Method and apparatus for classifying an accident process with respect to a two-wheeled vehicle

Abstract

The invention relates to a method for classifying an accident process of a two-wheeler, in particular a bicycle. Here, the method according to the invention can be run as an algorithm in a device with an analytical evaluation unit in order to display a collision or a fall of the two-wheeler to the rider or a third party by means of correspondingly generated and / or emitted information. Here, the device can be used in a two-wheeler like a bicycle, in particular in an e-bike. However, it goes without saying that a use in a motorcycle or other single-track vehicle is also possible.

Description

Technical Field

[0001] The present invention relates to a method and apparatus for classifying or verifying the reliability of accident processes involving two-wheeled vehicles, particularly accidents involving electric bicycles. Background Technology

[0002] A series of sensors mounted on two-wheeled vehicles (such as bicycles), and especially e-bikes, can be used to monitor not only the operation of the two-wheeled vehicle but also its driving status. For example, speed sensors can sense wheel speed and thus deduce the vehicle's speed. Furthermore, inclination sensors, or even acceleration sensors or speed sensors, can sense the tilt or inclination on the road, and sensors on the bicycle's pedal cranks can sense driving activity.

[0003] Analysis and evaluation of these sensed sensor parameters can identify critical situations or even accidents, for example, by monitoring the behavior of these sensor parameters over time. These identified critical situations or accidents can be used to automatically request assistance or transmit at least one piece of information to a third party. Summary of the Invention

[0004] Here, the objective of the present invention is to classify accidents so that accidents presenting as critical situations can be distinguished from wheel falls without direct collision.

[0005] This invention relates to a method for classifying the accident process of two-wheeled vehicles, particularly bicycles. The method according to the invention can be run as an algorithm in a device with an analysis and evaluation unit to display the collision or fall of the two-wheeled vehicle to the rider or a third party by means of correspondingly generated and / or emitted information. This device can be used in two-wheeled vehicles, such as bicycles, or especially in electric bicycles. However, it is also obviously applicable to motorcycles or other monorail vehicles.

[0006] To classify the accident process of a two-wheeled vehicle or a corresponding monorail vehicle, the acceleration of the two-wheeled vehicle in its plane of motion is sensed. Optionally, motion in the vertical plane can also be sensed. To sense acceleration in the plane of motion, for example, two mutually orthogonal acceleration directions can be sensed. This sensing can be continuous, providing new sensor parameters continuously, or it can be time-controlled within a predefined time range or at predefined time points. Alternatively, the sensing can be controlled based on predefined or specifically identified operating conditions. A first integral parameter is then formed or generated based on the time integration of a first sensor parameter, representing, for example, acceleration in the longitudinal or travel direction of the two-wheeled vehicle. Correspondingly, a second integral parameter is formed or generated based on a second sensor parameter, representing, for example, acceleration in the lateral direction relative to the longitudinal or travel direction of the two-wheeled vehicle. Here, the core of the invention lies in classifying accidents, for example, identified by other methods, as follows: to what extent do they involve serious accidents or only involve relatively harmless two-wheeled vehicle falls without anticipated injury to the two-wheeled rider. For this purpose, the classification and thus identification of serious accidents are performed based on first and second integral parameters.

[0007] Here, classification can be performed by comparing the first and / or second integration parameters with at least one corresponding common or separately assigned threshold. Alternatively or additionally, the classification result can be configured to relate to a direct comparison of the two integration parameters.

[0008] Here, if, for example, a first integral parameter representing acceleration in the longitudinal direction of the two-wheeled vehicle, or its value, exceeds a first threshold, the method according to the invention can identify a collision and thus an accident with the presumed personal injury. Optionally or additionally, a collision can also be identified if the first integral parameter, or its value, is greater than a second integral parameter, particularly by a factor of several. In this case, for example, it might indicate that the (negative) acceleration, or braking in the longitudinal direction, is sudden and significantly greater than a sideways rollover of the two-wheeled vehicle. Here, if a collision is identified, the method can generate information to be forwarded to the rider or an external location to call for assistance.

[0009] However, the method according to the invention can also identify accident conditions in which no personal injury is expected, such as when a two-wheeled vehicle only tilts to the side. For this purpose, the method checks during classification whether a second integral parameter, or its value, exceeds a second threshold. Optionally or additionally, the method may check whether the second integral parameter is greater than the first integral parameter, particularly by a large multiple. Such behavior might mean that the two-wheeled vehicle performs a lateral tilting motion and has essentially no forward movement or braking in the longitudinal direction.

[0010] Optionally, the method can also sense the roll and pitch angles of the two-wheeled vehicle, or the time-varying changes of these roll and pitch angles in the sense of rotational speed. Similarly, a classification of the two-wheeled vehicle's motion can be derived from the sensed roll or pitch angles or the corresponding rotational speeds.

[0011] In an extended embodiment of the invention, a mass-specific energy parameter is derived from the first and second integral parameters, which can also be considered for classification. For this purpose, additionally, a common rollover parameter is derived from the roll and pitch angles, which is also included in the classification. To identify a collision, it is checked how much the energy parameter exceeds an energy threshold and the rollover parameter exceeds a tilt threshold. First, if the energy threshold is exceeded and then the rollover parameter threshold is exceeded, the method generates information indicating a collision with potential personal injury. However, if both the energy value and the rollover parameter remain below the corresponding threshold, the method generates information representing only a two-wheeled vehicle, particularly a lateral fall, or even an upright bicycle. To distinguish lateral collisions, additional rollover thresholds can be considered. For upright bicycles, typical parameters fall within the 0-45° rollover parameter range.

[0012] Optionally, the method can be configured to generate information representing a two-wheeled vehicle collision based on energy values ​​above a corresponding threshold and rollover parameters below a corresponding threshold.

[0013] Optionally, an energy threshold for identifying a collision can be set based on the roll parameter, roll angle, and / or pitch angle, especially their absolute values. Here, the energy threshold can be set to be low when the values ​​of the roll parameter, roll angle, and / or pitch angle are small, and as the values ​​of the roll parameter, roll angle, and / or pitch angle increase, the energy threshold first increases slowly and then, for example, exponentially or quadratically. If this energy threshold is exceeded, a collision is identified. Alternatively, an energy threshold for identifying an impact can be used, which is initially set high when the values ​​of the roll parameter, roll angle, and / or pitch angle are small, and the energy threshold first decreases slowly and then similarly exponentially or quadratically as the values ​​of the roll parameter, roll angle, and / or pitch angle increase.

[0014] Further advantages will arise from the subsequent description of the embodiments or from the dependent claims. Attached Figure Description

[0015] Figure 1 The diagram schematically illustrates a two-wheeled vehicle equipped with a mobile terminal device and a coordinate system in which the two-wheeled vehicle moves during normal driving activities. Figure 2 The block diagram illustrates one possible implementation of the device according to the invention. Figure 3 The flowchart illustrates a possible embodiment of the method according to the present invention. Detailed Implementation

[0016] To illustrate the invention, a two-wheeled vehicle in the form of a bicycle 10 is taken as a starting point, wherein other monorail vehicles, such as electric bicycles, motorcycles, electric scooters, two-wheeled scooters, or even automobiles, may also be equipped with the invention. In the case of the present bicycle 10, a smartphone 20 is configured as a mobile terminal device, mounted on the steering mechanism of the bicycle 10 and configured to sense and process sensor data. Here, the smartphone 20 serves, for example, as a navigation instrument, as a display of driving dynamic parameters, and / or as a drive mechanism for controlling the bicycle 10. To perform the method according to the invention, the smartphone 20 utilizes sensor parameters provided by sensors available in the smartphone 20. Alternatively or optionally additionally, the smartphone 20 may also access sensors mounted on the bicycle 10. Here, for example, an acceleration sensor, as well as angle and rotation speed sensors, are conceivable on one of the wheels and / or on the frame. The bicycle generally moves forward in the longitudinal direction x on a track within the motion plane xy. Through cornering and curves, the motion plane xy is extended by an additional lateral direction y. Movement in the direction of the vertical axis z takes place, for example, in the category of uphill and downhill travel. Additional motions that may affect the vertical axis direction are caused by rotation in the x-axis direction or by uneven road surfaces. All these motions in the z-axis direction can be distinguished from each other by changes in velocity, i.e., the characteristic order of magnitude of dz / dt. Compared to the lateral tilt of a bicycle, which is essentially along the x-axis, such as riding uphill or downhill, these motions are associated with a longer time constant. In contrast, road unevenness is characterized by very small motions in the z-axis over short periods of time.

[0017] The sensors within the smartphone (20 inches) are aligned with their own coordinate systems. Therefore, by mounting the smartphone on the steering mechanism, recalibration is necessary with respect to the coordinate system determined by motion. Location-resolved sensor parameters from the smartphone's tachometer or accelerometer can be used to sense this defined coordinate system during normal driving activities. One possibility for performing this calibration to align the sensor parameters with the bicycle's motion plane (xy) is, for example, Euler angle estimation methods.

[0018] exist Figure 2The diagram schematically illustrates a device according to the invention, which implements the method according to the invention. Here, the device may be configured with an analysis and evaluation unit 100 including a memory 110, while sensors 120, 130, and 140 are configured separately, allowing the corresponding sensor parameters to be read through the analysis and evaluation unit 100. Alternatively, the device may also have one or more sensors in addition to the analysis and evaluation unit 100 and the memory 110, thus sensing at least a portion of the sensor parameters necessary for the method according to the invention. In this case, for example, referring to the smartphone 20 mentioned at the beginning as a device according to the invention, the device already includes an acceleration sensor 120, an angle sensor 130, and / or a rotational speed sensor 140. The associated memory 110 may store not only the sensed sensor parameters but also derived parameters, such as integral parameters or energy parameters and tilt parameters. Furthermore, corresponding thresholds may be stored in the memory 110.

[0019] Using at least one first sensor 120, the analysis and evaluation unit 100 senses acceleration parameters representing the motion of the two-wheeled vehicle. It is particularly advantageous here that the first sensor 120 senses acceleration in at least two directions, x and y, in the form of first and second sensor parameters, respectively, on the plane of motion. Alternatively, it may be configured to sense first and second sensor parameters corresponding to its own acceleration sensor for each of the two directions of motion.

[0020] In another embodiment, additional sensors may be provided, capable of classifying the accident process or verifying the reliability of the classification. For this purpose, an angle sensor 130 may be provided, which senses sensor parameters for the roll and pitch angles of the two-wheeled vehicle. Alternatively, two angle sensors may be provided, independently sensing the sensor parameters for the roll and pitch angles. In addition to the angle sensor 130, or alternatively, the analysis and evaluation unit 100 may sense the sensor parameters of the speed sensor 140. Here, the first speed ω around the longitudinal axis x... x Or the direction of travel of the two-wheeled vehicle and the second rotational speed ω around the y-direction transverse to the longitudinal axis x. y This is of particular interest for classifying the accident process. As is already present in the acceleration sensor 120 or the angle sensor 130, it can also be configured in the speed sensor 140, with the separate speed sensor 140x sensing the first and second speeds ω. x and ω y .

[0021] If the method identifies an accident process as being related to a collision or personal injury, the analysis and evaluation unit 100 generates information to display this status. In its simplest case, this can be done by notifying the rider via a corresponding optical or acoustic display 160. Alternatively, the information can be sent to an external receiver 150 to request assistance. This could involve a central dispatch center, pre-selected contacts, or, more generally, assistance services. Optionally, if a collision is detected, another unit 170 within the device 100 or on the two-wheeled vehicle can be activated. For example, this could involve shutting off the drive unit, switching the power source to no current, or issuing a prepared emergency signal.

[0022] If the method identifies only one type of fall that poses no danger, similar information can be released. Therefore, it is possible to configure the forwarding of information regarding the severity of the fall to proactively assist services. This also allows for control over the level of intervention in the components and devices 170 of the two-wheeled vehicle.

[0023] Subsequently, according to Figure 3 The flowchart illustrates a possible process of the method according to the invention. This method, or the associated algorithm, can particularly distinguish between frontal or lateral collisions of two-wheeled vehicles with other vehicles, persons, animals, and articles, and especially lateral falls. Thus, not only can the severity of the accident be classified, but also the course or progression of the accident can be estimated. The classification and estimation of the accident course can be used directly or through other methods to take assistance measures.

[0024] After the method begins, in the first step 300, at least the acceleration parameter a is sensed. x and a y The acceleration parameters represent the acceleration of the two-wheeled vehicle in the x and y directions. These two directions are essentially the direction of motion of the two-wheeled vehicle, or longitudinal direction x, and the y direction, which is orthogonal to the x direction in the typical plane of motion of the two-wheeled vehicle. Here, the necessary acceleration sensor can be fixedly mounted to the two-wheeled vehicle or can be part of a display and control device that can be temporarily fastened to the bicycle by the rider or carried by the rider to sense motion during riding. For example, to sense these and other sensor parameters, it is conceivable to use these sensors in mobile terminal devices, such as smartphones, or human machine interfaces (HMIs), such as those used to control the drive mechanism of electric vehicles.

[0025] Optionally, additional sensor parameters can be sensed in step 300, which can be used in other embodiments. For example, the speed sensor parameter ω can be sensed. x and ω y The speed sensor parameter represents the rotational speed about the x or y direction and thus senses the (lateral) rollover of the two-wheeled vehicle or the two-wheeled vehicle tipping forward due to the steering mechanism. Additionally or alternatively, the roll angle... And / or the pitch angle θ can also be sensed using a suitable sensor. Instead of sensing the roll and pitch angles independently, they can also be derived from the corresponding speed sensor parameters.

[0026] Alternatively, the sensing of the sensor parameters necessary for this method can also be performed in a separate method, so that the current sensor parameters are continuously provided to the method according to the invention.

[0027] If the sensing of sensor parameters, such as accelerometer parameters, is performed via a detachable mobile terminal device, it may be necessary to calibrate the alignment of the sensor parameters, as the spatial orientation defined in the mobile terminal device is inconsistent with the direction of motion based on the mounting and orientation when the device is secured. Calibration, for example, using Euler angle estimation, is suitable for this purpose. Alternatively, it may be configured to compensate for the gravitational acceleration component in the sensed acceleration parameters.

[0028] In the following step 310, the sensed acceleration sensor parameters are based on

[0029] Δv x =∫a x dt

[0030] Δv y =∫a y dt

[0031] Accordingly, integration is performed to identify not only velocity changes in the x-direction but also velocity changes in the y-direction. Consideration of these two spatial directions is sufficient to identify collisions where the motion of the two-wheeled vehicle is primarily situated in these two spatial directions. Motion in the z-axis direction is essentially caused by uneven ground or by uphill or downhill inclines. Advantageously, the integration range is selected based on the time recording of the sensor parameters. With a sampling rate of 100Hz, it may be sufficient to integrate 5 to 10 sensor parameters separately, allowing for an integration time of 50 to 100 ms. Furthermore, such a duration is sufficient to distinguish between a collision and a (lateral) fall (“ground impact”).

[0032] Next, in step 330, based on the velocity change Δv x or Δv yIdentify the spatial direction in which the main change in velocity occurs. Here, if it is identified that the velocity change primarily occurs in the y-direction, then the velocity change Δv in the y-direction... y The threshold SW has been exceeded. y If the change in the y-direction is the main part, especially if it is many times larger, then the two-wheeled vehicle (lateral) overturning is identified, so that the rider can be notified in step 350 or the information "two-wheeled vehicle overturned" can be generated in general.

[0033] Optionally, a corresponding threshold SW can be used. x and SW y The choice makes a distinction between "whether the collision occurs from the front (or rear) or from the side." Therefore, in Δv y When the value of is very large, a side collision can be identified because a higher speed change can be expected in a side collision compared to a two-wheeled vehicle tipping over. However, if the system identifies: speed change Δv y Very small, for example, by comparing it to another smaller threshold SW. y Then it can be identified that there is neither a collision nor an impact to the ground.

[0034] If, in step 330, it is identified that a frontal or lateral collision has occurred, for example, by sensing a higher velocity change Δv in the x-direction than in the y-direction. x This method identifies collisions (either longitudinally or laterally). Additionally, a threshold SW can be set here. x The threshold SW is exceeded in order to further verify the credibility of the collision. Different thresholds SW can also be used here. x，n This allows for the classification of the severity of the collision.

[0035] After a collision is identified in step 330, a further step 380 can generate information about the collision, which can then be forwarded, for example, to a third party. Therefore, a portable mobile terminal device can be used to call for assistance via radio or simply to notify designated personnel. Furthermore, additional information can be forwarded based on the severity of the collision, facilitating care for cyclists who have suffered injuries in the collision. Moreover, the mobile terminal device used can be the same device, and the necessary sensor parameters are also sensed by the same device.

[0036] After identifying a fall or collision, the method can either end or restart from step 300.

[0037] In another embodiment, after a collision with a two-wheeled vehicle is detected, in optional step 390, at least one component of the two-wheeled vehicle can be manipulated to prevent further damage to the rider and / or the two-wheeled vehicle. Here, for example, the drive mechanism of the electric bicycle can be switched off. However, it is also conceivable to manipulate optical and / or acoustic warning devices to notify other road users and / or call for assistance. Accordingly, similar manipulation of components of the two-wheeled vehicle can also be set in optional step 360 when a fall is detected. Here, if the rider has already stopped the two-wheeled vehicle and is not nearby, the fall can be notified to the rider.

[0038] As explained at the beginning, there is a possibility of sensing additional sensor parameters to verify the reliability of collision identification and / or improve the distinction relative to a simple two-wheeled vehicle falling over. Therefore, in step 310, it is possible to...

[0039]

[0040] From the sensed roll angle The tilt parameter γ is derived from the pitch angle θ. This tilt parameter represents the orientation of the two-wheeled vehicle. This tilt parameter γ can be determined based on...

[0041]

[0042] With the estimated energy parameter E kin Together with the change in velocity Δv x and Δv y The relevant parameter is used in step 330 to perform a reliability check on the collision identification by means of the overturning parameter γ. Here it is considered that the overturning parameter γ (and the overturning threshold SW) are used in step 330. k (Comparison) higher and subsequently energy parameter E kin (with energy threshold SW) E In cases where the rollover parameter γ is relatively high, a two-wheeled vehicle collision is possible. Conversely, if the rollover parameter γ is below the tilt threshold SW... k And the energy parameter E kin Below the energy threshold SW E This identifies when a two-wheeled vehicle falls or overturns, especially when it falls sideways. If a threshold is exceeded by one value and another value does not exceed the corresponding threshold, a reliability check must be performed using additional sensor parameters to identify a collision or fall of the two-wheeled vehicle. Similarly, this type and additional reliability checks can be performed in step 330.

[0043] Alternatively, a side collision can also be identified in step 330. For this, firstly, a large roll angle is used. For example, by using a threshold Comparative identification of (lateral) tipping or overturning may be associated with an increased speed change in the y-direction. However, if a small roll angle is identified... In the small roll angle In cases where both exist simultaneously (e.g., through a threshold SW) Ekin，y (Comparison) Higher kinetic energy E kin，y It can identify lateral collisions rather than simply falling to the side.

[0044] Another implementation may include, in step 330, the existence of a temporal correlation between the various sensed angles or between one or more thresholds to identify a frontal or lateral collision or the fall of a two-wheeled vehicle. Thus, a frontal collision can be identified if a significant pitch angle is first identified, followed by a roll angle within a short timeframe. This might indicate that the two-wheeled vehicle struck an obstacle and then the rider tilted to the side with the vehicle. Correspondingly, successive time delays in the longitudinal x-direction and lateral y-direction may also indicate a frontal collision.

[0045] Alternatively, a credibility check can be performed separately after the identification in step 330 to verify the identification in step 330. Here, the identification of a fall can be checked in a separate subsequent step 340 and the identification of a collision can be checked in step 370. For this purpose, additional sensor parameters or changes in these additional sensor parameters over time can be used. Thus, it can be identified, for example, shortly after a collision is identified, whether the two-wheeled vehicle is moving further in the longitudinal direction at an interpretable speed, so that the collision is characterized as unreliable. In this case, the method may end or restart at step 300. Similarly, the rider raising the two-wheeled vehicle or the identified 180° turn for maintenance conditions in step 340 can be interpreted as a fall of the two-wheeled vehicle. In the latter case, locking or disabling the drive unit in step 360 may be helpful for replacing the chain. However, if it is identified in step 340 that there is no fall, the method may end or new sensor parameters can be sensed at step 300 for reassessment.

[0046] In another embodiment, prior to the classification or identification of the collision or fall condition in step 330, an upstream step 320 may be performed, in which the threshold used, such as SW, is determined. k SW E SW x Or SW yThe energy threshold E can be determined based on the operating parameters of the two-wheeled vehicle and / or according to driving parameters or surrounding environmental parameters, such as the variance of sensed rotational speed parameters in the x and / or y directions. It is also conceivable to determine multiple thresholds or calculated or generated values ​​for the sensor parameters, for example, to classify the severity of a collision. Optionally, an energy threshold E can be set. kin，x or E kin，y At least one of these components is variable. Therefore, it is conceivable that the corresponding energy threshold increases or decreases with the roll parameter and / or the absolute roll or pitch angle. Thus, the energy threshold for identifying a collision can be selected such that it is chosen to be low when the values ​​of the roll parameter, roll angle, and / or pitch angle are small, while the energy threshold increases similarly as the roll parameter, roll angle, and / or pitch angle increase, for example, exponentially or quadratically. If the sensed or calculated energy exceeds the energy threshold, a collision is identified. Alternatively or additionally, an energy threshold for identifying impacts on the ground can be used. Here, when the values ​​of the roll parameter, roll angle, and / or pitch angle are low, a high energy threshold can be initially set, which decreases as the roll parameter, roll angle, and / or pitch angle increase, for example, with an exponential or quadratic correlation decrease. Here, if the energy threshold is exceeded, an impact can be identified.

[0047] It is also conceivable to use a probabilistic model instead of a threshold, or to use a probabilistic model when deriving the threshold. Here, for example, an internal or external database, or a single sensed sensor parameter, or a derived value, can be utilized. Therefore, it is conceivable that the analysis and evaluation unit 100 has a corresponding database in memory 110 or can access such a database by means of, in particular, a wireless radio connection.

[0048] In step 330, identification or classification can also be performed directly from a single sensed roll angle. Pitch angle θ, rotational speed parameters, especially their variance and / or orientation energy parameters E kin，x or E kin，y To do this, appropriate thresholds can be set.

Claims

1. A method for classifying the accident process of two-wheeled vehicles, wherein, The method includes at least: sensing a first sensor parameter (a x ) and second sensor parameters (a y The first and second sensor parameters represent the acceleration of the two-wheeled vehicle (10) in two directions in the plane of motion of the two-wheeled vehicle (10), and according to the first sensor parameter (a x The time integration produces the first integration parameter (v) x ), and according to the second sensor parameter (a y The time integration produces the second integration parameter (v) y ), and according to the first integral parameter (v) x ) and the second integral parameter (v y The method performs classification of the accident process, wherein the method at least senses a roll angle (φ), which represents the rotation of the two-wheeled vehicle about a longitudinal axis, and senses a pitch angle (θ), which represents the rotation of the two-wheeled vehicle about a lateral axis, and performs the classification of the accident process based on the roll angle and the pitch angle, wherein the method generates an energy parameter (E) based on a first integral parameter and a second integral parameter. kin ), and generate roll parameters (γ) based on the roll angle and the pitch angle, and generate roll parameters (E) based on the energy parameters (E). kin The energy parameter (γ) and the overturning parameter (γ) are used to classify the accident process, wherein if the energy parameter exceeds the energy threshold (SW) E And the tilting parameter (γ) exceeds the tilting threshold (SW). k ), generating information representing the collision of the two-wheeled vehicles, or if the energy parameter (E kin The energy threshold (SW) does not exceed the stated energy threshold. E And the tilting parameter (γ) does not exceed the tilting threshold (SW). k This generates information representing the sideways fall of the two-wheeled vehicle.

2. The method according to claim 1, characterized in that, The method involves: converting the first integral parameter (v) into... x ) and / or the second integration parameter (v y ) and at least one associated threshold (SW) x SW y Compare with, and / or compare the first integration parameter (v) with, and / or compare with, the first integration parameter (v) ... x ) and the second integral parameter (v y ) for comparison, Perform the classification.

3. The method according to claim 1 or 2, characterized in that, The method senses the acceleration parameter as the first sensor parameter (a x The acceleration parameter represents the acceleration of the two-wheeled vehicle in the longitudinal direction (x), and the sensed acceleration parameter is used as a second sensor parameter (a). y The acceleration parameter represents the acceleration of the two-wheeled vehicle in the lateral direction (y) relative to the longitudinal direction (x).

4. The method according to claim 3, characterized in that, If the first integral parameter (v) x Exceeding the first threshold and / or the first integral parameter (v) x ) exceeds the second integral parameter (v) y If the collision rate is several times higher than the collision rate, the method generates information representing the collision of the two-wheeled vehicle when performing the classification.

5. The method according to claim 3, characterized in that, If the second integral parameter (v) y If the second threshold is exceeded, and / or the second integral parameter (v) is exceeded. y ) exceeds the first integral parameter (v) x The method generates information representing the lateral fall of the two-wheeled vehicle during the classification process, which is several times greater than the classification process.

6. The method according to claim 1 or 2, characterized in that, The classification is set to consider the corresponding thresholds for the roll angle and / or pitch angle being exceeded by the roll angle and / or pitch angle.

7. The method according to claim 1 or 2, characterized in that, The method at least senses a first rotational speed (ω). x The first rotational speed represents the rotation of the two-wheeled vehicle around the longitudinal axis, and the sensed second rotational speed (ω) y The second rotational speed represents the rotation of the two-wheeled vehicle around the horizontal axis, and according to the first rotational speed (ω) x ) and the second rotational speed (ω) y Perform the classification of the accident process.

8. The method according to claim 1 or 2, characterized in that, The two-wheeled vehicle mentioned is a bicycle.

9. The method according to claim 6, characterized in that, Consider: The corresponding threshold is exceeded within a predetermined time interval.

10. The method according to claim 7, characterized in that, The method is set such that if the first rotational speed (ω) x Exceeding the first speed threshold (SWω) x ), generating information representing the lateral fall of the two-wheeled vehicle, or if the second rotational speed (ω) y Exceeding the second speed threshold (SWω) y This generates information representing a collision between the two-wheeled vehicles.

11. Equipment for classifying the accident process of two-wheeled vehicles, wherein, The device has an analysis and evaluation unit (100) that implements the method according to any one of claims 1 to 10, wherein the analysis and evaluation unit (100) senses at least a first sensor parameter and a second sensor parameter, the first sensor parameter and the second sensor parameter representing the acceleration of the two-wheeled vehicle (10) in two directions in the plane of motion of the two-wheeled vehicle (10), generates a first integral parameter based on the time integral of the first sensor parameter, generates a second integral parameter based on the time integral of the second sensor parameter, and performs a classification of the accident process based on the first integral parameter and the second integral parameter.

12. The device according to claim 11, characterized in that, The two-wheeled vehicle mentioned is a bicycle.

13. The device according to claim 11, characterized in that, The analysis and evaluation unit classifies the accident process by outputting information representing the collision or lateral fall of the two-wheeled vehicle.

14. A two-wheeled vehicle having the equipment according to any one of claims 11 to 13.

15. The two-wheeled vehicle according to claim 14, characterized in that, The two-wheeled vehicle mentioned is a bicycle.

Images