DE102024102818B4 - METHOD FOR DETECTING MECHANICAL IMPACTS ON A VEHICLE - Google Patents

Abstract

In a method for detecting mechanical impacts (10) on a vehicle (1), sensor data are acquired by means of a detection device (2), wherein the sensor data represent an impact (3) on the detection device (2) caused by a mechanical impact acting from outside on an outer shell of the parked vehicle (1). The sensor data comprise effective components of the impact on the detection device (2) in various directions of three-dimensional space. The acquired sensor data are processed, wherein each effective component is represented by a time-dependent function of a damped oscillation, from which an acceleration function is derived in each case. A start time of the acceleration functions corresponds to a time at which the mechanical impact on the vehicle (1) occurred. The acceleration functions are evaluated at the start time in order to obtain at least one feature of a force of the impact (3) on the detection device (2), the at least one obtained feature of the force of the impact (3) is processed by means of an evaluation model in order to obtain at least one feature of the mechanical action (10) on the vehicle (1), and an evaluation result is output which represents the at least one feature of the mechanical action (10) on the vehicle (1).

Description

The present invention relates to a method and system for detecting mechanical impacts on a vehicle, in particular on the outer shell of a parked vehicle, as well as to a module for a vehicle equipped with such a system.

Parking a vehicle in public areas or parking garages can occasionally be associated with untoward incidents that can result in damage to the vehicle's exterior, such as dents in the sheet metal or scratches in the paint. A common problem is other vehicles that strike or ram the parked vehicle when entering or leaving a parking space. Parking into tight spaces can further increase the risk of damage. In supermarket parking lots, rolling shopping carts that are not properly parked in the collection points can strike and damage parked vehicles. Careless handling of shopping carts can also lead to a collision with a stationary vehicle. People can accidentally strike a parked vehicle, especially if they are distracted or the vehicle is parked in a poorly lit area. In addition, vehicles can also be intentionally damaged. Depending on the location, natural causes such as falling branches, hail, or storm damage can also cause damage to the vehicle's exterior.

When such an incident occurs, it often goes undetected at first. Often, the person responsible walks away from the damaged vehicle without reporting it or leaving a message. They may not even have noticed that the parked vehicle was damaged. If the damage was caused without the involvement of another person or vehicle, the vehicle owner is responsible for discovering the damage themselves. However, a driver will not always inspect their vehicle from the outside before getting in and driving off.

State of the art:

WO 2006/074672 A1 This document describes a method for controlling a safety system in a vehicle. The output signal from at least one impact sensor is subjected to frequency analysis. The signal spectrum is compared with reference patterns, taking into account the current vehicle speed and/or the outside temperature. If an impact with a person to be protected (in particular, pedestrians and/or cyclists) is detected with a specified probability, the control unit outputs a signal to trigger protective measures.

DE 10 2021 209 257 A1 This document proposes a method for detecting damage to the exterior of a vehicle. Damage is divided into two groups (greater severity and lesser severity). The vehicle's acceleration and/or yaw rate are measured using an acceleration sensor and/or yaw rate sensor. Damage in the first group (greater severity) is detected when the acceleration or yaw rate exceeds a threshold. Damage in the second group (lesser severity) is detected by comparing the measured values with values learned for the specific vehicle type. This enables reliable detection and localization of even minor damage.

It is an object of the present invention to provide a solution for detecting mechanical effects on a vehicle, in particular on the outer shell of a parked vehicle.

This object is achieved according to the teaching of the independent claims. Various embodiments and further developments of the invention are the subject of the dependent claims.

A first aspect of the solution presented here relates to a method, in particular a computer-implemented method, for detecting mechanical impacts on a vehicle, wherein the vehicle has at least one detection device configured to detect accelerations in three-dimensional space. In the method, sensor data are detected by means of the detection device, wherein the sensor data represent an impact on the detection device caused by a mechanical impact from outside on an outer shell of the parked vehicle, wherein the sensor data comprise effective components of the impact on the detection device in different directions of the three-dimensional space. The detected sensor data are processed, wherein each effective component is represented by a time-dependent function of a damped oscillation, from which an acceleration function is derived, wherein a start time of the acceleration functions corresponds to a point in time at which the mechanical impact on the vehicle occurred. The acceleration functions are evaluated at the starting point in time to obtain at least one characteristic of a force of the impact on the detection device, and the at least one obtained characteristic of the force of the impact is processed using an evaluation model to obtain at least one characteristic of the mechanical impact on the vehicle. An evaluation result is then output, which represents the at least one characteristic of the mechanical impact on the vehicle.

The method according to the first aspect is therefore based, in particular, on using sensor data from a detection device to infer mechanical impacts on the vehicle. A mechanical impact on the outer shell of a parked vehicle, for example, caused by another parking vehicle, can be detected in this way. The mechanical impact is not measured directly at the point where it impacts the vehicle. Rather, indirect detection occurs because the mechanical impact propagates through the body and other structures of the vehicle, thereby causing an impact on the detection device, which may, for example, be located centrally in the vehicle.

Such a detection device, which may in particular comprise an acceleration sensor, is often already installed in the vehicle. For example, it may be an inertial measurement unit (IMU) installed in many vehicles, allowing the method to be implemented at low cost. The method is also robust because it can be applied to various vehicles, provided they are equipped with a corresponding detection device. Since the amount of sensor data from the detection device is relatively small, this method can also provide an efficient method, particularly compared to a method in which a vehicle would have to be equipped with a large number of sensors distributed across the outer shell.

As used herein, the term "vehicle" refers specifically to a passenger car, including all types of motor vehicles, hybrid and battery-powered electric vehicles, as well as vehicles such as sedans, vans, buses, trucks, delivery vans, and the like. A vehicle may be described as "parked" specifically when it is stationary and its engine is switched off.

The term "outer shell" of a vehicle as used here refers specifically to the part of the vehicle's body that constitutes the exterior of the vehicle and is therefore accessible from the outside and potentially susceptible to external mechanical influences. This outer part of the body, also called the outer skin, includes, for example, bumpers, fenders, hood, tailgate, side panels, doors, the A-, B-, and C-pillars, and the roof. The outer shell can also include other parts of the vehicle's exterior, such as exterior mirrors, headlights, flaps, panels, aprons, convertible top, etc., as well as window panes, which may be exposed to external mechanical influences.

The term "mechanical impact" used here refers specifically to an external impact on the vehicle's outer shell that is of a mechanical nature. This can be, for example, a shock, a collision, a blow, an impact, a collision, a jolt, and the like. The mechanical impact can be caused intentionally or unintentionally by another vehicle, a person, an object, or the like.

The term "evaluation model" used here refers to a model, particularly a mathematical one, that uses the characteristics of the impact force on the detection device as input variable(s) to provide an evaluation result based thereon, in this case at least one characteristic of a mechanical impact on a vehicle. The evaluation model can in particular be a "machine learning model" (or equivalently "machine learning model" (ML model)), which here is understood to mean in particular a mathematical, particularly statistical, model for making predictions or decisions created by means of at least one machine learning algorithm on the basis of example data referred to as training data, without the algorithm(s) being explicitly programmed to make such predictions or decisions. In particular, decision tree-based machine learning models (English "decision trees") or artificial neural networks are machine learning models.

The terms "comprises,""includes,""includes,""has,""has,""with," or any other variation thereof, where used, are intended to cover non-exclusive inclusion. For example, a method or apparatus that comprises or has a list of elements is not necessarily limited to these elements, but may include other elements not expressly listed or which are inherent in such a method or apparatus.

Furthermore, unless explicitly stated to the contrary, "or" refers to an inclusive "or" and not an exclusive "or." For example, a condition A or B is satisfied by one of the following conditions: A is true (or present) and B is false (or absent), A is false (or absent) and B is true (or present), and both A and B are true (or present).

The terms "a" or "an" as used herein are defined as "one or more." The terms "another" and "another," and any other variations thereof, are defined as "at least one other."

The term “plurality” or “several” as used here shall be understood to mean “two or more”.

The terms "configured" or "set up" to perform a specific function (and respective variations thereof) are to be understood within the meaning of the invention as meaning that the corresponding device is already in a configuration or setting in which it can perform the function or is at least adjustable - i.e. configurable - so that it can perform the function after being set accordingly. The configuration can be carried out, for example, by appropriately setting parameters of a process sequence or of switches or the like for activating or deactivating functionalities or settings. In particular, the device can have a plurality of predetermined configurations or operating modes, so that configuration can be carried out by selecting one of these configurations or operating modes.

In the following, various exemplary embodiments of the method are described, each of which, unless expressly excluded or technically impossible, can be combined with each other as well as with the other aspects of the present solution described.

In some embodiments, a Fast Fourier Transform (FFT) is applied to the acceleration components to obtain a frequency spectrum for each acceleration component, with the acceleration components being evaluated only for frequencies where the amplitude exceeds a threshold value. By applying an FFT, the frequency components of the vibration or acceleration functions can be calculated. In this way, low-frequency components can be ignored, which, for example, could complicate or distort the detection of a mechanical impact as background noise. By appropriate filtering, the components that are more likely to originate from a mechanical impact on the outer shell of the parked vehicle can be extracted for further processing. Since the method can detect mechanical impacts on a parked vehicle in particular, further disturbances, such as vibrations due to a running engine, are minimized anyway. In other words, the sensor data at the detection device for the parked vehicle are most likely caused primarily by an external mechanical impact on the vehicle.

In one embodiment of the invention, the at least one characteristic of the force of the impact comprises a direction and a magnitude of the force. The direction and magnitude of the force of the impact on the detection device ("impact force") are characterizing features suitable for deriving a mechanical effect on the vehicle that is the cause of this impact force.

In some embodiments, the direction of the impact force is represented by two angles obtained by angular operations on the acceleration components at the initial time. In this way, the direction of the impact force can be easily characterized. The direction of the force can be described, for example, by a first angle in a first plane, such as an x-y plane, and a second angle to this plane. These angles generally represent a portion of spherical coordinates (spatial polar coordinates). The magnitude of the impact force can then also be represented in spherical coordinates.

In some embodiments, the magnitude of the impact force is determined based on an oscillating mass and the acceleration components at the starting time. The oscillating mass is part of a damped oscillation function, which, as explained above, forms the basis for calculating the acceleration components and represents a tool for describing the impact force on the detection device. The oscillating mass is appropriately selected as a parameter for the damped oscillation function and can then be used to easily calculate the magnitude of the impact force.

In some embodiments, the at least one feature of the mechanical impact on the vehicle comprises a location on the vehicle at which the mechanical impact on the vehicle occurred. Depending on the mechanical impact on the vehicle, more precisely the external mechanical impact on the outer shell of the parked vehicle, corresponding sensor data is generated at the detection device, from which an impact force can be derived as described. Advantageously, a location on the vehicle or on the outer shell of the vehicle at which the mechanical impact on the vehicle occurred can be determined in this way. This can be an indication of a precise point of impact or at least an area of the outer shell on which the mechanical impact acted, such as the driver's door, the tailgate, a specific fender, etc. This makes it easier for a user to locate and inspect any damage.

In some embodiments, the at least one feature of the mechanical impact on the vehicle includes an amount of force of the mechanical impact. The sensor data of the detection device or the impact force determined therefrom can also provide an indication of the amount of force of the mechanical impact, i.e., how strong the mechanical impact was. Thus, based on this information, a user can determine whether damage to their vehicle is to be expected and, if applicable, how severe this damage might be.

In some embodiments, based on the evaluation result, a message is sent to a user's terminal device, which message contains at least a statement that a mechanical impact has occurred on the vehicle. In other words, as soon as a mechanical impact is detected, the user can first be informed that it has occurred at all. It goes without saying that further details about the mechanical impact can then also be communicated, for example where on the vehicle the impact occurred and how strong it was, as also explained below. Alternatively or additionally, a message can also be sent to another location, such as a reporting center, a service center or the like, so that appropriate measures can be initiated, for example dispatching a security service to the vehicle or the like.

In some embodiments, the notification is only sent to the user's terminal if the magnitude of the impact force exceeds a predefined threshold. For example, the user will not be notified if, for example, a mechanical impact on the vehicle has occurred, but it is so minor that damage is highly unlikely. The user will only be notified if more significant mechanical impacts occur. This can be any type of notification, alarm, or the like. For example, a notification may be sent that the driver's door has been impacted. The user can then directly check the damage to their vehicle.

In some embodiments, the evaluation model represents a correlation between the mechanical impact on the vehicle and the impact on the detection device. By providing such a correlation, a mechanical impact on the vehicle can be inferred from a specific impact on the detection device. It should be noted that the impact on the detection device, particularly with regard to its magnitude and direction, does not necessarily have to be the same as the mechanical impact on the vehicle due to the complex vehicle structure. This means that an impact on the vehicle from one side can manifest itself as an impact on the detection device from a completely different side, since the vehicle structure, such as body parts, connecting parts, suspensions, cladding, etc., can redirect the forces in the vehicle. However, the correlation in the evaluation model makes it possible to infer the mechanical impact on the vehicle from the sensor data of the detection device, which represent the impact on the sensor device.

In some embodiments, the evaluation model includes a plurality of assignments based on empirically determined assignments, wherein a plurality of specific mechanical Impacts on the vehicle are each assigned to a corresponding resulting impact on the detection device. In this way, a simple evaluation model can be provided. While it is conceivable in principle to calculate a mechanical impact on the vehicle from the sensor data, this is a very complex task due to the complex structure of a vehicle. The evaluation model can be trained using a large number of assignments. Targeted mechanical impacts on the vehicle, or more precisely its outer shell, can be applied and the associated sensor data from the detection device can be recorded. In this way, a large number of assignments can be provided on the basis of which an evaluation model can be created and trained. An evaluation model based on machine learning can then be able to derive other mechanical impacts from the sensor data from the detection device.

A second aspect of the present solution relates to a system for detecting mechanical impacts on a vehicle. The system comprises a detection device for detecting accelerations and a data processing device configured, in particular by means of a corresponding computer program, to carry out the method according to the first aspect for detecting mechanical impacts on the vehicle.

A third aspect of the present solution relates to a computer program or computer program product comprising instructions which, when executed on the data processing device of the system according to the second aspect, cause the system to carry out the method according to the first aspect.

The computer program can, in particular, be stored on a non-volatile data storage medium. This is preferably a data storage medium in the form of an optical data storage medium or a flash memory module. This can be advantageous if the computer program as such is to be handled independently of a processor platform on which the one or more programs are to be executed. In another implementation, the computer program can be present as a file on a data processing unit, in particular on a server, and can be downloaded via a data connection, for example the Internet or a dedicated data connection, such as a proprietary or local network. Furthermore, the computer program can have a plurality of interacting individual program modules. The modules can, in particular, be configured or at least usable in such a way that they are executed on different devices (computers or processor units) that are geographically spaced from one another and connected to one another via a data network, in the sense of distributed computing.

The system according to the second aspect can accordingly comprise a program memory in which the computer program is stored. Alternatively, the system can also be configured to access an external computer program, for example, available on one or more servers or other data processing units, via a communication connection, in particular to exchange data with it that is used during the execution of the method or computer program or that represents outputs of the computer program.

A fourth aspect of the present solution relates to a module for a vehicle equipped with a system according to the second aspect, wherein the module is configured to be installed in a vehicle. The module can in particular be a roof module (also called a roof control unit) or at least a part thereof, which can be configured to be installed in a roof lining of a vehicle, for example, near an interior mirror. The roof module can have further elements, such as lights, sensors for interior monitoring, or control elements that can be operated by the driver or passenger of a vehicle. In particular, an inertial measurement unit (IMU; see below) can be included in the module. The position near the center of the vehicle can be advantageous for the accuracy of measurements of the overall movements of the vehicle.

Modern vehicles can be equipped with a variety of detection devices (sensors) that help increase safety, improve driving comfort, and increase efficiency. These include a series of external sensors for detecting the vehicle's surroundings. Internal sensors, on the other hand, are capable of measuring certain parameters related to the vehicle itself. This includes, for example, the so-called inertial measurement unit (IMU). This measurement unit can, in particular, have an acceleration sensor (accelerometer) and, if necessary, a gyroscope sensor for measuring acceleration or angular rates. This allows the orientation, speed, and position of the vehicle to be tracked in real time, which in It can be used in various applications, from electronic stability control to driver assistance systems. The IMU can be used as a detection device in a system according to the second aspect to capture signals caused by a mechanical impact on the vehicle. Of particular interest here are the sensor data from the acceleration sensor, i.e., data relating to translational movements. The method according to the first aspect allows for the evaluation of the IMU signals, so that they can be utilized to detect mechanical impacts on the vehicle.

A fifth aspect of the present solution relates to a vehicle with a module according to the fourth aspect, which is arranged in particular as a roof module in a roof area, for example in the roof lining of the vehicle.

The features and advantages explained with regard to the first aspect of this solution also apply accordingly to the other aspects of the solution.

Further advantages, features and possible applications of this solution will become apparent from the following detailed description in conjunction with the drawings.

• 1 eine Ansicht eines geparkten Fahrzeugs mit einer Beschädigung;
• 2 schematisch eine beispielhafte Ausführungsform eines Verfahrens zum Erkennen von mechanischen Einwirkungen auf die Außenhülle eines geparkten Fahrzeugs;
• 3 eine beispielhafte gedämpfte Schwingungsfunktion; und
• 4 eine beispielhafte Darstellung einer resultierenden Stoßkraft auf die Detektionseinrichtung.

It shows:

• 1 a view of a parked vehicle with damage;
• 2 schematically shows an exemplary embodiment of a method for detecting mechanical impacts on the outer shell of a parked vehicle;
• 3 an exemplary damped oscillation function; and
• 4 an exemplary representation of a resulting impact force on the detection device.

In the figures, like reference numerals designate like, similar, or corresponding elements. Elements shown in the figures are not necessarily drawn to scale. Rather, the various elements shown in the figures are depicted in such a way that their function and general purpose will be understood by those skilled in the art. Connections and couplings between functional units and elements shown in the figures can also be implemented as indirect connections or couplings, unless expressly stated otherwise. Functional units can, in particular, be implemented as hardware, software, or a combination of hardware and software.

1 shows, by way of example, a vehicle 1, wherein a mechanical impact 10 occurs at an impact point 11 on the passenger door. Such a mechanical impact on the parked vehicle can be caused by colliding objects that hit the vehicle in question, such as another parking vehicle. With the solution presented here, mechanical impacts can be detected, using a simple, inexpensive sensor that is present in many vehicles, for example in the form of an inertial measurement unit (IMU). The detection device 2 used can therefore in particular comprise an acceleration sensor, which can be designed as a microelectronic mechanical system (MEMS) and delivers acceleration signals in three dimensions. The IMU 2 can be installed at a central location in the vehicle 1, for example as a roof module in the roof lining of the vehicle, as in 1 shown.

In 2 An overview of the process for detecting a mechanical impact on a vehicle is shown. The method is initiated by a mechanical impact 10 on the vehicle 1 (step 101), which occurs from outside on the outer shell of the vehicle 1, for example, by another object colliding with the vehicle 1. Sensor data, in particular acceleration data, are read from the detection device 2 (step 102) and subjected to an FFT (step 103) in order to calculate the frequency spectrum. In a next step 104, amplitude A and phase P are then filtered according to a specific frequency band and amplitude threshold value, for example, to filter out noise. From this, a force of an impact 3 on the detection device 2 is then calculated (step 105). From this, the magnitude of the mechanical impact 10 and the location 11 on the vehicle 1 at which the mechanical impact 10 impacted the vehicle 1 can then be estimated (step 106). For this purpose, an evaluation model can be used, which can be a machine learning model. The recognition is then completed (step 107), and a corresponding message can then be sent to a user, for example, to a mobile device. like a smartphone. The individual steps, especially those for calculating the impact force, are explained in more detail below. Finally, a possible way to generate a corresponding evaluation model is also explained.

Depending on the impact impulse of the mechanical action 10 on the vehicle 1, a corresponding impact with a certain impact force 3 on the sensor 2 is to be expected, whereby the resulting acceleration signal can be a damped, oscillating signal, as in 3 The damped oscillating signal is characterized by a number of parameters, in particular positive peak value (PP), negative peak value (NP), peak-to-peak amplitude (Amp), period time (t _P ) and decay time (t _d ).

The following is an example of a mathematical derivation to determine the formula for the damped vibration pulses that could be generated from the acceleration signals. The measured accelerations result from the impact on the detection device, which in turn was caused by the mechanical impact on the vehicle. The reaction to the impact on the detection device 2, which, as mentioned, can contain a MEMS accelerometer, can be formulated as a one-dimensional damped vibration in any spatial dimension. Newton's law of motion for a corresponding vibration system with a moving body suspended between two springs is (e.g., in the x-direction): $$F-M\ddot{x}=\left(K_1+K_2\right)x+B\dot{x}$$ $$F=\left(K_1+K_2\right)x+B\dot{x}+M\ddot{x}$$ where K ₁ and K ₂ are the stiffness coefficients of the springs of the suspension system, B is the friction coefficient and M is the mass of the moving body.

Let the applied impulse force be F=A . δ(t), where A is a specific constant value and δ(t) is the unit impulse. $$A.\delta \left(t\right)=\left(K_1+K_2\right)x+B\dot{x}+M\ddot{x}$$

A Laplace transformation is applied, assuming the initial conditions to be zero: $$A=\left(K_1+K_2\right)X\left(s\right)+BSX\left(s\right)+M{S}^{2}X\left(s\right)$$ $$x\left(t=0\right)=0$$ $$\dot{x}\left(t=0\right)=0$$

With the following features of the Laplace transform: $$\mathcal{L}\left\{f\left(t\right)\right\}={\displaystyle \underset{0}{\overset{\infty }{\int }}f\left(t\right)e^{-st}dt=\mathcal{F}\left(s\right)}$$ $$\mathcal{L}\left\{\delta \left(t\right))\right\}={\displaystyle \underset{0}{\overset{\infty }{\int }}\delta \left(t\right)e^{-st}dt}=e^0=1$$ $$\mathcal{L}\left\{\dot{f}\left(t\right)\right\}={{\displaystyle \underset{0}{\overset{\infty }{\int }}\dot{f}\left(t\right)e^{-st}dt=f\left(t\right)}{e}^{-st}|}_0^{\infty }+s{\displaystyle \underset{0}{\overset{\infty }{\int }}f\left(t\right)e^{-st}dt}=s\mathcal{F}\left(s\right)-f\left(t=0\right)$$ $$X\left(s\right)=\frac{A}{M{S}^2+BS+\left(K_1+K_2\right)}$$ $$X\left(s\right)=\frac{A/M}{S^2+\frac{B}{M}S+\frac{\left(K_1+K_2\right)}{M}}$$

Now be $$a_1=A/M$$ $$a_2=B/M$$ $$a_3=\left(K_1+K_2\right)/M$$ then: $$X\left(s\right)=\frac{a_1}{S^2+a_{2}S+a_3}=\frac{a_1}{{\left(S+\frac{a_2}{2}\right)}^2+\left(a_3-\frac{a_2{}^2}{4}\right)}$$ $$\mathcal{L}\left\{\text{sin}\left(at\right)\right\}={{\displaystyle \underset{0}{\overset{\infty }{\int }}\text{sin}}\left(at\right)e^{-st}dt\left[-\frac{\text{cos}\left(at\right)}{a}\left(e^{-st}\right)\right]|}_0^{\infty }+{\displaystyle \underset{0}{\overset{\infty }{\int }}\frac{\text{cos}\left(at\right)}{a}\left(-s{e}^{-st}\right)dt}$$ $$\begin{array}{c}\mathcal{L}\left\{\text{sin}\left(at\right)\right\}=\left(\frac{1}{a}-0\right)-\frac{s}{a}{\displaystyle \underset{0}{\overset{\infty }{\int }}\text{cos}\left(at\right)\left(e^{-st}\right)dt}\\ {=\frac{1}{a}-\frac{s}{a}\left[\frac{\text{sin}\left(at\right)}{a}\left(e^{-st}\right)\right]|}_0^{\infty }+\frac{s}{a}{\displaystyle \underset{0}{\overset{\infty }{\int }}\frac{\text{sin}\left(at\right)}{a}\left(-s{e}^{-st}\right)dt}\\ =\frac{1}{a}-\frac{s^2}{a^2}{\displaystyle \underset{0}{\overset{\infty }{\int }}\text{sin}\left(at\right)\left(e^{-st}\right)dt}=\frac{1}{a}-\frac{s^2}{a^2}\mathcal{L}\left\{\text{sin}\left(at\right)\right\}\end{array}$$ $$$$ $$\mathcal{L}\left\{\text{sin}\left(at\right)\right\}=\frac{1}{a}-\frac{s^2}{a^2}\mathcal{L}\left\{\text{sin}\left(at\right)\right\}$$ $$\mathcal{L}\left\{\text{sin}\left(at\right)\right\}=\frac{a}{s^2+a^2}$$ $$\mathcal{L}\left\{f\left(t\right)e^{-at}\right\}={\displaystyle \underset{0}{\overset{\infty }{\int }}f\left(t\right)}{e}^{-at}{e}^{-st}dt={\displaystyle \underset{0}{\overset{\infty }{\int }}f\left(t\right)e^{-\left(s+a\right)t}dt=\mathcal{F}}\left(s+a\right)$$ $$X\left(s\right)=\left(\frac{a_1}{\sqrt{a_3-\frac{a_2{}^2}{4}}}\right)\left(\frac{\sqrt{a_3-\frac{a_2{}^2}{4}}}{{\left(S+\frac{a_2}{2}\right)}^2+\left(a_3-\frac{a_2{}^2}{4}\right)}\right)$$

By applying the two derivation rules: $$\mathcal{L}\left\{\text{sin}\left(at\right)\right\}=\frac{a}{s^2+a^2}$$ $$\mathcal{L}\left\{f\left(t\right)e^{-at}\right\}=\mathcal{F}\left(s+a\right)$$

Then x(t) can be derived as follows: $$x\left(t\right)=\left(\frac{a_1}{\sqrt{a_3-\frac{a_2{}^2}{4}}}\right)\left(e^{-\left(\frac{a_2}{2}\right)t}\right)\left(\text{sin}\left(\left(\sqrt{a_3-\frac{a_2{}^2}{4}}\right)t\right)\right)$$

Thus, x(t) can be formulated as a damped oscillating signal whose physical parameters can be represented as follows. $$x\left(t\right)=K\left(e^{-Dt}\right)\left(\text{sin}\left(\left(\frac{2\pi }{T}\right)t\right)\right)$$

K is the initial displacement amount $$K=\left(\frac{a_1}{\sqrt{a_3-\frac{a_2{}^2}{4}}}\right)$$

D is the damping exponent $$D=\left(\frac{a_2}{2}\right)$$

T is the periodic time $$T=\frac{2\pi }{\sqrt{a_3-\frac{a_2{}^2}{4}}}$$

By substitution with the basic physical parameters, the parameters of the damped oscillating signal are represented in the form of the basic physical parameters of the mechanical system: $$a_1=A/M$$ $$a_2=B/M$$ $$a_3=\left(K_1+K_2\right)/M$$ $$K=\left(\frac{A}{\sqrt{\left(K_1+K_2\right)M-\frac{B^2}{4}}}\right)$$ $$D=\left(\frac{B}{2M}\right)$$ $$T=\left(\frac{2\pi M}{\sqrt{\left(K_1+K_2\right)M-\frac{B^2}{4}}}\right)$$

The derivatives are then calculated to obtain the formulas for the acceleration signal: $$x\left(t\right)=K\left(e^{-Dt}\right)\left(\text{sin}\left(\left(\frac{2\pi }{T}\right)t\right)\right)$$ $$\dot{x}\left(t\right)=K\left(\frac{2\pi }{T}\right)\left(e^{-Dt}\right)\left(\text{cos}\left(\left(\frac{2\pi }{T}\right)t\right)\right)-KD\left(e^{-Dt}\right)\left(\text{sin}\left(\left(\frac{2\pi }{T}\right)t\right)\right)$$ $$\begin{array}{l}\ddot{x}\left(t\right)=-K\left(\frac{2\pi }{T}\right)\left(e^{-Dt}\right)\left(\text{sin}\left(\left(\frac{2\pi }{T}\right)t\right)\right)-KT\left(\frac{2\pi }{T}\right)\left(e^{-Dt}\right)\left(\text{cos}\left(\left(\frac{2\pi }{T}\right)t\right)\right)\\ -KD\left(\frac{2\pi }{T}\right)\left(e^{-Dt}\right)\left(\text{cos}\left(\left(\frac{2\pi }{T}\right)t\right)\right)+K\left(D^2\right)\left(e^{-Dt}\right)\left(\text{sin}\left(\left(\frac{2\pi }{T}\right)t\right)\right)\end{array}$$ $$\ddot{x}\left(t\right)=K\left(D^2-{\left(\frac{2\pi }{T}\right)}^2\right)\left(e^{-Dt}\right)\left(\text{sin}\left(\left(\frac{2\pi }{T}\right)t\right)\right)-2KD\left(\frac{2\pi }{T}\right)\left(e^{-Dt}\right)\left(\text{cos}\left(\left(\frac{2\pi }{T}\right)t\right)\right)$$ $$\ddot{x}\left(t\right)=K\left(e^{-Dt}\right)\left[\left(D^2-{\left(\frac{2\pi }{T}\right)}^2\right)\left(\text{sin}\left(\left(\frac{2\pi }{T}\right)t\right)\right)-2D\left(\frac{2\pi }{T}\right)\left(\text{cos}\left(\left(\frac{2\pi }{T}\right)t\right)\right)\right]$$ $$\ddot{x}\left(t\right)=Amp\times \left[\text{sin}\left(\left(\frac{2\pi }{T}\right)t-\phi \right)\right]\times \left(e^{-Dt}\right)$$ where Amp is the amplitude of the oscillation signal and φ is its phase shift $$\text{tan}\left(\phi \right)=\frac{2D\left(\frac{2\pi }{T}\right)}{D^2-{\left(\frac{2\pi }{T}\right)}^2}$$ $$\begin{array}{l}\sqrt{{\left(D^2-{\left(\frac{2\pi }{T}\right)}^2\right)}^2+{\left(2D\left(\frac{2\pi }{T}\right)\right)}^2}=\sqrt{D^4+{\left(\frac{4{\pi }^2}{T^2}\right)}^2-2D^2\left(\frac{4{\pi }^2}{T^2}\right)+4D^2\left(\frac{4{\pi }^2}{T^2}\right)}\\ =\sqrt{D^4+{\left(\frac{4{\pi }^2}{T^2}\right)}^2-2D^2\left(\frac{4{\pi }^2}{T^2}\right)}=\sqrt{{\left(D^2+\frac{4{\pi }^2}{T^2}\right)}^2}=D^2+\frac{4{\pi }^2}{T^2}\end{array}$$ $$\text{sin}\left(\phi \right)=\frac{2D\left(\frac{2\pi }{T}\right)}{D^2+{\left(\frac{2\pi }{T}\right)}^2}$$ $$\text{cos}\left(\phi \right)=\frac{D^2-\left(\frac{2\pi }{T}\right)}{D^2+{\left(\frac{2\pi }{T}\right)}^2}$$ $$Amp=K\times \left(D^2+{\left(\frac{2\pi }{T}\right)}^2\right)$$

Thus, the resulting acceleration signal from several different input impulse forces (n) can be represented as a summation of damped vibration signals with several different amplitudes (Amp _i ), frequencies (f _i =1/T _i ), phase shifts (φ _i ) and damping coefficients (D _i ) as follows: $$\ddot{x}\left(t\right)={\displaystyle \sum _{i=1}^{\text{n}}\left[Am{p}_i\times \left[\text{sin}\left(\left(\frac{2\pi }{T_i}\right)t-{\phi }_i\right)\right]\times \left(e^{-D_{i}t}\right)\right]}$$ $$\ddot{x}\left(t\right)={\displaystyle \sum _{i=1}^{\text{n}}\left[Am{p}_i\times \left[\text{sin}\left(\left(2\pi f_i\right)t-{\phi }_i\right)\right]\times \left(e^{-D_{i}t}\right)\right]}$$

By performing a Fourier transform to determine the frequency spectrum of the acceleration signal, the signal can be analyzed in each frequency component. Using the fast Fourier transform (FFT), the amplitude (Amp _i ) and phase shift (φ _i ) can be determined at each frequency (f _i ). By filtering out the frequencies in the spectrum with low amplitude components (noise) and selecting only those frequencies with significant amplitudes, the most significant effects of the applied forces can be identified.

The resulting acceleration vector can be formulated based on the set of filtered frequencies for each acceleration vector in the Cartesian 3D coordinates. The set of filtered frequencies contains m, p, and q frequencies for the acceleration components in the x, y, and z directions, respectively. $$\ddot{x}\left(t\right)={\displaystyle \sum _{i=1}^m\left[\left(Am{p}_x{}_{{}_i}\right)\times \left[\text{sin}\left(\left(2\pi f_{x_i}\right)t-{\phi }_{x_i}\right)\right]\times \left(e^{-\left(D_{x_{{}_i}}\right)t}\right)\right]}$$ $$\ddot{y}\left(t\right)={\displaystyle \sum _{j=1}^p\left[\left(Am{p}_y{}_{{}_j}\right)\times \left[\text{sin}\left(\left(2\pi f_{y_j}\right)t-{\phi }_{y_j}\right)\right]\times \left(e^{-\left(D_{y_{{}_j}}\right)t}\right)\right]}$$ $$\ddot{z}\left(t\right)={\displaystyle \sum _{k=1}^q\left[\left(Am{p}_{z_k}\right)\times \left[\text{sin}\left(\left(2\pi f_{z_k}\right)t-{\phi }_{z_x}\right)\right]\times \left(e^{-\left(D_{z_k}\right)t}\right)\right]}$$

The magnitude and direction of the impact force at the detection device can now be determined. Since the starting time is defined as the beginning of the impact forces of the mechanical action on the vehicle, it can be assumed that the impact impulse forces reach their peak values at (t=0). Therefore, the acceleration signals at time (t=0) are analyzed to determine the magnitude and phase of the resulting acceleration and thus obtain an estimate of the direction and magnitude of the impact force, without taking into account the subsequent reflections of the suspension forces, which occur particularly from body parts and other vehicle structures.

By inserting t=0 into the equations for the acceleration signals, we now obtain the following equations: $$a_x=\ddot{x}\left(t=0\right)={\displaystyle \sum _{i=1}^m\left[\left(Am{p}_{x_i}\right)\times \text{sin}\left(-{\phi }_{x_i}\right)\right]}$$ $$a_y=\ddot{y}\left(t=0\right)={\displaystyle \sum _{j=1}^p\left[\left(Am{p}_{y_j}\right)\times \text{sin}\left(-{\phi }_{y_j}\right)\right]}$$ $$a_z=\ddot{z}\left(t=0\right)={\displaystyle \sum _{k=1}^q\left[\left(Am{p}_{z_k}\right)\times \text{sin}\left(-{\phi }_{z_k}\right)\right]}$$

The resulting force vector based on the significant acceleration components at the selected frequencies at t=0 can be generated as follows. $${\stackrel{\rightharpoonup }{F}}_{res}=M\times \stackrel{\rightharpoonup }{Acc}=M\times \left[a_x\stackrel{\rightharpoonup }{i}+a_y\stackrel{\rightharpoonup }{j}+a_z\stackrel{\rightharpoonup }{k}\right]$$ where (i ^⇀ , j ^⇀ and k ^⇀ are the unit vectors in the directions x, y and z respectively.

The magnitude and direction of the impact force are then as follows. $$\Vert {\stackrel{\rightharpoonup }{F}}_{res}\Vert =M\times \Vert \stackrel{\rightharpoonup }{Acc}\Vert =M\times \sqrt{a_x{}^2+a_y{}^2+a_z{}^2}$$ $$Direction\left({\stackrel{\rightharpoonup }{F}}_{res}\right)=\underset{\_}{{\stackrel{\rightharpoonup }{F}}_{res}}=\left\{\alpha ,\beta \right\}$$ where the angles α and β are defined as follows. These are also in 4 illustrated. $$\alpha ={\text{tan}}^{-1}\left(\frac{a_y}{a_x}\right)={\text{tan}}^{-1}\left(\frac{{\displaystyle \sum {}_{j=1}^p\left[\left(Am{p}_{y_j}\right)\times \text{sin}\left(-{\phi }_{y_j}\right)\right]}}{{\displaystyle \sum {}_{k=1}^q\left[\left(Am{p}_{x_i}\right)\times \text{sin}\left(-{\phi }_{x_i}\right)\right]}}\right)$$ $$\alpha ={\text{tan}}^{-1}\left(\frac{{\displaystyle \sum {}_{j=1}^p\left[\left(Am{p}_{y_j}\right)\times \text{cos}\left(\frac{\mathrm{\pi }}{2}+{\phi }_{y_j}\right)\right]}}{{\displaystyle \sum {}_{k=1}^q\left[\left(Am{p}_{x_i}\right)\times \text{cos}\left(\frac{\mathrm{\pi }}{2}+{\phi }_{x_i}\right)\right]}}\right)$$ $$\beta ={\text{tan}}^{-1}\left(\frac{a_z}{\sqrt{a_x{}^2+a_y{}^2}}\right)$$ $$\beta ={\text{tan}}^{-1}\left(\frac{{\displaystyle \sum {}_{k=1}^q\left[\left(Am{p}_{z_k}\right)\times \text{sin}\left(-{\phi }_{z_k}\right)\right]}}{\sqrt{{\left({\displaystyle \sum {}_{i=1}^m\left[\left(Am{p}_{x_i}\right)\times \text{sin}\left(-{\phi }_{x_i}\right)\right]}\right)}^2+{\left({\displaystyle \sum {}_{j=1}^p\left[\left(Am{p}_{y_j}\right)\times \text{sin}\left(-{\phi }_{y_j}\right)\right]}\right)}^2}}\right)$$ $$\beta ={\text{tan}}^{-1}\left(\frac{{\displaystyle \sum {}_{k=1}^q\left[\left(Am{p}_{z_k}\right)\times \text{cos}\left(-{\phi }_{z_k}\right)\right]}}{\sqrt{{\left({\displaystyle \sum {}_{i=1}^m\left[\left(Am{p}_{x_i}\right)\times \text{cos}\left(\frac{\mathrm{\pi }}{2}+{\phi }_{x_i}\right)\right]}\right)}^2+{\left({\displaystyle \sum {}_{j=1}^p\left[\left(Am{p}_{y_j}\right)\times \text{cos}\left(\frac{\mathrm{\pi }}{2}+{\phi }_{y_j}\right)\right]}\right)}^2}}\right)$$

Note that the calculations of the angles α and β are represented as cosines of the angles and not as sines, since the phases of a sinusoidal signal are defined as a cosine wave whose peak for zero phase begins at t=0. Thus, the phases extracted from the FFT are (π/2+φzk), (π/2+φzk), and (π/2+φzk), from which π/2 must be subtracted and then multiplied by -1 to be used in the sine operations in the calculations of the acceleration components, as derived at the beginning.

From the magnitude and direction, which can be derived as explained above, a corresponding causal mechanical impact on the vehicle can be deduced. In particular, the location on the vehicle where the impact occurred and the magnitude of the impact force must be determined.

Both forces, i.e., the impact force on the detection unit and the force of the mechanical impact, must be related to each other according to the finite element analysis of the vehicle chassis with all its internal and external suspensions and connections. Such a relationship is complex to formulate and requires further simulation analysis when all internal and external physical connections of the chassis are given.

A simplified way of establishing such a correlation is by creating a correlation in an evaluation model. For this purpose, traces of acceleration signals are recorded that correspond to specific impact forces at specific locations on the vehicle's outer surface. A test setup can be used in which a specific object, e.g., a pendulum ball, impacts the vehicle in a targeted manner, and the corresponding acceleration signals are recorded, particularly the direction and force of the resulting impact force. The location and angle of impact of the object are then also known and specified. This is repeated for a number of different locations, angles, and, if applicable, masses, to generate a correlation table. This can then be used in an evaluation model, including by utilizing artificial intelligence.

Results can be considered, for example, at frequencies in the range of 0 to 800 Hz. It should also be noted that the direction of the resulting acceleration (impact force) at the detection unit (IMU) does not have to be symmetrical to the impact forces of the mechanical action, which are approximately symmetrical on the right and left sides of the vehicle. This could be due to the asymmetric Mechanical chassis joints or loads within the vehicle may result in asymmetric transients transmitted from the vibrations of the impact forces to the IMU sensor. Therefore, more data may be needed to validate the relationship between the location of a mechanical impact and the location of the impact at the IMU. Consequently, the locations of the input impact forces and the corresponding direction of the resulting forces at the IMU are fed into an inference engine to estimate a closed-form relationship between them. This relationship formula can then be used to mimic the finite element analysis of the vehicle frame and its associated joints and suspensions in a simplified and efficient manner.

While at least one exemplary embodiment has been described above, it should be appreciated that a wide variety of variations exist. It should also be understood that the described exemplary embodiments are merely non-limiting examples and are not intended to limit the scope, applicability, or configuration of the devices and methods described herein. Rather, the foregoing description will provide one skilled in the art with guidance for implementing at least one exemplary embodiment, it being understood that various changes in the operation and arrangement of the elements described in an exemplary embodiment may be made without departing from the subject matter as defined in the appended claims, as well as their legal equivalents.

LIST OF REFERENCE SYMBOLS

1

vehicle

2

Detection device

3

resulting force

10

mechanical impact

11

Point of action

100

Method for detecting a mechanical impact on a vehicle

101-107

individual processes or procedural steps within the framework of procedure 400

Claims (15)

A method for detecting mechanical impacts (10) on a vehicle (1), wherein the vehicle (1) has at least one detection device (2) configured to detect accelerations in three-dimensional space, the method comprising: - detecting sensor data by means of the detection device (2), wherein the sensor data represent an impact (3) on the detection device (2) caused by a mechanical impact applied from outside to an outer shell of the parked vehicle (1), wherein the sensor data comprise effective components of the impact on the detection device (2) in various directions of the three-dimensional space; - processing the detected sensor data, wherein each effective component is represented by a time-dependent function of a damped oscillation, from which an acceleration function is derived, wherein a start time of the acceleration functions corresponds to a time at which the mechanical impact on the vehicle (1) occurred; - Evaluating the acceleration functions at the starting time in order to obtain at least one characteristic of a force of the impact (3) on the detection device (2), wherein the at least one characteristic of the force of the impact (3) comprises a direction and a magnitude of the force; - Processing the at least one obtained characteristic of the force of the impact (3) using an evaluation model in order to obtain at least one characteristic of the mechanical action (10) on the vehicle (1); and - Outputting an evaluation result representing the at least one characteristic of the mechanical action (10) on the vehicle (1). Procedure according to Claim 1 , where a Fast Fourier Transform is applied to the acceleration components to obtain a frequency spectrum for each acceleration component whereby the acceleration components are only evaluated for frequencies at which the amplitude exceeds a threshold value. Procedure according to Claim 1 , where the direction of the force of the impact (3) is represented by two angles obtained by angular operations of the acceleration components at the starting time. Procedure according to Claim 1 or 3 , wherein the magnitude of the force of the impact (3) is obtained from an oscillating mass and the acceleration components at the starting time. Method according to one of the preceding claims, wherein the at least one feature of the mechanical action (10) on the vehicle (1) comprises a location (11) on the vehicle (1) at which the mechanical action on the vehicle (1) occurred. Method according to one of the preceding claims, wherein the at least one feature of the mechanical action (10) on the vehicle (1) comprises an amount of a force of the mechanical action (10). Method according to one of the preceding claims, wherein, based on the evaluation result, a message is sent to a user's terminal device, which message contains at least one statement that a mechanical action (10) has taken place on the vehicle (1). Procedure according to Claim 6 and Claim 7 , wherein the message is sent to the user's terminal only if the magnitude of the force of the action (10) exceeds a predetermined threshold value. Method according to one of the preceding claims, wherein the evaluation model represents an association between the mechanical action (10) on the vehicle (1) and the impact on the detection device (2). Procedure according to Claim 9 , wherein the evaluation model contains a plurality of assignments based on empirically determined assignments, wherein a plurality of specific mechanical effects (10) on the vehicle (1) are each assigned to a corresponding resulting impact (3) on the detection device (2). System for detecting mechanical effects on a vehicle (1), wherein the system comprises a detection device (2) for detecting accelerations and a data processing device configured to carry out the method according to one of the preceding claims. Computer program comprising instructions which, when executed on the data processing device of the system according to Claim 11 cause the system to perform the procedure according to one of the Claims 1 until 10 to execute. Computer program product, comprising a non-volatile data carrier on which instructions are stored which, when executed on the data processing device of the system according to Claim 11 cause the system to perform the procedure according to one of the Claims 1 until 10 to execute. Module for a vehicle (1), comprising the system according to Claim 11 , wherein the module is designed to be installed in a vehicle (1). Vehicle comprising a module according to Claim 14 , which is arranged as a roof module in a roof area of the vehicle (1).