Method for commissioning a sensor unit, sensor unit, and movement unit comprising a sensor unit
The method automatically aligns sensor units with the actual motion vector using a combination of acceleration and environment-based sensors, addressing measurement inaccuracies and environmental challenges for precise motion detection in mobile machinery.
Patent Information
- Application Number
- PCT/EP2026/062916
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-05-05
- Filing Date
- 2026-05-05
- Publication Date
- 2026-11-12
AI Technical Summary
Existing sensor units for detecting the movement of mobile machinery suffer from measurement inaccuracies due to manual alignment issues, misalignment of mounting points, and unreliable external references, particularly in challenging environments.
A method for commissioning a sensor unit that automatically aligns the sensor coordinate system with the actual motion vector using a combination of an acceleration sensor and an environment-based sensor, such as radar, to correct for assembly errors and environmental conditions, enabling precise and reliable determination of motion.
Enables precise and reliable detection of motion vectors with automatic calibration, compensating for mounting misalignments and environmental interference, ensuring accurate speed and motion measurements even under variable conditions.
Smart Images

Figure EP2026062916_12112026_PF_FP_ABST
Abstract
Description
[0001] Method for commissioning a sensor unit, sensor unit and motion unit with sensor unit
[0002] The present invention relates to a method for commissioning a sensor unit, as well as a sensor unit and a motion unit with a sensor unit.
[0003] In the prior art, sensor units for detecting the movement of a motion unit, in particular a mobile working machine or a mobile component of a working machine, are known, wherein the sensor unit may comprise an inclination sensor and an acceleration sensor. The sensor unit is connected to the motion unit.
[0004] Typically, the sensor unit for detecting translational movements is aligned manually, which can lead to significant measurement inaccuracies, especially with complex motion sequences or misaligned mounting points. Manual adjustment is also error-prone and time-consuming, as the actual motion vector cannot always be precisely determined. Additionally, measured acceleration values can be distorted by sensor misalignment.
[0005] For example, in practice it often happens that the sensor unit on a mobile device is not mounted exactly along the direction of travel, which can lead to an offset between the actual motion vector of the machine and the orientation of the sensor coordinate system.
[0006] Further methods are known in the art that combine measurement data with data from an external reference system, such as a global navigation satellite system, to compensate for assembly inaccuracies or system-related drift during operation. However, satellite signals are often not reliably available, especially for mobile machinery such as agricultural or construction equipment, for example in forests, near buildings, or in valleys. Furthermore, wheel-based speed sensors are known, but these are often inaccurate on machinery due to high wheel slip on unpaved, muddy, or wet surfaces. Therefore, there is a need for a robust and reliable method that operates accurately even under these challenging conditions.
[0007] The object of the present invention is to provide a method for commissioning a sensor unit which, while avoiding the disadvantages known from the prior art, enables a precise and reliable determination of the movement of a motion unit. In particular, it aims to enable fast, reliable, and automated axis adjustment of the sensor coordinate system to a motion vector of the motion unit.
[0008] Furthermore, the task includes providing a sensor unit and a motion unit.
[0009] The problem is solved with regard to the method by the features of claim 1, with regard to the sensor unit by the features of claim 10 and with regard to the motion unit by the features of claim 12.
[0010] Advantageous embodiments of the invention are specified in the dependent claims. According to the invention, a method for commissioning a sensor unit is claimed, comprising an acceleration sensor and / or preferably an inclination sensor, which is / are connected to a motion unit in any mounting orientation, and wherein the method comprises the following commissioning steps in a preferred sequence to perform a translational axis calibration of a sensor coordinate system:
[0011] - Execution of a translational movement of the movement unit, especially at least temporarily
[0012] - Capturing the acceleration and / or rotational velocity values occurring during translational movement, preferably also inclination values, in order to determine a motion vector, - Teaching in a correction factor to transform an axis of the original sensor coordinate system in the direction of the determined
[0013] Motion vector. It is preferably provided that the speed of the motion unit is determined by means of the accelerometer and / or by means of an environment-based sensor, in particular a radar sensor, wherein in the event of a failure, unusability, or interference of a sensor signal of the environment-based sensor, in particular a radar signal, the speed determination is carried out at least temporarily, preferably completely to bridge the interference, based on acceleration values that are transformed in the direction of the motion vector by means of the learned correction factor. Thus, in the event of interference, the speed is determined at least temporarily by means of the accelerometer.
[0014] In other words, the sensor unit is preferably aligned in a fixed, but arbitrary, Cartesian sensor coordinate system. Targeted translational movements of the machine or the affected body part are used to acquire measurements, from which the effective motion vector is determined. A correction factor can then be derived from this, transforming the original sensor coordinate system to the direction of movement. In particular, this allows the actual motion vector to be determined, whereby a correction factor can be derived using trigonometric calculations, especially an arctangent function to determine the yaw angle for movements in the x-direction.
[0015] A correction factor within the meaning of the present invention is preferably understood to be a mathematical transformation rule designed to adapt the sensor coordinate system, at least with respect to one axis, to a determined translational motion vector of the motion unit. The correction factor can, in particular, be configured as a rotation angle, a direction vector, or a transformation matrix and describes a spatial relationship between the original sensor coordinate system and the actual direction of motion of the motion unit.
[0016] Preferably, it can be determined from the rotational speed values or rotation rates of the tilt sensor whether a translational movement is taking place, from which the motion vector can subsequently be derived. In particular, a translation can be detected if the rotational speed values approach zero.
[0017] Preferably, automated calibration is enabled, allowing precise detection of translational movements even with an initial sensor offset from the motion vector – with the acceleration values being further corrected by the roll and pitch values determined by the tilt sensor. In a preferred first step, a translational movement of the motion unit is performed, at least temporarily, in the actual direction of movement. In a preferred second step, the relevant measurement data during the movement – in particular the acceleration and / or rotational velocity values – are acquired. This data is evaluated to determine the actual motion vector and to optimally adapt the sensor coordinate system to the actual direction of movement.In the preferred third step, a correction factor is taught in to transform an axis of the original sensor coordinate system to the determined translational motion vector. This conversion is preferably performed using a transformation matrix, enabling precise measurement of the translational motion regardless of the initial mounting orientation. The correction factor can also be manually validated or adjusted if necessary. For further optimization, the measured acceleration values can preferably be corrected using the angle values from the tilt sensor to ensure precise alignment of the sensor coordinate system with the actual motion vector under varying operating conditions.
[0018] Preferably, the motion unit can be a mobile component of a machine and / or a body part of a mobile machine. The method according to the invention has the advantage of automatic calibration and adaptation of the sensor coordinate system to an actual direction of movement, so that the sensor unit can be installed in any given mounting position, even if this is spatially offset from an ideal axis of movement of the motion unit. This allows the sensor unit to be used for the precise measurement and adjustment of the axis position during translational movements of the motion unit.
[0019] In other words, the method enables automatic axis calibration and compensation for assembly errors. It analyzes acceleration and / or rotational speed values to detect or identify sections of pure linear motion along one of the main axes of the motion unit. These sections serve as a reference. Trigonometric calculations based on this reference motion determine a correction factor that not only aligns the sensor data with the true direction of motion during this phase, but also corrects systematic deviations in sensor orientation due to assembly inaccuracies, preferably by incorporating additional tilt values.
[0020] An environment-based sensor according to the present invention is a sensor configured to acquire information about the vehicle or motion environment of the motion unit and to derive a relative motion between the motion unit and the environment from this information. The environment-based sensor uses the vehicle environment as an external reference and is preferably functionally independent of the accelerometer. The acquisition of the environment information can be active or passive.
[0021] The term "environment-based sensor" encompasses, in particular, sensors that scan the environment by transmitting and receiving signals, such as radar, lidar, ultrasonic, and / or time-of-flight sensors, as well as imaging sensors that determine relative motion based on image sequences or environmental features. This can also include camera-based time-of-flight sensors.
[0022] Such an environment-based sensor can preferably be configured to acquire velocity information through active interaction with the environment of the moving unit. In particular, the sensor is configured to transmit signals to the environment, preferably the ground, and to receive signals reflected from the environment in order to derive a relative motion between the moving unit and the environment. Such sensors include, for example, radar, lidar, and / or ultrasonic sensors. Based on this relative motion, velocity information for the moving unit is determined.
[0023] The combination of a calibrated accelerometer with such an environment-based sensor leads to an unexpected increase in precision and reliability. This is particularly relevant for field machinery, where the functionality of environment-based sensors is often impaired by environmental conditions, especially wet surfaces, loose ground, and dense vegetation. The invention solves this problem through a synergistic fusion: The calibrated inertial system can provide reliable velocity information, especially when the signal from the environment-based sensor fails momentarily or becomes implausible. Conversely, when the environment-based sensor measures a valid signal, it provides a precise, preferably drift-free, reference to the actual velocity of the moving unit.
[0024] It should be taken into account that the environment-based sensor preferably provides an absolute, drift-free reference of the motion of the motion unit, while the accelerometer offers high resolution, particularly for short-term or dynamic movements, but may be prone to drift in the long term. The calibration and combination of both sensor systems according to the invention achieves a synergistic effect, compensating for the respective disadvantages of the individual sensors and combining their advantages, thus enabling a particularly robust and reliable determination of speed and motion.
[0025] Accelerometers provide highly dynamic measurements, which, however, can be subject to drift when integrated over time to determine velocity due to offset and noise components. Conversely, environment-based sensors, such as radar sensors, provide immediate and drift-free velocity information, although its availability can depend on external environmental conditions. The environment-based sensor is preferably designed to detect the velocity of the moving unit relative to its environment as the primary measurement, with the environment serving as an external spatial reference. For this purpose, the environment-based sensor preferably emits electromagnetic or acoustic signals, particularly in the direction of the ground, and evaluates reflected or backscattered signal components to determine immediate, non-integrated, and therefore drift-free velocity information.In particular, with an environment-based sensor designed as a radar sensor and / or ultrasonic sensor, the speed is determined using a.
[0026] Doppler frequency shift of the reflected signals. In contrast, the accelerometer is preferably used only as a supplement, particularly when the sensor signal of the environment-based sensor is unavailable or unusable due to external environmental conditions, insufficient signal quality, or temporary interference. In this case, the velocity is determined at least temporarily based on the acceleration values of the accelerometer, thus ensuring continuous operation even in the event of a failure or limited usability of the environment-based sensor.
[0027] In summary, the method according to the invention enables advantageously automatic and reliable calibration of the sensor coordinate system based on actual translational movements of the motion unit, without requiring a defined installation position or manual adjustment. This allows mounting misalignments and operational changes in orientation to be compensated for, ensuring precise detection of the actual direction and speed of movement even under difficult environmental conditions.
[0028] According to a preferred embodiment, the acceleration sensor is used to determine the speed of the motion unit, wherein, following commissioning, the detected acceleration values are continuously adjusted by the current inclination values, in particular inclination angles, in order to correct an influence on the speed determination due to a changing orientation of the sensor coordinate system relative to a gravitational vector.
[0029] Advantageously, this also ensures precise measurement of translational acceleration even under variable operating conditions.
[0030] In particular, even if the commissioning is affected by the movement unit traveling over a hill, a precise speed calculation can still be performed.
[0031] In other words, following the correction of mounting offsets, the acceleration values are continuously adjusted during operation using the current tilt angles. This allows for a precise determination of the actual translational acceleration even under variable operating conditions – such as cornering, uneven terrain, or other dynamic changes in tilt.
[0032] Preferably, acceleration and / or rotational speed values are evaluated to detect a temporary translational movement segment of the motion unit. This offers the advantage of flexible calibration by enabling the detection and adaptation to any movement. Therefore, it is not necessary for the movement to be predefined for calibration. In other words, the movement can be performed arbitrarily, with a time-limited translational movement segment being automatically detected and evaluated. An operator performing the movement thus does not need to pay attention to a precise direction of movement. Preferably, inclination values can also be used for reference or correction.
[0033] Any movement can be determined by measuring the rotation rate of the tilt sensor. In particular, a rotation rate approaching zero indicates a purely translational movement.
[0034] According to a further development, it may be preferred that the speed of the motion unit is determined by means of the acceleration sensor and / or by means of an environment-based sensor, in particular a radar sensor, wherein in the event of a failure or unusability of a sensor signal of the environment-based sensor, in particular a radar signal, the speed determination is carried out at least temporarily on the basis of acceleration values which are transformed in the direction of the motion vector by means of the learned correction factor and corrected on the basis of current inclination values.
[0035] Advantageously, in the event of a failure or detected unusability of the signal from the environment-based sensor, the velocity determination is performed, at least temporarily, solely based on the acceleration values. These are transformed, as previously described, using the learned correction factor and preferably, as also previously described, corrected based on current inclination values to ensure highly accurate inertial velocity determination.
[0036] Preferably, the usability of a sensor signal, in particular a radar signal, is evaluated on the basis of at least one sensor signal quality parameter, wherein the sensor signal quality parameter includes a received strength, a signal-to-noise ratio, a number of valid signal reflections, a stability of a determined speed signal and / or a temporal consistency of several speed measurements, and that, depending on the sensor signal quality parameter, the system automatically switches between an environment-based speed determination, an acceleration-based speed determination and / or a combined speed determination.
[0037] To intelligently control the fusion of the two sensor systems, the usability of the signal from the environment-based sensor is preferably evaluated using at least one signal quality parameter. This ensures that the best possible speed information is used at all times.
[0038] For further development, it may be preferable to measure the speed of the moving unit using a radar sensor and preferably also, or at least temporarily, using an accelerometer. This improves the reliability and accuracy of the speed measurement, whereby the two measurement methods can partially overlap redundantly or only be used temporarily if one of the measurement methods fails. For example, the speed measurement using the accelerometer can then be used if the radar signal fails or cannot be evaluated for speed measurement. This can occur, for example, if a reflected radar signal is disrupted due to external environmental conditions, especially aqueous surfaces.In this context, it is preferred if the usability of the radar signal is evaluated on the basis of at least one radar signal quality parameter, wherein the radar signal quality parameter includes a received strength, a signal-to-noise ratio, a number of valid radar reflections, a stability of a determined speed signal and / or a temporal consistency of several radar measurements, and that, depending on the radar signal quality parameter, switching is automatically performed between a radar-based speed determination, an acceleration-based speed determination and / or a combined speed determination.
[0039] Preferably, the calibration can be carried out taking into account environmental conditions such as vibrations or uneven terrain by filtering or compensating the sensor data accordingly.
[0040] Additionally, it may be preferred that the translational movement in several axes of the sensor coordinate system is simultaneously detected and calibrated.
[0041] This allows for improved precision in complex motion profiles.
[0042] Furthermore, it is advantageous to read in pre-existing motion profiles of the motion unit. This can increase or simplify the efficiency of the calibration by utilizing existing motion profiles. In particular, the procedure can be adapted to different types of mobile machinery as motion units by using machine-specific motion profiles or calibration routines.
[0043] Furthermore, it is conceivable that the calibration could be performed in real time during the operation of the mobile machine. In this case, the calibration could be dynamically adapted to changing operating conditions.
[0044] The invention further relates to a sensor unit for determining the speed of a motion unit, which is connected to the motion unit in any mounting orientation and wherein the sensor unit is put into operation or is operated by means of a previously mentioned method.
[0045] Preferably, the sensor unit may include an accelerometer and an environment-based sensor, in particular a radar sensor, to determine the speed of the moving unit, wherein, in the event of a failure or unusability of a sensor signal of the environment-based sensor, in particular a radar signal, the speed determination is carried out at least temporarily on the basis of acceleration values.
[0046] Preferably, the sensor unit comprises an accelerometer and a radar sensor to determine the speed of the moving unit, using the speed values from the accelerometer if a radar signal is not available.
[0047] Furthermore, the invention also relates to a motion unit with a previously mentioned sensor unit, wherein the motion unit is designed as a mobile working machine.
[0048] The invention will now be explained in more detail using exemplary embodiments and with reference to the drawings.
[0049] They show schematically:
[0050] Fig. 1: Perspective view of a mobile work machine with an offset installed acceleration and tilt sensor,
[0051] Fig. 2: Perspective view of a moving body part of a construction machine with an offset acceleration and tilt sensor.
[0052] Figure 1 shows a motion unit 10, in particular a mobile field machine 14, with a sensor unit 16 mounted in an orientation that deviates from an ideal direction of motion b. The illustration demonstrates that a sensor coordinate system (x1,y1) does not exactly coincide with a motion vector (x2,y2) or a principal axis of the motion unit 10, which is a common occurrence in practice. This offset is a typical application scenario for a method of translational axis calibration, since the sensor unit 16, despite its spatially offset mounting, is intended to enable precise measurements of the actual direction of motion b.
[0053] For the commissioning of the sensor unit 16, comprising an accelerometer and preferably an inclination sensor, a calibration procedure is carried out to perform a translational axis calibration of the sensor coordinate system (x1,y1). The procedure comprises the following steps: performing a translational movement of the motion unit 10, in particular along the direction of movement b; recording the acceleration and / or rotational velocity values occurring during the translational movement, preferably also inclination values, in order to determine a motion vector (x2,y2), (x4,y4); and teaching in a correction factor to transform an axis of the original sensor coordinate system (x1,y1), (x3,y3) in the direction of the determined motion vector (x2,y2), (x4,y4).
[0054] Figure 2 shows another embodiment using the example of a crane 11. Here, the sensor unit 16 with acceleration and tilt sensor is also mounted in a non-ideal position on a crane receiver 12 as a motion unit 10, such that the sensor coordinate system (x3,y3) deviates from the intended axis of motion b of the crane receiver 12 (x4,y4). Figure 2 further illustrates that automatic calibration may also be necessary for a moving part of a construction machine in order to detect the actual translational motion vector (x4,y4) and to compensate for any mounting misalignments of the sensor unit 16.
[0055] Preferably, the speed of the motion unit 10 is determined by means of a radar sensor, wherein, at least in the event of a disturbance of the radar signal, the calibrated acceleration values of the accelerometer are used. Reference numeral list
[0056] 10 movement units
[0057] 11 crane
[0058] 12 crane handles
[0059] 14 mobile work machines for field use
[0060] 16 sensor units
[0061] b Axis of motion of the motion unit
[0062] (x1,y1), (x3,y3) Cartesian coordinate system of the sensor unit (x2,y2), (x4,y4) motion vector and Cartesian coordinate system of the motion unit
Claims
Patent claims 1. Method for commissioning a sensor unit (16), comprising an acceleration sensor and / or tilt sensor, which is / are connected in any mounting orientation to a motion unit (10), in particular a mobile working machine (14), and wherein the method comprises the following commissioning steps to perform a translational axis calibration of a sensor coordinate system (x1,y1), (x3,y3): • Performing a translational movement of the motion unit (10), • Recording the acceleration and / or rotational velocity values occurring during translational motion in order to determine a motion vector (x2,y2), (x4,y4), • Teaching in a correction factor to transform one axis of the original sensor coordinate system (x1,y1),(x3,y3) in the direction of the determined motion vector (x2,y2), (x4,y4), wherein the speed of the motion unit (10) is determined by means of the accelerometer and / or by means of an environment-based sensor, in particular a radar sensor, wherein in the event of a failure or unusability of a sensor signal of the environment-based sensor, in particular a radar signal, the speed determination is carried out at least temporarily on the basis of acceleration values which are transformed in the direction of the motion vector by means of the learned correction factor.
2. Method according to claim 1, characterized by that the acceleration sensor is used to determine the speed of the motion unit (10), wherein the sensor unit (16) additionally includes an inclination sensor, wherein, following commissioning, the detected acceleration values are continuously adjusted by current inclination values in order to correct an influence on the speed determination due to a changing orientation of the sensor coordinate system (x1,y1), (x3,y3) relative to a gravitational vector.
3. Method according to claim 1 or 2, characterized by that the acceleration and / or rotational velocity values are evaluated in order to detect a temporary translational movement section of the motion unit (10) and to capture the movement vector (x2,y2), (x4,y4) for this movement section in order to perform the calibration based on a freely chosen movement.
4. Method according to any one of claims 1 to 3, characterized by that in the event of a failure or unusability of a sensor signal of the environment-based sensor, in particular a radar signal, the speed determination is carried out at least temporarily on the basis of acceleration values, which are transformed in the direction of the motion vector by means of the learned correction factor and corrected on the basis of current inclination values.
5. Method according to claim 4, characterized by that the usability of a sensor signal, in particular a radar signal, is evaluated on the basis of at least one sensor signal quality parameter, wherein the sensor signal quality parameter includes a received strength, a signal-to-noise ratio, a number of valid signal reflections, a stability of a determined speed signal and / or a temporal consistency of several speed measurements, and that, depending on the sensor signal quality parameter, switching is performed, in particular automatically, between an environment-based speed determination, an acceleration-based speed determination and / or a combined speed determination.
6. Method according to any one of claims 1 to 5, characterized by that the speed of the motion unit (10) is measured by means of the accelerometer and additionally by means of a radar sensor in order to improve the precision and reliability of the speed determination.
7. Method according to any one of claims 1 to 6, characterized by that the translational movement in several axes of the sensor coordinate system (x1,y1), (x3,y3) is simultaneously detected and calibrated in order to precisely map complex motion profiles.
8. Method according to any one of claims 1 to 7, characterized by that previously known motion profiles of the motion unit (10) are read in, which are taken into account when determining the motion vector (x2,y2), (x4,y4).
9. Method according to any one of claims 1 to 8, characterized by that the calibration is performed in real time during the operation of the motion unit (10), in particular a mobile working machine (14), in order to compensate for dynamic changes in the sensor position or the operating conditions.
10. Sensor unit for determining the speed of a motion unit (10), which is connected to the motion unit (10) in any mounting orientation, and wherein the sensor unit (16) is commissioned or operated by means of a method according to one of claims 1 to 9, wherein the sensor unit (16) comprises an acceleration sensor and an environment-based sensor, in particular a radar sensor, to determine the speed of the motion unit (10), wherein in the event of a failure or unusability of a sensor signal of the environment-based sensor, in particular a radar signal, the speed determination is carried out at least temporarily on the basis of acceleration values.
11. Sensor unit according to claim 10, characterized by that the sensor unit (16) has an accelerometer and a radar sensor to determine the speed of the motion unit (10), using the speed values of the accelerometer if a radar signal is not available.
12. Motion unit with a sensor unit (16) according to one of claims 10 or 11, wherein the motion unit (10) is designed as a mobile working machine (14) or a working machine with a mobile component, wherein the sensor unit (16) is connected to the mobile component or a body part of the mobile working machine (14) in any mounting orientation.
