amplitude and frequency of signal with predetermined values. The method includes the step of analysing, by the slave control unit, the instances at which said changes in the obtained signature of signal occurs. There is a step of directed at comparing the changes in obtained signature of signal with pre-determined values further provides identification of a first major change in measurement signature, said first major change is measured with high precision. This method further includes transferring, to the slave control unit, the first major change in the signature of a signal from the sensing unit. The method includes the step of determining a state of the glazing, wherein the said state is obtained by determining the impact level and one or more parameters of glazing. This method includes secondary analysing performed by a computational unit, using machine learning approaches to compare the change in state of the glazing at different instances based on historical data, and further to assess a risk factor of the glazing. It also includes calibrating, by the slave control unit, a data of a previous impact against a physical model as a base line for estimating data for a consequent impact on the glazing. The one or more aspects of the present invention provides a solution for a real-time machine learning model on the edge, in which relevant stakeholders or users will get immediate response regarding the state of the glazing of the vehicle such as the windshield. It provides a data model in addition to a physical model, in which for a subsequent impact on the glazing, the data model of the previous impact calibrated against physical model of the previous impact may be used as base line for estimating impact of subsequent hit. A machine learning model is used for forecasting of breakage and glazing service prompts to prevent breakage. Alternatively, suitable state estimators (such as and not limited to a recursive approach) may be used.

[0019] The significant features of the present invention and the advantages of the same will be apparent to a person skilled in the art from the detailed description that follows in conjunction with the annexed drawings.

[0020] BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The following briefly describes the accompanying drawings, illustrating the technical solution of the embodiments of the present invention or the prior art, for assisting the understanding of a person skilled in the art to comprehend the invention. It would be apparent that the accompanying drawings in the following description merely show some embodiments of the present invention, and persons skilled in the art can derive other drawings from the accompanying drawings without deviating from the scope of the disclosure.

[0022] FIG. 1 illustrates a block diagram of the system according to an embodiment of the present invention.

[0023] FIG. 2 illustrates a schematic diagram of the sensors being disposed on a windshield according to an embodiment of the present invention.

[0024] FIG. 3 illustrates a schematic diagram of a sensor within a laminated glass according to an embodiment of the present invention.

[0025] FIG. 4 illustrates a block diagram of the method according to an embodiment of the present invention. FIG. 5 illustrates a schematic diagram of a system architecture according to an embodiment of the present invention.

[0026] Persons skilled in the art will appreciate that elements in the figures are illustrated for simplicity and clarity and have not necessarily been drawn to scale. For example, the dimensions of some of the elements in the figures may be exaggerated relative to other elements to help to improve understanding of embodiments of the disclosure.

[0027] DETAILED DESCRIPTION

[0028] The present disclosure is now discussed in more detail referring to the drawings that accompany the present application. It would be appreciated by a skilled person that this description to assist the understanding of the invention, but these are to be regarded as merely exemplary.

[0029] The terms and words used in the following description are not limited to the bibliographical meanings and the same are used to enable a clear and consistent understanding of the invention. Accordingly, the terms / phrases are to be read in the context of the disclosure and not in isolation. Additionally, descriptions of well-known functions and constructions are omitted for clarity and conciseness.

[0030] The one or more embodiments of the present invention disclose a solution for real-time glass damage detection in a glazing of a vehicle. It provides a solution for automated impact force detection. Said impact force may be defined as the impact created on the glazing due to stone impact, stress, or any other jerk impact detection on the glazing. The real-time detection provides several advantages inclusive of fast-track insurance claim, quicker intimation to the stakeholders and thereby facilitating the further actions required for the glazing. The use of automated damage detection and cost estimation techniques can reduce both the amount of time and money spent on the insurance survey. In the present invention, edge computing has been advantageously used for the real time impact force detection on a glazing.

[0031] Reference is made to FIG. 1 that discloses a system (100) for impact force detection in a glazing (110) of a vehicle and prompting of further actions thereof. Said system comprises one or more impact detection sensing units (101), disposed on the glazing (110), configured to detect a force impact on the glazing (110). It further includes a slave control unit (102) configured to obtain input from said one or more impact detection sensing units (101) and to assess a force impact level on the glazing (110). The slave control unit (102) is further configured to determine a deviation of the obtained input from a standard data and generate an impact data, said impact data being indicative of a state of the glazing (110). The slave control unit (102) is configured to conditionally trigger a master control unit (103). This computation unit (104) is responsible to determine a risk level of the glazing (110) based on the state of the glazing (110) and to perform further actions. The present invention thus advantageously provides a data model in addition to a physical model, in case of second impact, the data model of the first impact calibrated against physical model of the first impact to be used as base line for estimating impact of second hit.

[0032] In an embodiment of the present invention, the slave control unit (102) is configured to obtain the force impact level by comparing a change in signature of an obtained signal from said one or more impact detection sensing units (101) with the standard data. The signature of a signal may be defined by at least the amplitude and frequency of the signal. Thus, the force impact level is obtained by comparing a change in amplitude, frequency obtained from one or more impact detection sensing units (101) with the standard data. Said standard data include predetermined values of amplitude and frequency for measurement of impact level. The slave control unit (102) is further configured to trigger the master control unit (103) if the obtained impact level is more than a set threshold value. The slave control unit (102) is thus configured to perform the pre-checks associated with the force impact analysis and verify if the change observed in the impact is significant compared to pre-determined information to send the data for further processing. In an implementation it may be regarded to compare Sn-i (decision function) ~ Sn(decision function) {here decision function is dependent on amplitude, frequency of the signal measured by the sensing units and n is represents the data at that instant of time}. In an alternate embodiment, the slave control unit (102) may directly be coupled to the computation unit (104). When a small rock hits a windshield, the cracks that appear are hardly a welcome sight, but the star-shaped pattern may contain information about the rock and the glass. Similarly, when a physical entity hits the glazing, it leaves its impact data or information on the glazing. The slave control unit (102) utilizes this information. Said control unit includes a pre-processing module (1021) configured for noise removal and performing pre-checks on the procured or sensed data. It further includes a data sensor processing module (1022) configured for processing the obtained data from sensing units. For further understanding, an example could be considered here. FIG. 2 depicts a schematic diagram of how the vehicle glazing (e.g., windshield) with the sensing units obtains data about an impact created on the glazing. If there is a stone impact on the windshield, then the signature of the measurements obtained from the sensing unit would be different from that of normal healthy conditions of the glazing. The difference in said measurement is observed and analysed by the slave control unit (102). Slave control unit (102) not only identifies for deviations in an instant stone impact measurement, but also checks for difference from the previous changes as well. It is thus, configured to compare change in signature of the measurements (5i, 82, 63, 64). The difference in the signature of measurements (61, 62, 63, 64) is identified by the slave control unit (102) itself. The very first major change in measurement signature in amplitude and frequency (signature of signal) is critical, and the measured data is of high precision. This measured data needs to be transferred to slave control unit (102) from the sensing unit (101). The slave control unit advantageously makes use of edge computing. The present invention thus includes a data model (with usage of historical data) in addition to the physical model. Thus, in case of second impact, the data model of the first impact calibrated against physical model of the first impact is be used as base line for estimating impact of second hit. The slave control unit (102) is configured to calibrate a data of a previous impact against a physical model as a base line for estimating data for a consequent impact on the glazing. The slave control unit (102) is further configured to compare the changes in obtained signature of signal with pre-determined values. The slave control unit (102) is still further configured to identify a first major change in obtained signature, said first major change is measured with high precision.

[0033] In an embodiment of the present invention, the computation unit (104) is wirelessly coupled with the master control unit (103) and is configured to assess a change in state of the glazing, state of glazing being identified at least with the force impact level and one or more physical parameters of glazing (110). Said physical parameters of the glazing may include nature of the substrate, curvature of the glass, coatings on it and the like. However, it is not limited to these. In an implementation, said computation unit (104) is a remote device in a cloud network. In another implementation, it may be integrated within the vehicle itself operably coupled either directly to the slave control unit (102) or the master control unit (103) or both. In an implementation, the computation unit (104) may include machine learning model / module (1041) and a risk level analyzing module (1042). In this embodiment, the computational unit (104) is configured to perform secondary analysis to compare the change in state of the glazing at different instances based on historical data, and further to assess a risk factor of the glazing. Such secondary analysis may include interpreting the risk factor basis the historical data using machine learning models. Alternatively, suitable state estimators (such as and not limited to a recursive approach) may be used as well.

[0034] The slave control unit (102) may thus transfer the data from sensing unit (101) and the change in measurement signatures to the computation unit (104). The data model may thus rely upon these historical data transferred to the computation unit (103), either directly from slave control unit (103) or via the master control unit (103). The computation unit (104) is configured to apply machine learning approaches to compare the change in state of the glazing (100) at different instances based on historical data, and further to assess a risk factor of the glazing. The computation unit (104) is configured to identify deviation characteristics for identifying the change in state of the glazing (100). Over the period, the computation unit (104) which may be a remote cloud-based device is configured to record all the outcomes of the processing or analysis in the cloud. In an implementation, the computation unit (104) is configured for service determination. It is configured to predict the breakage of the glazing which in turn helps in preventive service prompting and informing users such as insurance stakeholders for instance.

[0035] In an embodiment of the present invention, the computation unit (104) is configured to identify deviation characteristics for identifying the change in state of the glazing (100). It is further configured to classify the state of glazing (110) into at least three risk factor-based categories including low risk, medium risk and high risk. The computation unit (104) is configured to trigger a cautionary message via a communication unit (105) to one or more users for medium risk state and high-risk states. Additionally, the computation unit (104) is configured to trigger the master control unit (103) to display a cautionary message in a display unit (106) for high- risk state.

[0036] In an implementation of the present invention, the data obtained from the slave control unit (102) may be stored in a remote cloud storage device, in which said cloud storage device is operably configured to the computation unit (104). Therefore, the data available in the cloud may be used for precise detection of location of impact, and the risk level among others. Say for instance, using the data available in the cloud, the very first stone impact on the windshield may be traced and identify probable place of the impact. The computational unit (104) is then further configured to use statistical and machine learning methods to reconstruct or restructure the place of impact. Said computational unit may be either a local processing unit such as the Edge computing-based control unit or the a cloud based computational unit. Consequently, with the already computed results from the statistical and machine learning model over a period with a given stone impacted windshield, the computation unit (104) is configured to provide further actions such as the prediction of the estimate time for the complete replacement of windshield, intimating the users (the vehicle manufacturer, windshield manufacturer, the owner of the vehicle) the probable replacement time to avoid risk of safety while driving. It, thus, may display a suitable cautionary message in a display for risk state.

[0037] In an implementation of the present invention, the risk forecast could be categorised into different levels (such as low, medium, high) and based on risk estimate, in order to take a suitable further action. If the impact estimated is low or negligible, the record of the result of the analysis may simply be retained, if the impact is moderate, stakeholders such as the insurance or warranty users, vehicle owner may be informed and no immediate action may be required, but recommendations on service may be provided. However, if the impact estimate is high, an immediate action may be suggested to the vehicle owner besides intimating the stakeholders. Said immediate action may include displaying an alert in a display device of the vehicle, or sound alarms, however, not limited to this. Intimation to the stakeholders may be provided as a short message service into the registered mobile numbers of the stakeholders. The slave control unit which uses edge computing may be further configured to enable data directly to cloud through modules like Wi-Fi, GSM, narrowband loT connections and the like, however not limited to these. Information provided to warranty or insurance related stakeholders would advantageously help to save the cost, safety risk that might be higher when windshield breaks fully while driving.

[0038] In an embodiment of the present invention, the one or more impact detection sensing units (101) may be piezoelectric sensors (101a, 101b, 101c, lOld). Piezo or any other relevant sensors placed inside the glazing, and they provide measured data about present condition of the glazing. Said sensors may be disposed in the glazing (110) and is adapted to provide amplitude and frequency of signals sensed through the material of the glazing (110). In an implementation, said sensor may be disposed within the layers of the glazing via through-hole drilled, integrated via lamination and the like. FIG. 3 depicts a schematic diagram of sensing unit (101) being disposed in the glazing (110). The figure depicts a laminated glass used for windshield of the vehicle. Said laminated glass comprises two layers of glass (1105, 1107) sandwiched together with an interlayer (1106) such as Polyvinyl butyral (PVB). The sensing units (101) may be disposed on the glazing considering the visibility for driver, aesthetics, the mechanical integrity of the glazing and the layers. In an implementation, the sensing units may be so disposed that it is adapted to vibrate with the layers of laminated glass of the glazing to provide measurement signature of the signal for accurate impact detection. The glazing may include laminated glass, however, it is not limited to this and the glazing may be any of windshield, sidelite, backlite, quaterlite, or sunroof. The system may be suitably adapted for architectural glazing or facade as well.

[0039] In an implementation, said system (100) may be an internet of things (loT) with the master control unit (103) being communicatively coupled to the slave control unit (102). The slave control unit (102) may directly be communicatively coupled to the computation unit (104). Further, the slave control unit (102) may use Edge computing. It advantageously facilitates, the identification of windshield breakage patterns at the edge level, thereby reducing response time and enabling cloud processing for computationally intensive pattern recognition.

[0040] Reference is made to a method (200) for impact force detection in a glazing and prompting of further actions. Said method comprises at step S201, impact detecting sensing units obtains one or more parameters to determine a state of a glazing. The method further includes at step S202, a slave control unit detects deviation in the state of the glazing for obtaining an impact level. At step S303, the slave control unit conditionally triggers a master control unit for obtaining a risk factor on the state of the glazing, wherein the master control unit (103) is triggered, if an obtained impact level is more than a set threshold value. This conditional triggering is dependent on the decision function computation performed by the slave control unit. At step S204, the master control unit then communicates a cautionary message indicating the risk of the glazing. FIG. 4 depict the overview this method.

[0041] In an embodiment, the detection for deviation in the state of the glazing comprises slave control unit obtaining amplitude and frequency of signal sensed through the material of the glazing at S2021. It further includes the slave control unit comparing the changes in obtained signature (amplitude and frequency) of signal with pre-determined values at S2022. At step S2023, the slave control unit analyses the instances at which said changes in the obtained signature of signal occurs. This includes analysing the nature of the signal, particularly to observe if it is a continuous change (a continuous increase or decrease). At step S2022, the changes are compared. The changes in obtained signature of signal with pre-determined values are compared and further provided identification of a first major change in measurement signature, said first major change is measured with high precision. The method still further includes step S2024, involving transferring, to the slave control unit, the first major change in the signature of a signal from the sensing unit. At step S2025, the first major change in the signature of a signal is transferred from the sensing unit to the slave control unit. At step, S2025, a computational unit performs secondary analysing using machine learning approaches to compare the change in state of the glazing at different instances based on historical data, and further to assess a risk factor of the glazing. At step S2026, the slave control unit calibrates a data of a previous impact against a physical model as a base line for estimating data for a consequent impact on the glazing.

[0042] In the present invention is proposed a system that uses a data model in addition to the physical model for detection of impact. This advantageously improves the real-time responses and makes adept use of edge computation. Thus, provides use cases for service prompting, predictive maintenance of any products including windshield piezo electric signals. The insurance industry relies heavily on cost estimation. The use of automated damage detection and cost estimation techniques can reduce both the amount of time and fast track claims. This method can automate windshield claims. Thus, the claim will not require any manual intervention in a specific percentage of cases.

[0043] Example Embodiment: Reference is made to FIG. 5 that depicts a system for stone impact detection and prediction of windshield (110) breakage in a vehicle (300). The one or more impact force sensors like the piezoelectric sensors (101) are used to monitor the windshield in the vehicle body (301), and the measured data is transmitted to the slave control unit (102). The slave control unit (102) compares the measured data with an already existing established pattern or signature of a healthy windshield. Upon detection, the slave control unit (102) sends a complete measured data to the computation unit (104), which may be a cloud server via a wireless module (304). This wireless mode may be realized through internet protocols. The Cloud server utilizes a pre-trained machine learning (ML) model to analyze the data, predict the probability of windshield breakage, and categorize it into different risk levels. It is configured to identify a state change model / piezo and other characteristics of windshield, identify deviation characteristics variables at state change, use machine learning model to compare the deviation collected at different times, and compare the model output at the latest state with initial suggested output. Once these are done, it is capable of classifying the impact level in near-term, long-term basis and inform the stakeholders (303). The cloud server makes use of historical data. Stakeholders may be informed via Global System for Mobile communication, GSM module (302) communicatively coupled to the master control unit (103). Information regarding the identified risk levels may be communicated to the Vehicle Electronic control unit, ECU (305) and shared with the relevant stakeholders. The Vehicle ECU (305) receives notifications about high-impact collisions and severe breakage conditions both from the Edge level, i.e., Slave ECU and the Cloud server. The sensor data is transmitted to the cloud server only when patterns identical to breakage are detected. This approach optimizes the utilization of the cloud server's computational power by sending and processing the sensor data, while also benefiting from the presence of a pre-trained machine learning model stored in the memory of the server. Enabling timely decision-making, notifying the vehicle ECU about the initial detection of risky levels, and sharing cloud-processed results. This advantageously allows users to promptly respond to alert messages and leveraging cloud-based messaging ensures that all stakeholders receive timely notifications regarding these risky levels.

[0044] Industrial Application: The present invention finds use cases in the insurance industry that relies heavily on cost estimation. The use of automated damage detection and cost estimation techniques may reduce both the amount of time, thereby fast track claims. It provides additional alert to manufacturers for prompt service to the vehicle or the glazing. Furthermore, it includes means for safeguarding the driver from issues caused due to windshield breakage.

[0045] Some advantages of the present invention are enlisted in the following:

[0046] • The solution provides real-time response which is implemented using a real-time machine learning model on the edge, users get immediate responses. Furthermore, Edge Deployed machine learning models is faster.

[0047] • It makes use of both data model and physical model and hence, the accuracy of detecting windshield health and prediction is improved.

[0048] • It implements loT architecture and thereby improving reliability.

[0049] Many alterations and modifications of the present invention will no doubt become apparent to a person skilled in the art after having read the foregoing description. It is to be understood that the phraseology or terminology employed herein is for the purpose of description and not of limitation. It is to be understood that the description above contains many specifications, these should not be construed as limiting the scope of the invention but as merely providing illustrations of some of the personally preferred embodiments of this invention. Thus, the scope of the invention should be determined by the appended claims and their legal equivalents rather than by the examples given. Features that are described and / or illustrated with respect to one embodiment may be used in the same way or in a similar way in one or more other embodiments and / or in combination with or instead of the features of the other embodiments.

[0050] List of reference numerals appearing in the accompanying drawings and the corresponding features:

[0051] 100: system

[0052] 101 : impact detection sensing units

[0053] 1021 : pre-processing module

[0054] 1022: data sensor processing module

[0055] 102: slave control unit

[0056] 103: master control unit

[0057] 104: computation unit

[0058] 101a, 101b, 101c, lOld: piezoelectric sensors

[0059] 110: glazing

[0060] 1105,1107: glass substrate

[0061] 1105: interlayer

[0062] 300: vehicle

[0063] 301 : vehicle body

[0064] 302: GSM module

[0065] 303: stakeholders

[0066] 304: internet

[0067] 305: vehicle ECU

[0068] 1041 : ml model

[0069] 1042: risk level analyzing module

[0070] 200: method

[0071] S201-S204, S2021-S2026: method steps

Claims

CLAIMS1. A system (100) for impact force detection in a glazing (110) and prompting of further actions, wherein said system comprises: one or more impact detection sensing units (101), disposed on the glazing (110), configured to detect a force impact on the glazing (110); a slave control unit (102) configured to obtain input from said one or more impact detection sensing units (101) and to assess a force impact level on the glazing (110); wherein said slave control unit (102) is further configured to: determine a deviation of the obtained input from a standard data and generate an impact data, said impact data being indicative of a state of the glazing (110); and conditionally trigger a master control unit (103), said master control unit (103) being operably configured to a computation unit (104) to determine a risk level of the glazing (110) based on the state of the glazing (110) and to perform further actions.

2. The system (100) as claimed in claim 1, wherein said one or more impact detection sensing units (101) are piezoelectric sensors (101a, 101b, 101c, 101 d) disposed in the glazing (110) and is adapted to obtain a signature of signals sensed through the material of the glazing (HO).

3. The system (100) as claimed in claim in claim 1 or 2, wherein said signature of a signal is defined by at least the amplitude and frequency of the signal.

4. The system (100) as claimed in any one of claims 1-3, wherein the slave control unit (102) is configured to: obtain the force impact level by comparing a change in signature of the signal obtained from said one or more impact detection sensing units (101) with the standard data, said standard data include pre-determined values of the signature of the signal for measurement of impact level; and trigger the master control unit (103) if the obtained impact level is more than a set threshold value.

5. The system (100) as claimed in any one of claims 1-4, wherein the computation unit (104) is wirelessly coupled with the master control unit (103) and is configured to assess a change in state of the glazing, state of glazing being identified at least with the force impact level and one or more physical parameters of glazing (110).

6. The system (100) as claimed in any one of claims 1-5, wherein the computation unit (104) is configured to apply machine learning approaches to compare the change in state of the glazing (100) at different instances based on historical data, and further to assess a risk factor of the glazing.

7. The system (100) as claimed in any one of claims 1-6, wherein the computation unit (104) is configured to identify deviation characteristics for identifying the change in state of the glazing (100).

8. The system (100) as claimed in any one of claims 1-7, wherein the computation unit (104) is configured to classify the state of glazing (110) into at least three risk factor-based categories including low risk, medium risk and high risk.

9. The system (100) as claimed in any one of claims 1-8, wherein the computation unit (104) is configured to trigger a cautionary message via a communication unit (105) to one or more users for medium risk state and high-risk states.

10. The system (100) as claimed in any one of claims 1-9, wherein the computation unit (104) is configured to trigger the master control unit (103) to display a cautionary message in a display unit (106) for risk state.

11. The system (100) as claimed in any one of claims 1-10, wherein the slave control unit (102) is configured to calibrate a data of a previous impact against a physical model as a base line for estimating data for a consequent impact on the glazing.

12. The system (100) as claimed in any one of claims 1-11, wherein the slave control unit (102) is configured to compare the changes in obtained signature of signal with pre-determined values; andthe slave control unit (102) is further configured to identify a first major change in obtained signature, said first major change is measured with high precision.

13. The system (100) as claimed in any one of claims 1-12, wherein the computational unit (104) is configured to perform secondary analysis to compare the change in state of the glazing at different instances based on historical data, and further to assess a risk factor of the glazing.

14. A method (200) for impact force detection in a glazing and prompting of further actions, wherein said method comprising: obtaining (S201), by impact detecting sensing units, one or more parameters to determine a state of a glazing; detecting (S202), by a slave control unit, for deviation in the state of the glazing for obtaining an impact level; conditionally triggering (S203), by the slave control unit, a master control unit for obtaining a risk factor on the state of the glazing, wherein the master control unit (103) is triggered, if an obtained impact level is more than a set threshold value; and communicating (S204), by the master control unit, a cautionary message indicating the risk of the glazing.

15. The method (200) as claimed in claim 14, wherein detecting (S202) for deviation in the state of the glazing comprises: obtaining (S2021), by the slave control unit, signature of signal sensed through the material of the glazing, said signature of the signal is defined by at least amplitude and frequency of said signal; comparing (S2022), by the slave control unit, the changes in obtained signature of signal with pre-determined values; and analysing (S2023), by the slave control unit, the instances at which said changes in the obtained signature of signal occurs.

16. The method (200) as claimed in claim 14 or claim 15, wherein comparing (S2022) the changes in obtained signature of signal with pre-determined values further provides identification of a first major change in measurement signature, said first major change is measured with high precision.

17. The method (200) as claimed in any one of claims 14-16, comprises transferring (S2025), to the slave control unit, the first major change in the signature of a signal from the sensing unit.

18. The method (200) as claimed in any one of claims 14-17, comprises determining (S2026), by the slave control unit, a state of the glazing, wherein said state is obtained by determining the impact level and one or more physical parameters of glazing (110).

19. The method (200) as claimed in any one of claims 14-18, comprises secondary analysing (S2026), by a computational unit, using machine learning approaches to compare the change in state of the glazing at different instances based on historical data, and further to assess a risk factor of the glazing.

20. The method (200) as claimed in any one of claims 14-19, comprises calibrating (S2027), by the slave control unit, a data of a previous impact against a physical model as a base line for estimating data for a consequent impact on the glazing.

Citation Information

Patent Citations

  • Systems and Methods for Damage Detection

    US20200013160A1

  • Method for assessing an external event on an automotive glazing

    US20230025723A1

Cited By

  • Renewable resource classification management method and system based on image recognition

    CN120763781A