Methods for generating and / or processing motion data, methods for recording and / or evaluating motion sequences, methods for detecting and / or evaluating dynamic instabilities, methods for generating and / or optimizing algorithms and / or models, devices for executing such methods and computer programs

Abstract

Method for generating and / or processing motion data of a freely movable body (130), wherein at least two sensor units (104, 106, 108, 110), each comprising an angular velocity sensor and an acceleration sensor, are arranged at a fixed distance from one another on the body (130), and an angular acceleration vector is directly determined based on measurement signals from the at least two sensor units (104, 106, 108, 110) and by applying the laws of general motion of rigid bodies; method for recording and / or evaluating motion sequences of a freely movable body (130), wherein motion data of the body (130) generated and / or processed according to such a method are used; method for detecting and / or evaluating dynamic instabilities of a freely movable body (130), wherein motion data of the body (130) generated and / or processed according to such a method are used.Method for generating and / or optimizing algorithms and / or models for calculating the dynamics of freely moving bodies (130), wherein motion data of the body (130) generated and / or processed according to such a method are used, and device (100) for carrying out such a method with at least two sensor units (104, 106, 108, 110) that can be arranged fixedly apart from one another on a freely moving body (130), each comprising an angular velocity sensor and an acceleration sensor.

Description

[0001] The invention relates to a method for generating and / or processing motion data of a freely moving body. Furthermore, the invention relates to a method for detecting and / or evaluating motion sequences of a freely moving body. Furthermore, the invention relates to a method for detecting and / or evaluating dynamic instabilities of a freely moving body. Furthermore, the invention relates to a method for generating and / or optimizing algorithms and / or models for calculating the dynamics of freely moving bodies. Furthermore, the invention relates to a device for executing such methods. Furthermore, the invention relates to a computer program.

[0002] The publication "King, AI (2018). Measurement of Angular Acceleration. In AI King, The Biomechanics of Impact Injury (pp. 153-178). Springer International Publishing." deals with the measurement of angular acceleration. It first describes the traditional method and shows that it is numerically unstable. It then presents an alternative method that is numerically stable but requires more sensors.

[0003] The publication "Martin, PG, Hall, GW, Crandall, JR, & Pilkey, WD (1998). Measuring the Acceleration of a Rigid Body. Shock and Vibration, 5(4), 211-224." deals with the measurement of the acceleration of a rigid body and presents two methods for measuring the acceleration of a point on a rigid body with six degrees of freedom. The first method, called the peripheral system, uses three groups of accelerometers mounted orthogonally to each other and aligned with the point's axes. One of the groups consists of the three accelerometers mounted on a cube-shaped three-axis angular rate sensor (ARS). The second method, called the compact cube scheme, uses a single three-accelerometer / ARS cluster that can be mounted anywhere on the rigid body.

[0004] The publication "Ovaska, SJ, & Valiviita, S. (1998). Angular acceleration measurement: A review. IMTC / 98 Conference Proceedings. IEEE Instrumentation and Measurement Technology Conference. Where Instrumentation is Going (Cat. No. 98CH36222), 2, 875-880." provides an overview of sensors, methods, and algorithms available for measuring angular acceleration. The focus is on delay-sensitive real-time applications.

[0005] Document WO 2021 / 240567 A1 relates to a method for evaluating the accelerations or velocities acting on a user's head, comprising the steps of: providing a protective device suitable for being worn on the user's head and comprising one or more sensors installed on the protective device in a position such that they are not in direct contact with the head; detecting a plurality of signals by means of the one or more sensors, each signal comprising data relating to an acceleration and / or velocities; analyzing and processing the signals to determine the temporal progression of the acceleration and / or velocity experienced by the head.In order to enable an evaluation of the entire temporal progression of both the linear and rotational accelerations experienced by the head when calculating the linear and rotational accelerations acting on the head in the event of an impact, document WO 2021 / 240567 A1 proposes carrying out the analysis and data processing step using an analysis device comprising at least one artificial intelligence module consisting of at least one machine learning module and configured to carry out an initial training step, which provides for input data and corresponding output data to be input into the at least one artificial intelligence module, the data being data relating to the acceleration experienced by the protective device in the event of an impact, collision or accident.

[0006] The document WO 2017 / 118610 A1 relates to a method for determining a tipping fall or a risk of tipping fall, comprising the following steps: determining a center of gravity position of a center of gravity of the person, which represents an actual center of gravity of the person or an approximation thereof, determining a position of a standing base of the person or an approximation thereof, determining connection information regarding a spatial connection between the center of gravity and the standing base.In order to detect a risk of falling or a fall and to trigger the opening of an impact protection bag, the following steps are proposed in document WO 2017 / 118610 A1: detecting an angular position of at least one lower leg of a leg in relation to the associated thigh and an angular position of the same thigh in relation to the torso; calculating a relative position of an ankle joint or at least part of a foot or footwear of the leg in relation to the centre of gravity as connecting information, using angular positions between the lower leg and thigh and between the thigh and torso, as well as lengths of the lower leg and thigh, and a relative position between the hip joint of the leg and the centre of gravity; deriving a tipping fall or a risk of tipping fall from the extent to which it is true that the centre of gravity and the standing base are located in the direction of the centre of gravity acceleration relative to one another.

[0007] The invention is based on the object of structurally and / or functionally improving the aforementioned methods. Furthermore, the invention is based on the object of providing and / or structurally and / or functionally improving an aforementioned device. Furthermore, the invention is based on the object of providing and / or structurally and / or functionally improving an aforementioned computer program.

[0008] The problem is solved by a method having the features of claim 1. Furthermore, the problem is solved by a method having the features of claim 10. Furthermore, the problem is solved by a method having the features of claim 11. Furthermore, the problem is solved by a method having the features of claim 12. Furthermore, the problem is solved by a device having the features of claim 13. Furthermore, the problem is solved by a computer program having the features of claim 17. Advantageous embodiments and / or further developments are the subject of the subclaims.

[0009] The method according to the invention is designed for generating and / or processing motion data of a freely moving body. The method can be implemented in a computer program. Motion data can be data that represent physical kinematic variables, in particular location, speed and / or acceleration. The kinematic variables can be described or have been described with magnitude, direction and orientation. The kinematic variables can be described or have been described vectorially. The body can be an object that can be demarcated from its environment and that has a mass and a volume. The body can be a physical body. The body can be a human body. The body can be a machine, in particular a mobile machine. The body can be a robot, in particular a humanoid robot.The body can be displaced in three independent directions and rotated about three independent axes. The body can have six degrees of freedom.

[0010] At least two sensor units are arranged on the body at a fixed distance from one another. The at least two sensor units can also be referred to as a first sensor unit and at least one further sensor unit. In particular, four sensor units can be arranged on the body. The four sensor units can be referred to as a first sensor unit, a second sensor unit, a third sensor unit, and a fourth sensor unit. The first sensor unit can be arranged at a first location. The second sensor unit can be arranged at a second location. The third sensor unit can be arranged at a third location. The fourth sensor unit can be arranged at a fourth location. The at least one further sensor unit can be arranged at at least one further location. The sensor units can be arranged in such a way that different and independent configurations for evaluating the angular acceleration vector are possible.

[0011] The at least two sensor units can be arranged directly on the body. The at least two sensor units can be arranged on the body with the aid of a sensor unit carrier. The sensor units each have at least one angular velocity sensor and at least one acceleration sensor. The at least one angular velocity sensor can be designed to spatially record angular velocities. The at least one acceleration sensor can be designed to spatially record accelerations. With the aid of the at least one angular velocity sensor, a rotation can be detected quantitatively and in a spatially directed and oriented manner and output as a measurement signal. With the aid of the at least one acceleration sensor, a translation can be detected quantitatively and in a spatially directed and oriented manner and output as a measurement signal.With the help of the first sensor unit, a first rotation and a first translation can be detected at the first location and output as a measurement signal. With the help of the second sensor unit, a second rotation and a second translation can be detected at the second location and output as a measurement signal. With the help of the third sensor unit, a third rotation and a third translation can be detected at the third location and output as a measurement signal. With the help of the fourth sensor unit, a fourth rotation and a fourth translation can be detected at the fourth location and output as a measurement signal. With the help of the at least one further sensor unit, at least one further rotation and at least one further translation can be detected at the at least one further location and output as a measurement signal.

[0012] Based on measurement signals from the at least two sensor units and applying the laws of general motion of rigid bodies, an angular acceleration vector is determined directly. An angular velocity vector can be determined based on a measurement signal from an angular velocity sensor. An angular acceleration vector can be determined based on a measurement signal from an angular velocity sensor of a sensor unit. A linear acceleration vector can be determined based on a measurement signal from an acceleration sensor of a sensor unit. The linear acceleration vector, the angular velocity vector, and the angular acceleration vector can be described or have been described in a Cartesian coordinate system. The linear acceleration vector, the angular velocity vector, and the angular acceleration vector can be described or have been described by coordinates.The linear acceleration vector, the angular velocity vector and the angular acceleration vector can be described or described three-dimensionally.

[0013] Laws of general motion of rigid bodies that are used to directly determine the angular acceleration vector are fundamental rigid body mechanics equations of translational dynamics and rotational dynamics, in particular Newton's laws and Euler's equations, in particular equations that combine translational dynamics and rotational dynamics of a rigid body, in particular Newton-Euler equations.

[0014] In this context, "direct" determination of the angular acceleration vector means, in particular, that the angular acceleration vector is not calculated "indirectly" by numerical differentiation of measurement signals. However, numerical differentiation of measurement signals can also be performed and additionally taken into account when determining the angular acceleration vector.

[0015] The body can have a first body part and at least one further body part. The at least one further body part can be movably connected to the first body part. The at least two sensor units can be arranged on the first body part. At least one further sensor unit can be arranged on the at least one further body part. The at least one further sensor unit can be a sensor unit corresponding to the at least two sensor units. The at least one further sensor unit can have at least one angular velocity sensor and at least one acceleration sensor. The angular acceleration vector can be determined taking into account measurement signals from the at least one further sensor unit.

[0016] Redundant measurement signals can be generated using the sensor units. A measurement signal is redundant if a corresponding measurement signal for determining the angular acceleration vector already exists. Redundant measurement signals can be fused. Redundant measurement signals can be fused by first determining the coordinates of the angular acceleration vector in several independent ways and then determining the coordinate values ​​by averaging. Separate angular velocity vectors and measured three-dimensional linear acceleration vectors at different locations, in particular at four different locations, can be used to determine the angular acceleration vector.

[0017] One of the at least two sensor units can be defined and / or used as a hierarchically superior sensor unit. The remaining sensor units can be defined and / or used as hierarchically subordinate sensor units. Connection vectors emanating from the hierarchically superior sensor unit to the hierarchically subordinate sensor units can be linearly independent. The definition and / or use of the sensor units can be changed. The definition and / or use of the sensor units can be changed regularly and / or cyclically.

[0018] Motion data based on the measurement signals from the sensor units and / or the angular acceleration vector can be processed directly. Motion data based on the measurement signals from the sensor units and / or the angular acceleration vector can be processed in real time. Motion data based on the measurement signals from the sensor units and / or the angular acceleration vector can be stored for separate processing. Motion data based on the measurement signals from the sensor units and / or the angular acceleration vector can be processed separately in terms of time and / or location.

[0019] Based on the measurement signals from the sensor units and / or the angular acceleration vector, a spatial orientation, an angular velocity, and / or a linear acceleration of the body can be determined. A spatial orientation, an angular velocity, and / or a linear acceleration of the body can be determined as a rigid-body approximation.

[0020] When generating and / or processing motion data, at least one further piece of motion-based information can be taken into account in addition to the measurement signals from the sensor units and / or the angular acceleration vector. The at least one further piece of motion-based information can be provided using a sensor device, for example, a force plate or a magnetometer. The at least one further piece of motion-based information can be provided by calculation. For example, the at least one further piece of motion-based information can be a calculated and / or estimated spatial orientation, a calculated and / or estimated angular velocity, and / or a calculated and / or estimated linear acceleration.

[0021] Based on the measurement signals of the sensor units and / or the angular acceleration vector, a ground reaction force and / or its point of application can be determined.

[0022] Body movement data generated and / or processed according to such a method can be used to record and / or evaluate movement sequences of a freely moving body. A method according to the invention is designed for recording and / or evaluating movement sequences of a freely moving body. Body movement data generated and / or processed according to such a method can be used to detect and / or evaluate dynamic instabilities of a freely moving body. A method according to the invention is designed for detecting and / or evaluating dynamic instabilities of a freely moving body. The movement data can be used in the field of medical rehabilitation, in particular for the qualitative evaluation of adaptation phenomena after training, even under the effect of external perturbations.The movement data can be used to determine stability / instability, safety, joint loads, movement effectiveness and / or movement coordination, for example when walking, running, sprinting, jumping, carrying, lifting, etc., and / or to determine movement kinematics and loading on both the upper and lower body.

[0023] Body motion data generated and / or processed according to such a method can be used to generate and / or optimize algorithms and / or models for calculating the dynamics of freely moving bodies. A method according to the invention is designed to generate and / or optimize algorithms and / or models for calculating the dynamics of freely moving bodies. Discrepancies between models and real systems (machine or human) can be reduced by optimizing the underlying structural, mass distribution, and behavioral parameters.

[0024] The device according to the invention is designed to carry out such a method. The device can be designed as a mobile device. The device has at least two sensor units. The at least two sensor units can also be referred to as a first sensor unit and at least one further sensor unit. In particular, the device can have four sensor units. The four sensor units can be referred to as a first sensor unit, a second sensor unit, a third sensor unit, and a fourth sensor unit. The sensor units each have at least one angular velocity sensor and at least one acceleration sensor. The at least one angular velocity sensor can be designed to spatially record angular velocities. The at least one acceleration sensor can be designed to spatially record accelerations.The at least one acceleration sensor can have three uniaxially active sensor elements. The at least two sensor units can each have a sensor housing. The at least two sensor units can be arranged on a freely movable body at a fixed distance from one another.

[0025] The device can comprise a sensor unit carrier. The sensor unit carrier can be rigid at least in sections and / or at least directionally. The sensor unit carrier can be arranged on the body. The sensor unit carrier can be arranged on a human torso and / or can be designed in a vest-like manner. The at least two sensor units can be arranged on the sensor unit carrier at a fixed distance from one another.

[0026] The device may comprise a computer. The computer may be designed to execute a computer program. The computer may comprise a processor, a main memory, at least one signal input, at least one signal output, and / or a data memory. The sensor units may be connected to the computer in a signal-conducting manner. The device may comprise an electrical power supply. The electrical power supply may be embodied as a battery or accumulator.

[0027] The sensor units can each have a predetermined maximum mass relative to the body or a body part. The predetermined maximum mass of all sensor units of the device can be 100%, in particular 80%, in particular 60%, in particular 40%, in particular 20%, of the weight of the body or body part. The sensor units can each have a predetermined minimum distance from one another. The predetermined minimum distance can correspond to the size of the smallest possible two-point dimension of a sensor unit.

[0028] The computer program comprises program code sections with which such a method can be executed when the computer program is executed using such a device. The method can be implemented in the computer program.

[0029] In summary and in other words, the invention thus results in, among other things, an evaluation of instabilities in movements on a living object and a more precise analysis of dynamics on real machines / people using sensors for the additional measurement of the angular velocity vector for segments that can move arbitrarily in space or an "instability detection" for a person and dynamics calculation on real machines using sensors for measuring the kinematic state, among other things the angular acceleration vector, on body segments that can move arbitrarily.

[0030] Instabilities during human locomotion can be evaluated under arbitrarily acting external perturbations. Adaptation phenomena can be demonstrated through training. A more precise evaluation of the dynamics of living beings and real systems using a rigid body approximation can be enabled. An angular acceleration vector can be generated for bodies / body segments moving arbitrarily in space. A measurement can be based on a combination of angular velocity sensors. Their measurement signals can be processed using fundamental equations of rigid-body mechanics. The required physical quantities can be an angular velocity vector and a linear acceleration vector. Only physical quantities measured by the angular velocity sensors at a specific time can be used.Sampling can be performed at a rate such that the body's angular acceleration vector can be determined in real time. The method can be performed without filtering. The method can be performed without any numerical differentiation of measurement signals.

[0031] Dynamics can be determined. Derived variables such as center of pressure, ground reaction force, and / or stability criteria can be determined. A sensor can be mounted per movable segment. Measurement signals for linear acceleration, angular velocity, and / or spatial orientation can be generated redundantly. Sensor fusion can be enabled.

[0032] Four angular velocity sensors can be mounted on a rigid mounting fixture. One sensor can be designated as the root sensor. The choice can be arbitrary and can be cyclically swapped for sensor fusion. The connection vectors between the root sensor and the three other angular velocity sensors should be linearly independent. This ensures the three-dimensional measurement of angular acceleration. The angular velocity sensors can be spaced at predetermined minimum distances from each other. Using a variably designed mount, the sensors can be securely mounted on a body or body segment.

[0033] The method and / or device can be used to detect instabilities during locomotion, to qualitatively assess the severity of the instability, and to observe adaptation phenomena after training in humans, even under the influence of external perturbations. The measurement signals thus available, such as orientation in space, angular velocity and linear acceleration vectors, and the angular acceleration vector calculated by the sensors, can provide a basis. The method can be used for more precise calculations of the dynamics of real systems and living beings. Using the method and in combination with an optimization algorithm and a force plate, models of objects, particularly models containing structure, inertial data, and / or biomechanical data, can be optimized. The sensors can be wearable on relevant body segments thanks to a non-slip attachment.

[0034] The invention enables an increased sampling rate. Real-time determination of an angular acceleration vector is made possible. Delay-free execution of the method is made possible. Filtering processes can be eliminated. Numerical differentiation can be eliminated. Resistance to white noise is increased. Noise is reduced. Online evaluation of instabilities caused, for example, by external perturbations is made possible. Training effects and / or adaptation phenomena in a person can be quantitatively assessed. Statements about the severity of instability caused by a perturbation can be made individually for each test subject. The accuracy of dynamic calculations can be increased. The failure of individual sensor units can be compensated for. Measurement results can be improved. Location-independent use is made possible. The transfer from simulation to reality can be improved.The discrepancy between model and reality can be reduced, enabling independent functionality without the need for additional equipment.

[0035] In the following, embodiments of the invention are described in more detail with reference to figures, which show schematically and by way of example: Fig. 1 a device with a sensor unit carrier and four sensor units for generating and / or processing movement data and Fig. 2 an arrangement of four sensor units on a body for generating and / or processing movement data.

[0036] Fig. Figure 1 shows a device 100 with a sensor unit carrier 102 and four sensor units 104, 106, 108, 110 for generating and / or processing motion data. The sensor unit carrier 102 serves to hold the sensor units 104, 106, 108, 110 firmly arranged relative to one another and to arrange them on a human torso. For this purpose, the sensor unit carrier 102 is designed as a vest and has a rigid sensor holding section. The sensor holding section can be manufactured, for example, using a 3D printing process.

[0037] The sensor units 104, 106, 108, 110 each have an angular velocity sensor and an acceleration sensor, with the angular velocity sensor serving to spatially record angular velocities and the acceleration sensor serving to spatially record accelerations. The sensor units 104, 106, 108, 110 are arranged on the sensor holding section of the sensor unit carrier 102 at a fixed distance from one another at a predetermined minimum distance such that, starting from the sensor unit 104, the sensor units 106, 108, 110 lie on axes 112, 114, 116 that are approximately perpendicular to one another.

[0038] The device 100 also includes a computer 118, in this case a single-board computer such as a Raspberry Pi, and a power bank as an electrical power supply 120. The sensor units 104, 106, 108, 110 are signal-conductively connected to the computer 118 via USB connections 122, 124, 126, 128.

[0039] Fig. Figure 2 shows an arrangement of the four sensor units 104, 106, 108, 110 on a body 130 for generating and / or processing motion data. One of the sensor units, for example, sensor unit 104, is defined and used as a hierarchically superior sensor unit. The remaining sensor units 106, 108, 110 are defined and used as hierarchically subordinate sensor units. The definition and use of the sensor units 104, 106, 108, 110 is subject to regular change.

[0040] Starting from an origin O of a Cartesian coordinate system, the hierarchically superior sensor unit 104 is arranged at a location 0, a position of the sensor unit 104 is described by a position vector r0. The hierarchically subordinate sensor units 106, 108, 110 are arranged at locations 1, 2 and 3, the positions of the sensor units 106, 108, 110 are described by the connecting vectors r i described.

[0041] The sensor units 104, 106, 108, 110 are arranged in such a way that regardless of which of the sensor units 104, 106, 108, 110 is defined and used as the hierarchically superior sensor unit, the connection vectors r i are linearly independent from the higher-level sensor unit to the hierarchically subordinate sensor units.

[0042] A computer program is installed on the computer 118. This program includes program code sections that can be used to directly determine an angular acceleration vector based on measurement signals from the sensor units 104, 106, 108, 110 and applying the laws of general motion of rigid bodies when the computer program is executed. Using the angular acceleration vector, a spatial orientation, an angular velocity, and / or a linear acceleration of the body are determined. This allows movement sequences of the body 130 to be recorded and / or dynamic instabilities of the body 130 to be detected and evaluated.

[0043] "May" refers in particular to optional features of the invention. Accordingly, there are also further developments and / or embodiments of the invention that additionally or alternatively comprise the respective feature(s).

[0044] If necessary, features can also be selected from the combinations of features disclosed in the description and used alone or in combination with other features to define the subject matter of the claim, dissolving any structural and / or functional connection that may exist between the features. Reference symbol 100 device 102 sensor unit carriers 104 Sensor unit 106 Sensor unit 108 Sensor unit 110 Sensor unit 112 Axis 114 Axis 116 Axis 118 computers 120 Energy supply 122 USB connection 124 USB connection 126 USB connection 128 USB connection 130 bodies

Claims

[1] Method for generating and / or processing movement data of a freely movable body (130), wherein at least two sensor units (104, 106, 108, 110), each having at least one angular velocity sensor and at least one acceleration sensor, are arranged on the body (130) at a fixed distance from one another, and an angular acceleration vector is determined directly on the basis of measurement signals from the at least two sensor units (104, 106, 108, 110) and using laws of general motion of rigid bodies. [2] Method according to claim 1, characterized bythat the body (130) has a first body part and at least one further body part movably connected to the first body part, the at least two sensor units (104, 106, 108, 110) are arranged on the first body part, at least one further sensor unit having at least one angular velocity sensor and at least one acceleration sensor is arranged on the at least one further body part, and the angular acceleration vector is determined taking into account measurement signals from the at least one further sensor unit. [3] Method according to at least one of the preceding claims, characterized by that redundant measurement signals are generated and / or redundant measurement signals are fused using the sensor units (104, 106, 108, 110). [4] Method according to at least one of the preceding claims, characterized bythat one of the at least two sensor units (104, 106, 108, 110) is defined and / or used as a hierarchically superior sensor unit (104) and the remaining sensor units (106, 108, 110) are defined and / or used as hierarchically subordinate sensor units (106, 108, 110). [5] Method according to claim 4, characterized by that a definition and / or use of the sensor units (104, 106, 108, 110) is changed. [6] Method according to at least one of the preceding claims, characterized by that movement data based on the measurement signals of the sensor units (104, 106, 108, 110) and / or the angular acceleration vector are processed directly and / or stored for separate processing. [7] Method according to at least one of the preceding claims, characterized bythat a spatial orientation, an angular velocity and / or a linear acceleration of the body (130) is / are determined on the basis of the measurement signals of the sensor units (104, 106, 108, 110) and / or the angular acceleration vector. [8] Method according to at least one of the preceding claims, characterized by that when generating and / or processing movement data, at least one further piece of movement-based information is taken into account in addition to the measurement signals of the sensor units (104, 106, 108, 110) and / or the angular acceleration vector. [9] Method according to at least one of the preceding claims, characterized by that a ground reaction force and / or its point of application is / are determined on the basis of the measurement signals of the sensor units (104, 106, 108, 110) and / or the angular acceleration vector. [10] Method for detecting and / or evaluating movement sequences of a freely movable body (130), characterized by that movement data of the body (130) generated and / or processed according to a method according to at least one of claims 1 to 9 are used. [11] Method for detecting and / or evaluating dynamic instabilities of a freely moving body (130), characterized by that movement data of the body (130) generated and / or processed according to a method according to at least one of claims 1 to 9 are used. [12] Method for generating and / or optimising algorithms and / or models for calculating the dynamics of freely moving bodies (130), characterized by that movement data of the body (130) generated and / or processed according to a method according to at least one of claims 1 to 9 are used. [13] Device (100) designed to carry out a method according to at least one of the preceding claims, characterized bythat the device (100) has at least two sensor units (104, 106, 108, 110) which can be arranged at a fixed distance from one another on a freely movable body (130) and each have at least one angular velocity sensor and at least one acceleration sensor. [14] Device (100) according to claim 13, characterized by that the device (100) has a sensor unit carrier (102) which is rigid at least in sections and / or at least in a direction-dependent manner and which can be arranged on the body (130), and the at least two sensor units (104, 106, 108, 110) are arranged on the sensor unit carrier (102) at a fixed distance from one another. [15] Device (100) according to at least one of claims 13 to 14, characterized by that the device (100) comprises a computer and / or an electrical power supply. [16] Device (100) according to at least one of claims 13 to 15, characterized bythat the sensor units (104, 106, 108, 110) each have a predetermined maximum mass relative to the body (130) or a body part and / or a predetermined minimum distance from one another. [17] Computer program, characterized by that the computer program comprises program code sections with which a method according to at least one of claims 1 to 12 can be carried out when the computer program is executed with the aid of a device (100) according to at least one of claims 13 to 16.

Images