Method and device for capturing an object

The device and method enhance LFI systems by expanding laser beams for overlapping ranges and processing spectral information to reconstruct 3D point clouds, addressing efficiency limitations and enabling precise, robust, and fast measurements on diverse surfaces.

WO2026233643A1PCT designated stage Publication Date: 2026-11-12ROBERT BOSCH GMBH
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/EP2026/060563
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-05
Filing Date
2026-04-13
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

Conventional laser feedback interferometry (LFI) setups can only determine a single distance value using three sensors, limiting their efficiency in generating a point cloud.

Method used

A device and method that utilizes multiple LFI sensors with an optical unit to expand laser beams into a common overlapping range, processing spectral information to reconstruct a three-dimensional point cloud using a computationally efficient algorithm, potentially incorporating VCSEL sensors and a Levenberg Marquardt algorithm for minimization.

Benefits of technology

Enables the construction of an inexpensive, high-integrability, miniaturized 3D sensor with robustness to ambient light, capable of high-precision, non-contact, high-speed measurements on various surfaces, including glass and semiconductor structures, and compensates for environmental influences.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure EP2026060563_12112026_PF_FP_ABST
    Figure EP2026060563_12112026_PF_FP_ABST
Patent Text Reader

Abstract

The invention relates to a device (1) and a method for capturing an object (OBJ) by means of: a plurality of sensors (2) which each have a laser beam source (2A) for emitting a laser beam and a detector (2B) for detecting reflected laser beams, an optical unit (3) which is designed to expand the laser beam emitted by the laser beam source (2A) of a sensor (2) in such a way that the sensors (2) have a common overlapping measurement range (ÜMB), and an information processing unit (4) which is designed to process spectral information which is provided by the detectors (2B) of the sensors (2) in order to reconstruct a three-dimensional point cloud of the object (OBJ).
Need to check novelty before this filing date? Find Prior Art

Description

[0001] R.418009

[0002] - 1 -

[0003] Description

[0004] title

[0005] Method and device for detecting an object

[0006] The invention relates to a method and a device for detecting an object based on laser feedback interferometry (LFI) sensor technology.

[0007] State of the art

[0008] Laser feedback interferometry (LFI) has a wide range of applications. German patent DE10 2021 207540 A1, for example, describes a method for determining eye position. In this method, laser feedback interferometry measurements from an eye are received by a laser interferometry unit, with the measurements being based on at least one laser signal reflected from a component of the eye.

[0009] Laser feedback interferometry (LFI) is based on the interference of light, allowing for extremely precise measurement of minute deviations in the optical path length. A monochromatic laser source emits a coherent beam of light with a known wavelength, which can be reflected by an object. The laser wavelength can be modulated to measure distances. By analyzing interference patterns, changes in length or the position of the object being measured can be precisely determined. Tiny changes in the optical path length result in visible interference shifts, which can be detected by high-precision detectors. Fig. 1 illustrates a distance measurement, R.418009

[0010] - 2 -

[0011] which is based on the amplitude of a beat frequency fb. Figure 1 shows the spectrum of an LFI sensor with a very thin laser beam.

[0012] The conventional measuring setup shown in Fig. 2 allows the distance to an object OBJ to be measured. Using three LFI sensors, and given the known orientation of the sensors SEN relative to each other and their pose (e.g., via a normal vector SEN1n – SEN3n), a distance measurement can be performed, and the intersection point SCH-P shown in Fig. 2 can be reconstructed in three-dimensional space. As shown in Fig. 2, the sensors SEN direct a narrow measuring laser beam onto the object OBJ. A 3D measurement value can thus be generated. In the context of sensors SEN, the pose typically refers to the position and orientation of an object or system in space. The pose is described by parameters such as the three-dimensional coordinates (x, y, z) and the orientation (roll, pitch, yaw).

[0013] However, a disadvantage of the conventional measuring arrangement shown in Fig. 2 is that only a single distance value can be determined using three LFI sensors, and not a point cloud. Consequently, the conventional measuring arrangement shown in Fig. 2 is not efficient.

[0014] Disclosure of the invention

[0015] According to a first aspect, the invention provides a device for detecting an object with several sensors, each comprising a laser light source for emitting a laser beam and a detector for detecting reflected laser beams; an optical unit designed to extend the laser beam emitted by the laser light source of a sensor in such a way that the sensors have a common overlapping measuring range; and an information processing unit designed to process spectral information provided by the detectors of the sensors for the reconstruction of a three-dimensional point cloud of the object.

[0016] In the measuring device according to the invention, the laser beams are widened to increase the measuring range. A 3D point cloud is generated based on spectral R.418009.

[0017] - 3 -

[0018] Information provided by multiple static LFI sensors is calculated. This allows for the construction of an inexpensive 3D sensor device with high integrability and / or miniaturization potential.

[0019] The provided spectral information or data can be processed using a computationally efficient algorithm to reconstruct the 3D point cloud.

[0020] The measuring device according to the invention is also characterized by a high robustness to interference from ambient light of the measuring principle due to the coherent measuring method used.

[0021] In one possible embodiment of the device according to the invention for detecting an object, the sensors of the device have laser feedback interferometry (LFI) sensors.

[0022] The use of laser feedback interferometry (LFI) sensors offers several technical advantages that make them attractive for high-precision measurement applications.

[0023] LFI sensors offer high precision and resolution. They can measure movements and changes in length in the nanometer or even sub-nanometer range. Extremely small displacements can be detected through the use of laser interference.

[0024] LFI sensors allow for non-contact measurement. Because the measurement is based on laser light, it is contactless. This means there are no mechanical influences on the object being measured, and sensitive samples remain undamaged.

[0025] LFI sensors enable high-speed, real-time measurements. They allow for fast measurements with high sampling rates, making them ideal for dynamic processes or high-speed applications. R.418009

[0026] - 4 -

[0027] The interference method ensures consistent and stable measurements over long periods. Furthermore, temperature fluctuations and environmental influences can be compensated for through appropriate calibration.

[0028] LFI sensors are also small and can be easily integrated into existing systems.

[0029] LFI sensors enable measurements on various surfaces. They function on both reflective and partially transparent surfaces. Even challenging materials like glass or semiconductor structures can be measured.

[0030] In most applications, no calibration is necessary. LFI sensors can operate directly using the laser wavelength as a reference, thus eliminating the need for complex calibrations.

[0031] In one possible embodiment of the device according to the invention for detecting an object, the optical unit comprises an optical lens or a meta-optic component.

[0032] In one possible embodiment of the inventive device for detecting an object, an optical unit is provided for each sensor, which is designed to expand the laser beam emitted by the sensor's light source into a cone shape. This allows for individual expansion of the laser beam for each individual sensor.

[0033] In a possible alternative embodiment of the inventive device for detecting an object, a common optical unit is provided which is designed to expand the laser beams emitted by the laser beam sources of the sensors into a cone shape. This offers the advantage that only one optical component is required. This saves space and simplifies the assembly of the device. R.418009

[0034] - 5 -

[0035] In one possible embodiment of the inventive device for detecting an object, the sensors utilize VCSEL (Vertical Cavity Surface Emitting Laser) sensors. This saves space and facilitates integration.

[0036] In one possible embodiment of the inventive device for detecting an object, geometric information is stored for each sensor, which is processed by the information processing unit to reconstruct the three-dimensional point cloud of the object. The generated three-dimensional point cloud can be temporarily stored and processed in turn, for example to generate control signals in response to the detected object.

[0037] The geometric information relating to the sensors can be stored in a data memory of the device. This allows for easy configuration and reconfiguration of the device according to the invention for different sensor arrangements and applications.

[0038] In one possible embodiment of the device according to the invention for detecting an object, the information stored for each sensor includes a position of the sensor, a measuring radius of the sensor and opening angle of the laser beam emitted by the sensor and expanded in a conical shape.

[0039] In one possible embodiment of the inventive device for detecting an object, the information processing unit is designed to calculate the three-dimensional point cloud of the object based on the spectral information provided by the sensor detectors and on the basis of geometric information stored for the sensors. The calculation is preferably performed in real time.

[0040] In one possible embodiment of the device according to the invention for detecting an object, the information processing unit is designed to generate the three-dimensional point cloud of the object by executing an R.418009, based on the spectral information provided by the detectors of the sensors and on the basis of geometric information stored for each sensor.

[0041] - 6 -

[0042] The minimization algorithm is calculated iteratively. The minimization algorithm can be executed on a processor of the information processing unit.

[0043] Different minimization algorithms can be executed for different use cases.

[0044] In one possible embodiment of the device according to the invention for detecting an object, the minimization algorithm executed by the information processing unit is a Levenberg Marquardt algorithm (LMA).

[0045] In one possible embodiment of the device according to the invention for detecting an object, the emission wavelength of the laser beam emitted by the laser light source of the sensor is continuously changed.

[0046] Furthermore, according to another aspect, the invention provides a method for capturing an object comprising the following steps:

[0047] Expanding laser beams emitted from laser beam sources of sensors by an optical unit, such that the sensors have a common overlapping measuring range;

[0048] Detection of reflected laser beams by sensor detectors to provide spectral information; and

[0049] Processing the spectral information provided by the sensor detectors to reconstruct a three-dimensional point cloud of the object.

[0050] In one possible embodiment of the inventive method for detecting an object, the three-dimensional point cloud of the object is calculated on the basis of the spectral information provided by the detectors of the sensors and on the basis of geometric information stored for the sensors.

[0051] In one possible embodiment of the inventive method for detecting an object, the three-dimensional point cloud of the object is processed by R.418009

[0052] - 7 -

[0053] Execution of a minimization algorithm, in particular a Levenberg Marquardt algorithm, computed iteratively.

[0054] Further possible embodiments, developments, and implementations of the invention also include combinations of features of the invention described previously or subsequently with regard to the exemplary embodiments, even if not explicitly mentioned. In particular, the person skilled in the art will also consider individual aspects as improvements or additions to the respective basic form of the present invention. The above embodiments and developments can be combined with one another as appropriate.

[0055] The present invention will be explained in more detail below with reference to the exemplary embodiments shown in the schematic figures of the drawings.

[0056] This shows:

[0057] Fig. 1 shows a spectrum by representation of a conventional measurement method;

[0058] Fig. 2 shows a view illustrating a conventional measuring arrangement; Fig. 3 shows an embodiment of a device according to the invention for detecting an object;

[0059] Fig. 4 shows an embodiment of a device according to the invention for capturing an object;

[0060] Fig. 5A, 5B Diagrams to explain the functioning of the device according to the invention for detecting an object;

[0061] Fig. 6 shows a spectrum to illustrate the functioning of the device according to the invention for detecting an object;

[0062] Fig. 7 is a diagram illustrating the functioning of the device according to the invention for capturing an object; R.418009

[0063] - 8 -

[0064] Fig. 8 is a diagram to explain the functioning of the device according to the invention for detecting an object;

[0065] Fig. 9 shows an embodiment of a device according to the invention for capturing an object;

[0066] Fig. 10 shows an embodiment of a device according to the invention for capturing an object;

[0067] Fig. 11 is a flowchart illustrating an embodiment of the inventive method for detecting an object.

[0068] The accompanying drawings are intended to provide a further understanding of the embodiments of the invention. They illustrate embodiments and, in conjunction with the description, serve to explain the principles and concepts of the invention. Other embodiments and many of the advantages mentioned will become apparent with reference to the drawings. The elements of the drawings are not necessarily shown to scale.

[0069] In the figures of the drawing, identical, functionally equivalent and similarly acting elements, features and components - unless otherwise stated - are each provided with the same reference symbols.

[0070] The invention provides a device 1 for detecting an object OBJ. Figures 3 and 4 show possible embodiments of the sensor device 1 according to the invention. The device 1 comprises several sensors 2, each of which has a laser beam source 2A for emitting a laser beam and a detector 2B for detecting reflected laser beams. In the embodiment shown in Figure 3, the device 1 comprises three spaced-apart sensors 2-1, 2-2, 2-3. The number of sensors 2 can vary depending on the application. The positions of the sensors 2-1, 2-2, 2-3 are defined and preferably stored in a configuration memory 5. The sensor device 1 according to the invention shown in Figure 3 can be integrated into a mobile unit, for example, a robotic vacuum cleaner or the like, for detecting objects in the vicinity of the mobile unit. R.418009

[0071] - 9 -

[0072] An optical unit 3 is designed to expand the laser beam emitted by the laser light source 2A of a sensor 2 such that the various sensors 2 have a common overlapping measurement range ÜMB. In one possible embodiment, the optical unit 3 can be integrated into the sensor 2. Alternatively, the optical unit 3 is located in front of the laser light source 2A of the sensor 2, as shown in Fig. 4. The optics preferably cause a cone-shaped expansion of the laser beam emitted by the laser light source 2A of the sensor 2. In the measuring arrangement shown in Fig. 3, three light cones K1, K2, K3 are directed at the object OBJ, overlapping within an overlapping measurement range ÜMB.

[0073] The device 1 includes an information processing unit 4, which is designed to process spectral information provided by the detectors 2B of the sensors 2 for the reconstruction of a three-dimensional (3D) point cloud of the object OBJ. In one possible embodiment of the device 1 according to the invention for detecting an object, geometric information is stored for each sensor 2, which is processed by the information processing unit 4 to reconstruct the three-dimensional point cloud of the object OBJ. The geometric information relating to the sensors 2 can be stored in a data memory 5 of the device 1. This allows for easy configuration and reconfiguration of the device 1 according to the invention for different sensor arrangements and applications.

[0074] In one possible embodiment of the device 1 according to the invention for detecting an object OBJ, the information stored for each sensor 2 includes the position of the sensor 2, a measuring radius r of the sensor 2, and the opening angle of the laser beam emitted by the sensor 2 and expanded in a conical shape. In one possible embodiment of the device 1 according to the invention for detecting an object, the information processing unit 4 is designed to perform calculations based on the spectral information provided by the detectors 2B of the sensors 2 and on the basis of the information for each sensor 2.

[0075] - 10 -

[0076] to calculate the three-dimensional point cloud of the object OBJ from the stored geometric information.

[0077] The information processing unit 4 is designed to iteratively calculate the three-dimensional point cloud of the object OBJ by executing a minimization algorithm, based on the spectral information provided by the detectors of the sensors 2 and on the geometric information stored for each sensor 2. The minimization algorithm can be executed on a processor of the information processing unit 4.

[0078] Different minimization algorithms can be executed for different use cases.

[0079] In one possible embodiment of the device 1 according to the invention for detecting an object OBJ, the minimization algorithm executed by the information processing unit 4 is a Levenberg Marquardt algorithm (LMA).

[0080] In one possible embodiment of the device 1 according to the invention for detecting an object OBJ, an emission wavelength of the laser beam emitted by the laser beam source 2A of the sensor 2 is continuously changed.

[0081] Furthermore, according to another aspect, the invention provides a method for detecting an object with the following steps, as they are also schematically depicted in the flowchart according to Fig. 11.

[0082] In a first step S1, laser beams emitted by laser beam sources 2A of sensors 2 are widened by an optical unit 3 such that the sensors 2 have a common overlapping measuring range ÜMB.

[0083] In a second step S2, reflected laser beams are detected by detectors 2B of sensors 2 to provide spectral information.

[0084] In a third step S3, the spectral information provided by detectors 2B of sensors 2 is processed to reconstruct an R.418009.

[0085] - 11 -

[0086] three-dimensional point cloud of object OBJ. The three-dimensional point cloud of object OBJ is calculated based on the spectral information provided by detectors 2B of sensors 2 and on the basis of geometric information stored for sensors 2.

[0087] Fig. 3 shows an embodiment of the measuring device 1 according to the invention. In the illustrated embodiment, the device 1 comprises three LFI sensors 2-1, 2-2, 2-3, which are arranged relative to one another such that they have a common overlapping measuring range ÜMB. The overlapping measuring range ÜMB is shown schematically in Fig. 3. In the illustrated embodiment, each of the three sensors 2-i projects a cone-shaped laser beam onto the object OBJ. In one possible embodiment, the laser beam of a VCSEL can be expanded or narrowed by optics such that a target measuring range is achieved.

[0088] As schematically shown in Fig. 4, the laser beam emitted by the laser beam source 2A can, in one possible embodiment, be focused by a lens as an optical unit 3 into theta 0 and phi. <p Richtung um einen Winkel aufgeweitet werden. Dies kann bei einer möglichen Implementierung durch eine zylindrische Linse oder durch eine einfache Linse mit verschiedenen Brennweiten in x- und y-Richtung erfolgen.

[0089] Instead of a closed expansion using a normal diffractive or otherwise constructed lens, an alternative embodiment allows the use of a meta-optic component as the optical unit 3 to generate multiple laser beams, for example, a beam grid for point distribution in a plane or in space. The LFI sensors 2 consist of a laser light source 2A and a detector 2B. The detector 2B can be formed by a photodiode PD located in or adjacent to the laser cavity. In one possible embodiment, the laser light source 2A can be integrated into a VCSEL, as schematically illustrated in Fig. 10.

[0090] - 12 -

[0091] The laser is able to measure distance by detecting the interference of the emitted and backscattered light and measuring the beat frequency fb via detector 2B. The following relationship applies:

[0092]

[0093] The amplitude of the photodiode current ID corresponds to an optical power P0(k), where k is the wavenumber, damped by internal and external losses. Furthermore, the amplitude of the photodiode current ID is frequency-modulated by the cosine term. The distance d also appears here. In simplified terms, this results in:

[0094]

[0095] This formula is correct for a simple, infinitesimally small laser beam measuring an object OBJ. Using a Fourier transform, the distance-dependent frequency modulation of the photodiode current ID can be measured, and the beat frequency fb, which corresponds to the distance d, can be determined.

[0096] The wavenumber k= 2π / λ plays a crucial role in laser feedback interferometry (LFI), especially in conjunction with the Henry linewidth enhancement factor a and the resulting beat frequency fb.

[0097] The optical frequency of the laser can shift due to optical feedback. This shift depends on the optical phase d, which is related to the optical path length L of the feedback light via the wavenumber k:

[0098] d>= k L

[0099] Since the beat frequency fb depends on the temporal change of the phase d>, it follows that:

[0100] dΦ / dt =k · dL / dtR.418009

[0101] - 13 -

[0102] If we insert this into the relationship for the frequency shift Af due to the a-factor:

[0103] Δf = α / 4π · k · dΦ / dt

[0104] You will receive:

[0105] Δf = α / 4π · k · dL / dt

[0106] Since the beat frequency fb in LFI is typically determined by the frequency shift Af, the following applies:

[0107] fb ~ α / 2π · k · dL / dt

[0108] The wavenumber k amplifies the effect of the Henry factor a on the beat frequency.

[0109] About the equation:

[0110] fb = d ext αk / π

[0111] The distance d ext can be calculated from the measured beat frequency fb. In laser feedback interferometry (LFI), the beat frequency fb typically depends on the Henry linewidth enhancement factor (a-factor). This factor describes how strongly the laser linewidth is influenced by the optical gain in the semiconductor laser. Generally, a higher a-factor leads to greater line broadening of the laser spectrum, which in turn increases the bandwidth of the generated beat frequencies fb. Therefore, the larger the a-factor, the wider the generated beat frequency fb is usually in LFI.

[0112] In laser feedback interferometry (LFI), the beat frequency (fb) refers to the difference frequency between two coherent laser beams, generated by interference patterns. This frequency arises when the two laser beams collide and produce interference phenomena. (R.418009)

[0113] - 14 -

[0114] Beat frequency fb can be used to measure tiny changes in length or movement, as it is sensitive to phase changes that may be caused by changes in distance d or in the reflection of surfaces.

[0115] The measurement signal of a sensor 2 for a widened laser beam looks like this for, e.g., two planes with d1 and d2:

[0116]

[0117] + 2√η ref η ext,1 cos(2kd ext,1 )

[0118] + 2√η ref η ext,2 cos(2kd ext,2 )]

[0119] This implies that the same distances in the spectrum correspond to the same frequency. Furthermore, only unique distances to the object OBJ have an assignable frequency. The measurement signal exhibits a superposition of frequencies, resulting in the loss of the correlation between frequencies f and their respective distances d. Consequently, it is unclear how many equal distances d constitute a beat frequency fb (e.g., when illuminating a sphere, equidistant rings appear on the sphere's surface, all at the same distance from the laser).

[0120] Fig. 5A, 5B show a 2D example of equidistant measurements.

[0121] Fig. 5A shows an LFI sensor SEN that emits cone-shaped expanded laser beams,

[0122] Fig. 5B shows the distance d in mm as a function of a theta increment of the elevation angle theta starting from the forward axis of the sensor.

[0123] Curve I in Fig. 5B shows all distances d that can be measured along the eye. Curve II in Fig. 5B shows the distances actually measured. R.418009

[0124] - 15 -

[0125] Only the distances shown in curve II are measured by the LFI sensor 2. The corresponding spectrum is shown in Fig. 6. As can be seen in Fig. 6, the spectrum widens over the distance measurement range. For each bin in the spectrum, a corresponding distance can be calculated as described above. The amplitude can be used to estimate how many targets are required for the respective distance. A bin in a frequency spectrum denotes a discrete frequency range or interval in which the spectral analysis is performed. The width of a bin depends on the sampling rate and the length of the analyzed signal. The longer the time window, the finer the frequency resolution. Each bin contains an amplitude (and often a phase) that indicates the strength of the corresponding frequency in the signal.

[0126] In order to nevertheless measure the distance d to a measuring point on the target object OBJ, in a preferred embodiment, as shown schematically in Fig. 3, three sensors 2-1, 2-2, 2-3 are provided, or at least two sensors if information about the target object OBJ is available (for example, the target is a plane or has a flat surface).

[0127] For the first sensor 2-1, a solution sphere shell segment is obtained for a given distance (d1 meas) with a radius r, as schematically shown in Fig. 7. This intersects the target or object OBJ at N = {1, 2, M} points, depending on the geometry of the object OBJ illuminated by the first sensor 2-1. For the measuring range in which distances d are measured (dränge) for the second sensor 2-2, a space of solution sphere shell segments is obtained.

[0128] For the measured distance d1 meas from sensor 2-1, there is an intersection point SP between two shells in this space of solution sphere shell segments, where the radius d2meas is a free parameter for which this intersection point SP arises. Particularly with spherical target objects OBJ, depending on the positioning of sensors 2, at least two further intersection points SP can occur, meaning that the 3D point is not yet uniquely described in three spatial dimensions. For this reason, a third space of solution sphere shell segments is preferably defined starting from a third sensor 2-3R.418009

[0129] - 16 -

[0130] spanned. This is particularly important when more complex geometries of the OBJ objects are expected.

[0131] The three sensors 2-1, 2-2, 2-3 then yield the solution sphere shell segments shown in Fig. 8. The solution is determined by defining it as a minimization problem.

[0132] In one possible embodiment, the geometric information stored for sensors 2 includes:

[0133] SEN 2-1:

[0134] Sphere center (sensor position (x,y,z))

[0135] Radius r1 dlmeas

[0136] Opening angle of cone K1 [0min, 0max][phi_min, phi_max ].

[0137] SEN 2-2:

[0138] Sphere center (sensor position (x,y,z))

[0139] radius range [rmin, rmax] dmin dmax from dmeas

[0140] Opening angle of cone K2 [0min, 0max][phi_min,phi_max].

[0141] SEN 2-3:

[0142] Sphere center (sensor position (x,y,z))

[0143] radius range [rmin, rmax] dmin dmax from dmeas

[0144] Opening angle of the cone K3 [0min, 0max][phi_min,phi_max].

[0145] A point (x, y, z) is sought for the above-mentioned given parameters.

[0146] Spherical conditions in three-dimensional space:

[0147] Each point must be on or near a sphere.

[0148] For the first sensor 2-1, the following applies:

[0149]

[0150] The radius r:R.418009 is obtained for the second sensor 2-2.

[0151] - 17 -

[0152]

[0153] There is a lower limit and an upper limit to the radius r:

[0154] rradius, lower = max(0, r) min,i − r)

[0155] 'radius, upper = IIiax(0, T ^max,i)

[0156] Furthermore, the following cone conditions apply with regard to the opening cone K of the laser beam.

[0157] The intersection point SP must lie within the opening cone K of each sensor SEN.

[0158] Spatial coordinates can be defined relative to the respective sensor SEN:

[0159] Elevation angle starting from the forward axis of the sensor SEN < J> Azimuth angle in the local plane of the sensor SEN

[0160] Deviations:

[0161] For ö and <t>:

[0162]

[0163] The residual vector r (x,y,z) combines all residuals of pairwise sections of spheres or spheres SPR.418009

[0164]

[0165] ^sphere, 1

[0166] rradius, lower, 2rradius, upper, 2rradius, lower, 3rradius, upper, 3

[0167]

[0168] Each pair of balls or spheres SP intersects in a circle or at a single point.

[0169] For two spheres SPi and SPj defined by their centers Ci = (xi,yi,zi) and Cj = (xj,yj,zj) and associated radii ri and rj, the following conditions must be met.

[0170] The distance d between the two centers Ci, Cj must satisfy the following condition:

[0171] |r i - r j | ≤ d ≤ r i + r j ,

[0172] where d represents the Euclidean distance between the two centers Ci, Cj

[0173]

[0174] If the above condition is met, the following case distinction applies.

[0175] Falls:R.418009

[0176] - 19 -

[0177]

[0178] then the two spheres SPi, SPj intersect at exactly one point;

[0179] |r i − r j | < d < r i + r j

[0180] then the two spheres SPi, SPj intersect in a circle;

[0181] both balls SPi, SPj at one point;

[0182]

[0183] Then the two spheres SPi and SPj do not intersect or touch each other.

[0184] For three spheres SP1, SP2, and SP3, their respective intersections must be consistent. This means that the centers C and circular intersections must be co-planar.

[0185] The three centers C1, C2, C3 of the three spheres SP1, SP2, SP3 and their pairwise circular intersections must define a single plane. If the centers C are not coplanar, there can be no common intersection point SCH-P.

[0186] The intersection point SCH-P of the three spheres SP1, SP2, SP3 must satisfy the equations of all three spheres:

[0187]

[0188] If a solution (x,y,z) exists for this, it represents the intersection point SCH-P. R.418009

[0189] - 20 -

[0190] For minimization, one possible implementation uses a Levenberg-Marquardt algorithm (LMA) (another minimization algorithm such as Gauss-Newton would also work), which initializes iteratives based on a starting value, which can be determined, for example, for a mean distance from the sensors using the triangulation described earlier. The Levenberg-Marquardt algorithm (LMA) is an iterative optimization method used for nonlinear curve fitting. It combines the gradient descent method and the Newton method to find a robust and efficient solution for nonlinear least-squares problems.

[0191] The algorithm then iterates until a minimum is found. Once the sensor reaches its minimum, it has determined the radii d2 and d3, which are then used to determine the point (x,y,z) in space.

[0192] The algorithm can then be executed for the next dmeasl along dränge for sensor 2-1 until a solution has been found for all distances in the spectrum of sensor 2-1. To maximize the number of data points in 3D, the distance range between dmin and dmax can also be resampled with an arbitrarily fine grid. For each distance measurement in d1, the LM minimizer is started to find the corresponding point in space that is also detected by the other sensors.

[0193] Fig. 9 shows a possible embodiment of a measuring arrangement according to the invention with several sensors 2 under a common lens 3, in particular as a package with an ASIC.

[0194] Fig. 10 shows the VIP (VCSEL with integrated photodiode PD - here photodiode PD and laser) and the measurement method (LFI) from coupled emissions and returning laser beams into the laser cavity.

[0195] A VCSEL (Vertical Cavity Surface Emitting Laser) has a laser diode LD as its laser beam source 2A, which emits light in a cone-shaped beam vertically from the surface of a fabricated wafer. The VCSEL also has an integrated photodiode PD as a detector 2B. The VCSEL combines both a laser beam source 2A and an R.418009

[0196] - 21 -

[0197] Detector unit 2B in a single component. The VCSEL includes a semiconductor laser 2A that emits its light perpendicular to the surface of the chip. In one possible embodiment, the semiconductor laser 2A consists of several semiconductor layers, including two distributed Bragg reflectors (DBRs) that form the optical resonator structure.

[0198] A semiconductor photodiode PD, acting as detector 2B, is integrated directly below or next to the VCSEL. It can detect either the light reflected or transmitted by the VCSEL. The photodiode PD converts the received light into an electrical signal. This signal can be sampled and converted into measured values. The photodiode PD can monitor the emitted light and provide feedback to stabilize the laser power. Combining the laser beam source 2A and detector 2B on a single chip reduces the required space. Because the VCSEL operates at low currents, its energy consumption is low.

[0199] The VCSEL operates on the principle of the self-mixing effect. The VCSEL emits a narrowband laser beam perpendicularly from the surface of the component. In the device 1 according to the invention, the emitted laser beam is widened by optics and then strikes the object OBJ to be measured, whereby part of the light is reflected from its surface.

[0200] The light reflected from the object OBJ is coupled back into the laser resonator. This feedback causes the feedback light to interfere with the laser light originally present in the resonator. The superposition of the coupled light with the internal light field alters the phase and amplitude characteristics of the VCSEL. This modulation leads to measurable changes in the laser's output power, frequency, or phase.

[0201] In the weak feedback regime, the modulation is nearly linear and proportional to the phase shift, enabling precise distance measurements. Preferably, a photodiode PD is integrated into the same chip, which directly detects the modified output signal. By analyzing the self-mixing signals, precise information about the position, velocity, or vibration of the object OBJ can be extracted. A laser 2A with optical output power P0 emits an R.418009

[0202] - 22 -

[0203] a coherent laser beam directed towards the target or target object OBJ.

[0204] Fig. 10 illustrates the operation of a VIP and the resulting measurement method (Laser Feedback Interferometry (LFI)). The laser beam strikes the target object OBJ at an angle of incidence Y and is attenuated by volume scattering effects and absorption, described by a reflectivity R, and re-coupled into the laser. Text denotes the time the laser beam requires to travel the distance Lext to the target object OBJ. The returning light interferes with the electric field oscillating in the laser cavity, resulting in a modulation of the optical power.

[0205] By controlling the emission wavelength (Frequency-Modulated Continuous Wave (FMCW)), the change in distance between the laser and the target object OBJ can now be determined (more accurately) by reading the photodiode PD. A change in distance also leads to a measurable shift in wavelength (Doppler effect), which allows the velocity of the target object OBJ to be calculated.

[0206] The sensor device 1 according to the invention can be used in a variety of ways, particularly in mobile units. Possible areas of application include, for example, AR / VR headsets or robotic vacuum cleaners.

[0207] Although the present invention has been fully described above with reference to preferred embodiments, it is not limited thereto, but can be modified in many different ways.< / t>

Claims

R.418009 - 23 - Claims:

1. Device (1) for detecting an object (OBJ) with: several sensors (2), each comprising a laser beam source (2A) for emitting a laser beam and a detector (2B) for detecting reflected laser beams; an optical unit (3) designed to extend the laser beam emitted by the laser beam source (2A) of a sensor (2) such that the sensors (2) have a common overlapping measurement range (SMR); and with an information processing unit (4) designed to process spectral information provided by the detectors (2B) of the sensors (2) to reconstruct a three-dimensional point cloud of the object (OBJ).

2. Device for detecting an object according to claim 1, wherein the sensors (2) comprise laser feedback interferometry (LFI) sensors.

3. Device for detecting an object according to one of the preceding claims, wherein the optical unit (3) comprises an optical lens or a meta-optic component.

4. Device for detecting an object according to any one of the preceding claims 1 to 3, wherein for each sensor (2) an associated optical unit (3) is provided which is designed to expand the laser beam emitted by the light source (2A) of the sensor (2) in a conical shape. R.418009 - 24 - 5. Device for detecting an object according to any one of the preceding claims 1 to 3, wherein a common optical unit (3) is provided which is designed to expand the laser beams emitted by the laser beam sources (2A) of the sensors (2) in a conical shape.

6. Device for detecting an object according to any one of the preceding claims 1 to 5, wherein the sensors (2) comprise VCSEL (Vertical Cavity Surface Emitting Laser) sensors.

7. Device for detecting an object according to any one of the preceding claims 1 to 6, wherein geometric information is stored for each sensor (2) which is processed by the information processing unit (4) to reconstruct the three-dimensional point cloud of the object (OBJ).

8. Device for detecting an object according to claim 7, wherein the information stored for each sensor (2) comprises a position of the sensor (2), a measuring radius of the sensor (2) and opening angle of the laser beam emitted by the sensor (2) and expanded in a cone shape.

9. Device for detecting an object according to any one of the preceding claims 1 to 8, wherein the information processing unit (4) is designed to calculate the three-dimensional point cloud of the object (OBJ) on the basis of the spectral information provided by the detectors (2B) of the sensors (2) and on the basis of geometric information stored for the sensors (2).

10. Device for detecting an object according to claim 9, wherein the information processing unit (4) is designed to determine the three-dimensional R.418009 based on the spectral information provided by the detectors (2B) of the sensors (2) and on the basis of geometric information stored for the sensors (2). - 25 - To iteratively calculate the point cloud of the object (OBJ) by executing a minimization algorithm.

11. Device for detecting an object according to claim 10, wherein the minimization algorithm executed by the information processing unit (4) is a Levenberg Marquardt algorithm (LMA).

12. Device for detecting an object according to one of claims 1 to 11, wherein an emission wavelength of the laser beam emitted by the laser beam source (2A) of the sensor (2) is continuously changed.

13. Procedure for capturing an object with the following steps: Expanding (S1) of laser beams emitted from laser beam sources (2A) of sensors (2) by an optical unit (3) such that the sensors (2) have a common overlapping measurement range (TMR); Detection (S2) of reflected laser beams by detectors (2B) of the sensors (2) to provide spectral information; and Processing (S3) the spectral information provided by the detectors (2B) of the sensors (2) to reconstruct a three-dimensional point cloud of the object (OBJ).

14. Method for detecting an object according to claim 13, wherein the three-dimensional point cloud of the object (OBJ) is calculated on the basis of the spectral information provided by the detectors (2B) of the sensors (2) and on the basis of geometric information stored for the sensors (2). R.418009 - 26 - 15. Method for capturing an object according to claim 14, wherein the three-dimensional point cloud of the object (OBJ) is iteratively calculated by executing a minimization algorithm, in particular a Levenberg Marquardt algorithm.