IT202200025560A1 - Mobile Device Tracking System - Google Patents

Description

DESCRIPTION of the Industrial Invention entitled: ?Mobile device monitoring system?

DESCRIPTION TEXT

The present invention relates to a system and a method that allow to detect the movements of a smartphone, certify the date and time of a potential fall and quantify the potential damage suffered by the same.

In particular, the present invention relates to a monitoring system for a mobile device for detecting impact damage, which mobile device comprises at least one gyroscope, at least one accelerometer, means for connection to a communication network and processing means for executing a logic program configured to detect the signals emitted by said gyroscope and/or said accelerometer, apply a time stamp to them and send them via said connection means to remote processing means.

The intended field of application is that of insurance services and the primary purpose is to certify the damages suffered by the insured products.

There is currently an expanding insurance sector that uses IoT platforms, sensors and connected devices to collect data useful for monitoring the actual behavior of customers, on the basis of which the prices of the insurance product are dynamically calculated or the customer's behavior is verified in the event of a claim.

The use of monitoring technology, even in real time, allows not only a more accurate risk assessment compared to a traditional estimate and greater flexibility and customization of policies, but also to influence customer behavior, promoting virtuous behavior and decreasing the probability of risk.

This approach is well established in the automotive sector, and is now gaining traction in the home and healthcare sectors, where the growing popularity of wearable devices that track a user's activity and vital signs means it is theoretically possible for insurers to obtain ongoing data on their customers' lifestyles, which can be used to encourage and reward good behaviour in a way not dissimilar to that used in the automotive sector.

However, no insurance product has been designed for a class of devices that are increasingly essential in everyday life such as mobile devices, be they smartphones but also tablets or laptops.

However, the insurance of these product classes must mitigate the risk of fraud as much as possible.

Mobile devices are typically provided with at least one gyroscope, at least one accelerometer, means of connection to a communications network and processing means for the execution of a logic program configured to detect the signals emitted by said gyroscope and/or said accelerometer, apply a time stamp thereto and send them via said means of connection to remote processing means.

The accelerometer sensor measures the acceleration of the smartphone's movement on the three axes (X, Y, Z) and its primary use is to automatically move the device's orientation from portrait to landscape.

The gyro sensor detects the movements of the smartphone, and is used, for example, by video games to move the view left and right by moving the smartphone.

The prior art offers numerous applications that exploit smartphone sensors, but none of these are aimed at monitoring movements for insurance purposes.

There is currently a need, not satisfied by the state of the art, for a system and method for monitoring mobile devices that allows for the evaluation of actual accidental falls of the same.

The present invention aims to achieve these goals and to overcome the drawbacks of the state of the art with a monitoring system as described at the beginning, wherein further said remote processing means comprise at least one artificial intelligence unit configured for the analysis of the received signals and for the discrimination of voluntary movements of the device performed by a user and accidental ones.

The use of an artificial intelligence program to analyze the combined reading of the two sensors allows for the accurate reconstruction of the dynamics of the event that potentially damaged the mobile device, distinguishing between accidental falls and attempts to purposely damage the product.

In an exemplary embodiment, said remote processing means are configured to generate a probability value of the type of damage suffered by the mobile device.

This allows you to have an indication of the type of impact and the damage suffered by the mobile device.

In a further executive example, said remote processing means are configured to generate an estimate of the extent of damage suffered by the mobile device.

In this way, once the type of damage suffered by the mobile device has been indicated, it is possible to provide information about the extent of the damage.

In one embodiment, said remote processing means are provided with one or more storage units for the received data and the processed data.

The data received and the results obtained can be stored in this way for subsequent sending, consultation or further processing.

In a further embodiment, said remote processing means are provided with means of notification of the received data and/or the processed data to third-party users.

It is possible in this way to automatically generate notifications, for example to an insurance company.

The present invention also provides a method of monitoring a mobile device for detecting impact damage, which mobile device comprises at least one gyroscope and at least one accelerometer.

The method involves the following steps:

a) detection of the signals generated by the said gyroscope;

b) detection of the signals generated by the said accelerometer;

c) applying a timestamp to such signals;

d) sending signals to a remote service; e) analysis of said signals by said remote service using artificial intelligence algorithms;

f) discrimination between voluntary device movements performed by a user and accidental ones;

g) generating an impact damage assessment based on the said discrimination.

In one embodiment, the remote service generates a probability value of the type of damage suffered by the mobile device.

According to an executive example, the remote service generates an estimate of the extent of damage suffered by the mobile device.

In a further example of implementation, the remote service stores the received data and the processed data.

According to one embodiment remote service sends a notification to third party users.

These and other features and advantages of the present invention will become clearer from the following description of some non-limiting executive examples illustrated in the attached drawings in which:

Fig. 1 illustrates a block diagram of the system according to the present invention;

Fig. 2 shows a schematic representation of the method and management flows of the system which is the object of the present invention.

Figure 1 illustrates a schematic view of an executive example of the monitoring system of a mobile device 1, in particular a smartphone, for the detection of impact damage according to the present invention. The smartphone 1 comprises at least one gyroscope 10, at least one accelerometer 11, connection means 2 to a communication network, preferably the Internet, and processing means for the execution of a logic program 12 configured to detect the signals emitted by the gyroscope 10 and/or the accelerometer 11, apply a time stamp to them and send them via the connection means 2 to remote processing means 3.

The processing means 3 may be for example a cloud service or a remote server or a remote network.

The remote processing means 3 comprise at least one artificial intelligence unit 30 which constitutes a machine learning engine and is configured to execute artificial intelligence algorithms. The artificial intelligence unit preferably comprises one or more neural networks. The artificial intelligence algorithms perform an analysis of the received signals and a discrimination of voluntary movements of the smartphone 1 performed by a user and accidental ones.

The remote processing means 3 are configured to generate a probability value of the type of damage suffered by the smartphone 1.

The remote processing means 3 are configured to generate an estimate of the extent of damage suffered by the smartphone 1.

The remote processing means 3 are provided with one or more storage units for the received data and the processed data 31, in which a database of reconstructed events can be stored.

Remote processing facilities 3 are provided with means of notification of received data and/or processed data to third-party users.

When the detected event is within a set of parameters, the related data is transmitted, together with a "time-stamp" that certifies the date and time of the accident, to a cloud platform 3 that stores them and generates an alarm for the insurance service manager. The latter will be able to use the data received for easy and efficient claims management.

Figure 2 schematically illustrates the method and management flows of the system which is the object of the present invention.

The system and method are based on the installation of a native smartphone app 4 that is executed by the half-processors 12 and that, using the services offered by the smartphone operating system 1, continuously receives the data generated by the accelerometer sensor 11 and the gyroscopic sensor 10 that the majority of smartphones on the market are equipped with.

The app 4 can be installed on smartphone 1 when signing the insurance contract and advantageously performs a check on the integrity and functionality of smartphone 1 as soon as it is installed.

App 4 may also have user account and active contract management features, allowing for the signing of new contracts, payment and renewal of contracts, opening of claims and similar operations.

A flowchart of the method is illustrated in Figure 2. The machine learning engine 30 trains an AI-based diagnostic system that receives event notifications from the app 40 installed on the smartphone 1.

The AI-based diagnostic system allows to quantify the type and extent of potential damage suffered (for example: broken screen, broken body, etc.). To do this, the AI diagnostic system relies on training neural networks on data relating to events whose dynamics are known, to learn to reconstruct the nature of the event.

The insurance service manager can thus receive notification messages automatically generated when conditions pre-established by the method occur, through a combination of rules set by the system that take into account different parameters.

A first parameter concerns the extent of the decelerations suffered during the fall.

A second parameter concerns an estimate of the presumed point of impact of smartphone 1 with the ground.

A third parameter concerns the accelerations and movements preceding the impact.

An event received by the AI-based diagnostic system 30 from the app 40 is first analyzed in a first analysis block 50, which evaluates whether the event is indicative of a probable breakage 52 or whether it is a false positive 51. In the case of a probable breakage, the event is forwarded to a second analysis block 53, in which it is discriminated whether it is accidental damage 54 or intentionally caused damage 55.

In the event of damage caused intentionally, it is likely that the malicious user will not be able to produce the desired effect at the first attempt and will therefore try to repeat the operation several times. The time elapsed between one impact and the next is measured to distinguish this behavior pattern from that of, for example, an accidental fall, in which several impacts occur but in this case in very close time frames. It is therefore possible to set a time threshold between the recorded impacts beyond which it is unlikely that these impacts can be traced back to a single accidental fall.

In the event of a fall, the accelerometer reading detects the acceleration caused by the fall before the impact and only subsequently the impact. This is clearly different from damage caused by a blunt object on the stationary device, and can therefore be easily discriminated.

Preferably, the AI algorithm is initially calibrated by experimental tests. In particular, the amount of deceleration and the point of impact are initially set by experimental tests. At a later stage and following a sufficient number of AI processing of real accidental falls to smartphones, the system optimizes the calibration based on the observation of the actual correlation between accelerometer readings and actual damage.

From the above it is therefore evident that the invention is not limited to the embodiments described and illustrated herein by way of non-limiting examples, but may be varied and modified, as a whole and in its individual details, especially in terms of construction, according to the specific needs and conveniences of manufacture and use, within the scope of technical and functional equivalents, without thereby abandoning the guiding principle set out above and claimed below.

Claims (10)

1. Monitoring system for a mobile device (1) for detecting impact damage, which mobile device (1) comprises at least one gyroscope (10), at least one accelerometer (11), means of connection (2) to a communication network and processing means for executing a logic program (12) configured to detect the signals emitted by said gyroscope (10) and/or said accelerometer (11), apply a time stamp thereto and send them via said connection means (2) to remote processing means (3), characterised in that the said remote processing means (3) comprise at least one artificial intelligence unit (30) configured for the analysis of the received signals and for the discrimination of voluntary movements of the device performed by a user and accidental ones (1). 2. System according to claim 1, wherein said remote processing means (3) are configured to generate a probability value of the type of damage suffered by the mobile device (1). 3. System according to claim 2, wherein said remote processing means (3) are configured to generate an estimate of the extent of damage suffered by the mobile device (1). 4. System according to one or more of the preceding claims, wherein said remote processing means (3) are provided with one or more storage units for the received data and the processed data (31). 5. System according to one or more of the preceding claims, wherein said remote processing means (3) are provided with means for notifying third-party users of the received data and/or the processed data. 6. A method of monitoring a mobile device (1) for detecting impact damage, which mobile device comprises at least one gyroscope (10) and at least one accelerometer (11), characterised in that It involves the following steps: a) detection of the signals generated by the said gyroscope (10); b) detection of the signals generated by the said accelerometer (11); c) applying a timestamp to such signals; d) sending signals to a remote service; e) analysis of said signals by said remote service using artificial intelligence algorithms; f) discrimination of device movements (1) voluntary performed by a user and accidental; g) generation of an impact damage assessment based on said discrimination. 7. Method according to claim 6, wherein the remote service generates a probability value of the type of damage suffered by the mobile device (1). 8. System according to claim 7, wherein the remote service generates an estimate of the extent of damage suffered by the mobile device (1). 9. System according to one or more of the preceding claims 6 to 8, wherein the remote service stores the received data and the processed data. 10. System according to one or more of the preceding claims 6 to 9, wherein the remote service sends a notification to third-party users.