Three-dimensional sensing system, three-dimensional sensing method and sweeping robot

Abstract

The embodiment of the invention provides a three-dimensional sensing system, a three-dimensional sensing method and a sweeping robot. The three-dimensional sensing system comprises a zoom polarized light projector which comprises a light source module, a zoom collimating lens group and a diffractive optical element; the light source module comprises a first light-emitting area and a second light-emitting area, and each light-emitting area comprises a light-emitting hole for emitting light in the first polarization direction and a light-emitting hole for emitting light in the second polarization direction; the zoom collimating lens group is used for collimating light rays emitted from the light source module and changing the projection depth of field through focal length adjustment; the zoom polarized light projector is configured as follows: when first depth-of-field projection is carried out, the zoom collimating lens group is adjusted to a first focal length, and the first light-emitting area is lightened, so that a first structured light pattern is projected; when second depth-of-field projection is carried out, the zoom collimating lens group is adjusted to a second focal length, and the second light-emitting area is lightened, so that a second structured light pattern is projected; the two structured light patterns have the same field angle and speckle density.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the technical field of structured light, and more particularly, to a three-dimensional sensing system, a three-dimensional sensing method, and a robot sweeper. BACKGROUND

[0002] With the development of technologies such as autonomous driving, robot navigation, and face recognition, the demand for three-dimensional sensing systems is growing. Structured light technology has become one of the mainstream active three-dimensional sensing solutions due to its high precision and reliability. In this technology, a projector projects a specific structured light pattern such as a speckle onto a target space, and a receiver captures the deformed pattern modulated by the object. Through calculation, the three-dimensional information of the object can be obtained.

[0003] In application scenarios such as robot sweepers, it is necessary to both map the distant space and accurately avoid obstacles in the close range. This requires the three-dimensional sensing system to have different measurement depths of field. Existing technologies such as patent application N115989970A usually use two independent projectors to be responsible for the far and near depths of field. However, this solution has problems such as high cost, large system size, and complex light path calibration. In addition, in this existing technology, the field of view FOV and the speckle density of the patterns projected by the two projectors are usually different. When switching the depth of field for three-dimensional reconstruction, the algorithm needs to adapt to two different pattern parameters, increasing the computational complexity, and even causing "black holes" (data missing) in the depth map due to pattern mismatch. Another improvement solution is to set a zoom lens after the diffractive optical element DOE of a single projector, and to achieve different depth of field projection by changing the focal length. However, this way changes the characteristics of the outgoing light beam, resulting in different FOV and speckle density of the projected pattern at different focal lengths, which cannot solve the above algorithm adaptation problem. At the same time, zooming after the DOE will introduce additional optical aberrations, reducing the spot quality and energy utilization.

[0004] In addition, when detecting high-reflectivity objects such as mirrors and glass, the strong specular reflection light will cause the receiving end of the photosensitive chip to be overexposed, drowning the diffuse reflection information of the object surface, resulting in depth calculation failure.

[0005] Therefore, there is an urgent need for a new three-dimensional sensing solution that can solve the above problems. SUMMARY

[0006] One purpose of the present application is to provide a new technical solution for a three-dimensional sensing system, aiming to overcome the shortcomings of the prior art, and to provide a three-dimensional sensing system that can achieve different depth of field projection with a single light projector, and ensure the same field of view FOV and speckle density at different depths of field.

[0007] A second purpose of the present application is to provide a new technical solution for a three-dimensional sensing method.

[0008] The third object of the present application is to provide a new technical solution of a sweeping robot.

[0009] In a first aspect, the present application provides a three-dimensional sensing system, comprising a zoom polarized light projector; The zoom polarized light projector comprises: a light source module comprising a first light-emitting area and a second light-emitting area capable of independent control, each light-emitting area comprising a first light-emitting hole emitting light rays of a first polarization direction and a second light-emitting hole emitting light rays of a second polarization direction, wherein the first polarization direction and the second polarization direction are orthogonal to each other; a zoom collimating lens group arranged on the light-emitting side of the light source module for collimating the light rays emitted from the light source module and changing the projection depth of field by focal length adjustment; a single diffractive optical element arranged on the light-emitting side of the zoom collimating lens group for diffracting the collimated light beam to form a structured light pattern; The zoom polarized light projector is configured to: when performing first depth of field projection, adjusting the zoom collimating lens group to a first focal length f1 and lighting the first light-emitting area to project a first structured light pattern; when performing second depth of field projection, adjusting the zoom collimating lens group to a second focal length f2 and lighting the second light-emitting area to project a second structured light pattern; wherein the first structured light pattern and the second structured light pattern have the same field of view angle and speckle density.

[0010] Optionally, the first light-emitting area and the second light-emitting area are arranged in a superimposed arrangement or a side-by-side arrangement.

[0011] Optionally, the zoom collimating lens group is one of a motor-driven lens group, a liquid lens, or a liquid crystal lens.

[0012] Optionally, the minimum lateral period Ph and the minimum vertical period Pv of the diffractive optical element satisfy the following relationship: Ph=λ / 2sin(artan(L1 / (2*f1))); Pv=λ / 2sin(artan(H1 / (2*f1))); wherein λ is the wavelength of the light rays emitted from the light source module, L1 and H1 are the length and width of the first light-emitting area respectively, and f1 is the first focal length.

[0013] Optionally, the length L2 and the width H2 of the second light-emitting area satisfy the following relationship: L2=2*f2*tan(arsin(λ / 2*Ph)); H2=2*f2*tan(arsin(λ / 2*Pv)); Where f2 is the second focal length, and satisfies f1 < f2 and H1*L1 < H2*L2.

[0014] Optionally, the three-dimensional sensing system further includes: A receiver is configured to receive a structured light pattern emitted from the zoom-polarized light projector and reflected back from the target object; and, An analyzer is placed in the incident light path of the receiver to select light rays that pass through a specific polarization direction.

[0015] Optionally, the analyzer is a rotatable linear polarizer or a liquid crystal polarization modulation device.

[0016] Optionally, the receiver is configured to: capture the structured light pattern reflected back by the target object in frames when the zoom collimating lens group is at the first focal length f1 or the second focal length f2; The three-dimensional sensing system also includes a processor connected to the receiver and configured to perform the following operations: The analyzer is controlled such that its transmission axis is parallel to the first polarization direction during the first frame acquisition, so that the receiver can acquire light rays in the first polarization direction and form a first image. The analyzer is controlled such that its transmission axis is parallel to the second polarization direction during the second frame acquisition, so that the receiver can acquire light rays in the second polarization direction and form a second image. The first image and the second image are fused together, and depth calculation is performed based on the fused image to reconstruct the three-dimensional information of the target object.

[0017] Optionally, the receiver further includes a lens group, a narrowband filter, and a photosensitive chip arranged sequentially along the optical path.

[0018] Secondly, this application provides a three-dimensional sensing method, the three-dimensional sensing method comprising: A zoom-polarized light projector is controlled to project a structured light pattern onto a target object by adjusting the focal length of its zoom collimating lens group to the target focal length and illuminating the light-emitting area corresponding to the target focal length; wherein the target focal length is a first focal length f1 or a second focal length f2, and the structured light pattern includes two types of light rays with a first polarization direction and a second polarization direction that are orthogonal to each other. The structured light pattern reflected from the target object is captured in frames using a receiver, an analyzer located on its incident light path, and a processor connected to it. The processor controls the polarizer to make the transmission axis parallel to the first polarization direction when collecting the first frame, so that the receiver acquires light rays of the first polarization direction to form a first image; and The processor controls the polarizer to make the transmission axis parallel to the second polarization direction when collecting the second frame, so that the receiver acquires light rays of the second polarization direction to form a second image. The processor fuses the first image and the second image, and performs depth calculation based on the fused image to reconstruct three-dimensional information of the target object.

[0019] In a third aspect, the present application provides a sweeping robot, which comprises: The three-dimensional sensing system according to the first aspect.

[0020] The present application has the following advantages: The three-dimensional sensing system provided by the present application realizes the projection of structured light at different depths by using the optical architecture combining the zoom collimator lens group and the single diffractive optical element DOE and matching the light source module controlled independently in different regions, only with a single projector. The core advantage of the optical design is that it can output structured light patterns with consistent field of view FOV and speckle point density at different focal lengths, which fundamentally ensures the consistency of the patterns required for three-dimensional imaging, significantly improves the accuracy of depth calculation and the adaptability of the system. At the same time, the integrated design effectively simplifies the system structure, reduces the hardware cost and calibration complexity, and provides an efficient and reliable sensing solution for various devices that need three-dimensional environmental perception.

[0021] Other features of the present application and its advantages will become apparent from the following detailed description of exemplary embodiments thereof, taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0022] The accompanying drawings incorporated in and forming a part of the specification, illustrate embodiments of the present application and, together with the description, serve to explain the principles of the present application.

[0023] Figure 1 One of the optical structures and optical path diagrams of the zoom polarized light projector of the three-dimensional sensing system provided by the present application (small focal length and large field of view); Figure 2 The second of the optical structures and optical path diagrams of the zoom polarized light projector of the three-dimensional sensing system provided by the present application (large focal length and small field of view); Figure 3 One of the structural schematic diagrams of the light source module of the zoom polarized light projector of the three-dimensional sensing system provided by the present application (the first light emitting area and the second light emitting area are arranged in a superposed manner); Figure 4 Structure diagram of light source module of zoom polarized light projector of three-dimensional sensing system provided by the embodiment of the present application (first light emitting area and second light emitting area arranged side by side); Figure 5 Simulation result diagram of three-dimensional sensing system provided by the embodiment of the present application under different focal lengths (1); Figure 6 Simulation result diagram of three-dimensional sensing system provided by the embodiment of the present application under different focal lengths (2); Figure 7 Optical structure and optical path diagram of receiver (polarization analyzer is a linear polarizer) of three-dimensional sensing system provided by the embodiment of the present application (1); Figure 8 Optical structure and optical path diagram of receiver (polarization analyzer is a linear polarizer) of three-dimensional sensing system provided by the embodiment of the present application (2); Figure 9 Working principle diagram of receiver (polarization analyzer is a liquid crystal polarization modulation device) of three-dimensional sensing system provided by the embodiment of the present application.

[0024] Explanation of reference signs: A, zoom polarized light projector; B, receiver; 1, light source module; 11, first light emitting area; 12, second light emitting area; a, first light emitting hole; b, second light emitting hole; 01, light of first polarization direction; 02, light of second polarization direction; 2, zoom collimator lens group; 3, diffractive optical element; 4, polarization analyzer; 5, lens group; 6, narrow-band filter; 7, photosensitive chip. DETAILED DESCRIPTION

[0025] Various exemplary embodiments of the present application will now be described in detail with reference to the accompanying drawings. Note that the relative arrangement, numerical expressions, and numerical values of components and steps set forth in these embodiments are illustrative only and do not limit the scope of the present application unless otherwise specifically stated.

[0026] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way limiting of the application or its applications or uses.

[0027] Techniques and equipment known to those of ordinary skill in the relevant art can not be discussed in detail, but should be considered part of the specification where appropriate.

[0028] In all examples shown and discussed herein, any specific values should be interpreted as merely illustrative and not as limiting. Thus, other examples of the exemplary embodiments can have different values.

[0029] It should be noted that like reference numerals and letters refer to like items throughout the several views, and once an item is defined in one view, it should not have to be discussed further in subsequent views.

[0030] The three-dimensional sensing system, the three-dimensional sensing method and the sweeping robot provided by the embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0031] According to one embodiment of the present application, a three-dimensional sensing system is provided, referring to Figure 1 and Figure 2 , the three-dimensional sensing system comprises a zoom polar light projector A, the zoom polar light projector A comprises a light source module 1, a zoom collimating lens group 2 and a single diffractive optical element 3. The light source module 1 comprises a first light-emitting area 11 and a second light-emitting area 12 which can be independently controlled, each light-emitting area comprises a first light-emitting hole a emitting light rays 01 of a first polarization direction and a second light-emitting hole b emitting light rays 02 of a second polarization direction, wherein the first polarization direction and the second polarization direction are orthogonal to each other. The zoom collimating lens group 2 is arranged on the light-emitting side of the light source module 1, used for collimating the light rays emitted from the light source module 1, and changing the projection depth by adjusting the focal length. The single diffractive optical element 3 is arranged on the light-emitting side of the zoom collimating lens group 2, used for diffracting the collimated light beam to form a structured light pattern. The zoom polar light projector A is configured to: when performing first depth of field projection, adjust the zoom collimating lens group 2 to a first focal length f1 and light up the first light-emitting area 11 to project a first structured light pattern; when performing second depth of field projection, adjust the zoom collimating lens group 2 to a second focal length f2 and light up the second light-emitting area 12 to project a second structured light pattern; wherein the first structured light pattern and the second structured light pattern have the same field of view angle and speckle density.

[0032] The three-dimensional sensing system provided by the embodiments of the present application has one of the core optical components as a zoom polar light projector A. From the optical structure, it can be known in combination with Figure 1 and Figure 2 that the zoom polar light projector A mainly consists of a light source module 1, a zoom collimating lens group 2 and a single diffractive optical element 3. The above three elements in the zoom polar light projector A are described in detail below.

[0033] The light source module 1 design comprises two independently controllable light-emitting areas, which are a first light-emitting area 11 and a second light-emitting area 12. Specifically, referring to Figure 3 and Figure 4Each of the light-emitting areas in the light source module 1 is mixedly arranged with a first light-emitting hole a capable of emitting light rays of a first polarization direction 01 and a second light-emitting hole b capable of emitting light rays of a second polarization direction 02.

[0034] In the present application, the focal length adjustment of the zoom collimator lens group 2 is linked with the lighting of a specific light-emitting area in the light source module 1, so that the structured light pattern projected by the zoom polarized light projector A at different depths of field can maintain consistent field of view FOV and speckle point density.

[0035] For example, the first light-emitting area 11 corresponds to the first focal length f1 of the zoom collimator lens group 2, and the second light-emitting area 12 corresponds to the second focal length f2 of the zoom collimator lens group 2.

[0036] In the present application, the zoom collimator lens group 2 is used to collimate the light rays emitted from the light source module 1 and change the projection depth of field of the zoom polarized light projector A by adjusting its focal length. Specifically, collimation refers to the process of converting the light rays emitted by the light source module 1 into parallel light. This step is crucial because only parallel, uniform light can form a structured light pattern with regular shape and stable quality when incident on the subsequent diffractive optical element 3. Another function of the zoom collimator lens group 2 is to change the projection depth of field. Depth of field refers to the range of object distances that can be clearly imaged. By changing the focal length of the zoom collimator lens group 2, the convergence or divergence characteristics of the light rays can be controlled, thereby changing the distance in space where the projected pattern is clearest and has the most concentrated energy (i.e., the depth of field).

[0037] In the present application, the focal length of the zoom collimator lens group 2 is inversely proportional to the projection depth of field: the focal length increases, the depth of field decreases; the focal length decreases, the depth of field increases.

[0038] The diffractive optical element 3 is fixedly arranged on the light-emitting side of the zoom collimator lens group 2 and is used to diffract the collimated light rays to form a structured light pattern. The diffractive optical element 3 is a thin sheet with a micro-nano structure on its surface. When parallel collimated light passes through it, diffraction occurs, recombining a parallel light beam into a structured light pattern with specific characteristic points (such as speckles).

[0039] One design in the present application is to fix the diffractive optical element 3 after the zoom collimator lens group 2, which is an important design and is significantly different from the prior art, such as zooming after the diffractive optical element. Specifically: In the present application, the diffractive optical element 3 is fixed to work in the parallel light provided by the zoom collimator group 2 as expected in its design. This ensures that the basic form and optical quality of the generated structured light pattern are stable and consistent regardless of the change in focal length. This "zoom collimation first, then fixed diffraction" architecture is the fundamental guarantee to realize the core innovation of the present application - a single light projector that can project a high-quality pattern with consistent parameters at different depths.

[0040] In an example of the present application, the light source module 1 can adopt a VSEL light source.

[0041] The three-dimensional sensing system provided by the embodiments of the present application has a zoom polarized light projector A, which includes a light source module 1, a zoom collimator group 2, and a diffractive optical element 3. The working process of the three-dimensional sensing system is as follows: first, select the working mode according to the requirements, such as a small depth of field mode or a large depth of field mode, then adjust the focal length of the zoom collimator group 2 and control the lighting of the corresponding light-emitting area. The light emitted by the light source module 1 (including light with two different polarization directions) is collimated and zoomed by the zoom collimator group 2, and then exits in the form of a collimated light beam to the diffractive optical element 3. After being diffracted by the diffractive optical element 3, a structured light pattern with a specific spatial distribution is formed and projected onto the surface of a target object.

[0042] For example, in the large depth of field mode, the focal length of the zoom collimator group 2 can be adjusted to a first focal length f1, and the first light-emitting area 11 is controlled to be lit; in the small depth of field mode, the focal length of the zoom collimator group 2 can be adjusted to a second focal length f2, and the second light-emitting area 12 is controlled to be lit; wherein the first focal length f1 is smaller than the second focal length f2, and the light-emitting size of the first light-emitting area 11 is smaller than the light-emitting size of the second light-emitting area 12.

[0043] The three-dimensional sensing system provided by the embodiments of the present application realizes structured light projection at different depths by using a zoom collimator group 2 combined with a single diffractive optical element 3, and matching a light source module 1 with independent control in different areas. The core advantage of this optical design is that it can output a structured light pattern with consistent field of view angle FOV and speckle point density at different focal lengths, which fundamentally guarantees the consistency of the pattern required for three-dimensional imaging, significantly improves the accuracy of depth calculation and the adaptability of the system. At the same time, this integrated design effectively simplifies the system structure, reduces the hardware cost and calibration complexity, and provides an efficient and reliable sensing solution for various devices that require three-dimensional environmental perception.

[0044] The three-dimensional sensing system provided by the embodiments of the present application can be adapted to many technical fields that have an urgent need for three-dimensional environmental perception, such as floor cleaning robots, service robots, autonomous driving, AR / VR, intelligent monitoring, etc., and has a very broad application prospect.

[0045] Referring to Figure 3 and Figure 4 , the light source module 1 of the present application comprises a first light emitting area 11 and a second light emitting area 12, both of which are mixed and matched with first light emitting holes a emitting light rays of a first polarization direction 01 and second light emitting holes b emitting light rays of a second polarization direction 02, and the first polarization direction and the second polarization direction are orthogonal to each other. This design enables each light emitting area of the light source module 1 to output two orthogonal polarized lights at the same time, constituting a complete dual-polarized light source unit. For example, the first polarization direction is 90°, and the second polarization direction is 0°. Or vice versa.

[0046] The mixed and matched polarization structure of the light source module 1 is the key to realizing "projected light with dual polarization information". By independently controlling the lighting and extinguishing of the first light emitting area 11 or the second light emitting area 12, the present application can flexibly realize the switching between different depth of field modes. At the same time, since the projected light contains orthogonal polarization components, the receiver B (i.e. the receiving end of the three-dimensional sensing system) on the subsequent light path can respectively collect images of different polarization states by switching the angle of the polarizer in time, and then effectively suppress mirror reflection noise and extract diffuse reflection signals through image fusion, thereby significantly improving the three-dimensional perception accuracy in high-reflectivity scenes. The details will be described in the receiver B part, which will not be described here.

[0047] In some examples of the present application, referring to Figure 3 and Figure 4 , the first light emitting area 11 and the second light emitting area 12 are arranged in a stacked arrangement or a side-by-side arrangement.

[0048] In this example provided by the present application, referring to Figure 3 and Figure 4 , the first light emitting area 11 and the second light emitting area 12 can adopt two different arrangement modes.

[0049] In the stacked arrangement mode, referring to Figure 3 , the first light emitting area 11 and the second light emitting area 12 are arranged in a stacked manner in a direction perpendicular to the optical axis. As can be seen from Figure 3 , the light emitting holes of the first light emitting area 11 and the light emitting holes of the second light emitting area 12 partially overlap. This layout can reduce the overall area of the light source module 1, which is conducive to the miniaturization and integration of the system, and is especially suitable for compact devices such as floor cleaning robots that have strict volume restrictions.

[0050] In addition, it should be noted that, referring to Figure 3In the overlapping part of the first light-emitting area 11 and the second light-emitting area 12, the first light-emitting hole a of the first light-emitting area 11 corresponds to the position of the first light-emitting hole a of the second light-emitting area 12, and the second light-emitting hole b of the first light-emitting area 11 corresponds to the position of the second light-emitting hole b of the second light-emitting area 12.

[0051] In the parallel arrangement mode, referring to Figure 4 , the first light-emitting area 11 and the second light-emitting area 12 are arranged as spatially separated independent blocks. This mode avoids the overlap of the physical positions between the light-emitting areas, making the layout design, electrode lead and driving control more convenient, and is beneficial to reduce the process complexity and manufacturing cost.

[0052] In some examples of the present application, the zoom collimator lens group 2 is one of a motor-driven lens group, a liquid lens or a liquid crystal lens.

[0053] In this example provided by the present application, the zoom collimator lens group 2 can be implemented by any one of a motor-driven lens group, a liquid lens or a liquid crystal lens. The zoom collimator lens group 2 is located between the light source module 1 and the diffractive optical element 3, and one of its functions is to realize the switching of different projection depth of field by adjusting its focal length. For example, switching to large focal length and small depth of field (first depth of field) or small focal length and large depth of field (second depth of field), as shown in Figure 1 and Figure 2 .

[0054] Motor-driven lens group: the focal length is adjusted by changing the lens spacing through mechanical structure. The advantage of this mode is that the technology is mature and the control is accurate, which is a reliable mechanical solution to realize zoom.

[0055] Liquid lens: the focal length is changed quickly and silently by applying an electric field to change the surface curvature of the liquid lens, which has the advantages of fast response speed and no mechanical moving parts.

[0056] Liquid crystal lens: the focal length is electronically adjusted by changing the effective refractive index distribution of liquid crystal molecules through external electric field regulation, which has the characteristics of compact structure and low power consumption.

[0057] Although the above three ways have different physical mechanisms, they can all realize dynamic adjustment of focal length and ensure good spot quality during zooming. They respectively represent three technical paths of mechanical, fluid optics and electrically controlled optics, providing diversified choices for different volume, cost and response speed requirements of application scenarios.

[0058] It should be emphasized that no matter which way is adopted, they are located between the light source module 1 and the diffractive optical element 3, which is the key to ensure that the spot quality is not degraded.

[0059] In some examples of this application, the minimum transverse period Ph and the minimum vertical period Pv of the diffractive optical element 3 satisfy the following relationship: Ph=λ / 2sin(artan(L1 / (2*f1)))(1); Pv=λ / 2sin(artan(H1 / (2*f1)))(2); Wherein, λ is the wavelength of the light emitted from the light source module 1, L1 and H1 are the length and width of the first light-emitting area 11, respectively, and f1 is the first focal length.

[0060] In one design example of this application, the minimum lateral period Ph and minimum vertical period Pv of the diffractive optical element 3 are determined by the above-mentioned relationships (1) and (2). Wherein, λ is the emission wavelength of the light source module 1, L1 and H1 are the length and width of the first emission area 11, respectively, and f1 is the first focal length set by the zoom collimating lens group 2, suitable for, for example, a close-range projection scenario of 30m. The core of this design principle lies in associating the emission size of the first emission area 11 with the focal length of the zoom collimating lens group 2 to the structural parameters of the diffractive optical element 3, as explained below.

[0061] The expressions (L1 / (2*f1) and (H1 / (2*f1)) represent the tangent values ​​of the angles subtended by half the length and width of the first light-emitting area 11 with respect to the first focal length f1, respectively. By combining the arctangent and the sine function, the maximum diffraction angle θ allowed after collimation can be derived. Finally, the critical period λ / (2sinθ) is derived according to the grating equation. This period value is the diffraction limit corresponding to achieving the target field of view (FOV).

[0062] Through the above design, when the first light-emitting area 11 is lit and the zoom collimating lens group 2 is at the first focal length f1, the generated first structured light pattern can completely cover the preset FOV, while achieving uniform speckle distribution and high light energy utilization, thus providing a high-quality optical foundation for close-range three-dimensional perception.

[0063] In some examples of this application, the length L2 and width H2 of the second light-emitting region 12 satisfy the following relationship: L2=2*f2*tan(arsin(λ / 2*Ph))(3); H2=2*f2*tan(arsin(λ / 2*Pv))(4); Where f2 is the second focal length, and satisfies f1 < f2 and H1*L1 < H2*L2.

[0064] In this example of the application, to ensure that when the zoom collimating lens group 2 is switched to the second focal length f2 (e.g., for long-distance projection at 100m), the zoom polarized light projector A can still project a second structured light pattern with the same field of view (FOV) and speckle density as at the first focal length f1. This example controls the luminous size of the second luminous region 12, see equations (3) and (4).

[0065] The premise of this design is the use of a single, fixed diffractive optical element 3 (DOE). Once its microstructure period (Ph, Pv) is determined, the maximum diffraction angle that can be produced is fixed, thus determining the limit FOV that the system can cover.

[0066] The expressions (λ / 2*Ph) and (λ / 2*Pv) represent the sine value of the diffraction angle determined by the period of the diffraction optical element 3; the diffraction angle θ is obtained by the arcsine function; the tangent value tan(θ) and the focal length f2 together determine the length dimension L2 and width dimension H2 of the second light-emitting region 12 required to achieve the diffraction angle.

[0067] The essence of equations (3) and (4) is to perform reverse design under the constraint of fixed diffractive optical element 3. That is, based on the known period (Ph, Pv) of diffractive optical element 3 and the second focal length f2, the necessary theoretical size of the second light-emitting region 12 is calculated in reverse. This ensures that when a larger focal length is used, by illuminating the larger second light-emitting region 12, the light emitted by it, after being diffracted by diffractive optical element 3, can still fill the same FOV as the near-focal length mode and maintain a consistent speckle density.

[0068] In other words, in this example of the application, the period (Ph, Pv) and wavelength (λ) of the diffractive optical element 3 are set to known constants, and the size of the light source (i.e. the second light-emitting region 12) required to achieve the maximum diffraction angle that the diffractive optical element 3 can produce is deduced.

[0069] In this application, the diffractive optical element 3 is fixed, and its periods Ph and Pv can be determined by the emission size of the first emission region 11 and the first focal length f1. Alternatively, the periods Ph and Pv of the diffractive optical element 3 can also be designed to be determined by the emission size of the second emission region 12 and the second focal length f2. In this case, the emission size of the first emission region 12 needs to be further determined based on the determined Ph and Pv. The specific design can be flexibly adapted as needed.

[0070] In this example of the application, f1 < f2 and H1*L1 < H2*L2 are proposed. This clarifies the working mode: the first luminous area 11 (small size) is used with a small focal length (first focal length f1) for a large depth of field; the second luminous area 12 (large size) is used with a large focal length f2 (second focal length f2) for a small depth of field.

[0071] See Figure 5 One of the simulation results shown is Figure 5 The left and right images show the structured light patterns projected by the diffractive optical element 3 at projection distances of 100m and 30m, using different focal lengths. The left image shows the first depth-of-field pattern (corresponding to a long distance of 100m and a large focal length), and the right image shows the second depth-of-field pattern (corresponding to a short distance of 30m and a small focal length). Comparing the two images, it can be seen that despite the different projection distances, the generated patterns maintain a consistent field of view (FOV).

[0072] See Figure 6 The second simulation result shown is... Figure 6 The two images on the left and right correspond to respectively Figure 5 The local light spot distribution of 10m×10m extracted from the 100m and 30m projection patterns is shown. It can be seen that the light spot distribution density of the two depth-of-field patterns is basically the same at different projection distances, indicating that the three-dimensional sensing system can maintain a uniform and matched light spot density in different working modes.

[0073] See Figure 5 and Figure 6 The simulation results shown verify that the zoom polarized light projector A of this application can achieve consistent FOV and speckle density at different depths of field.

[0074] See some examples in this application. Figure 7 and Figure 8 The three-dimensional sensing system further includes a receiver B and an analyzer 4. The receiver B receives the structured light pattern emitted from the zoom polarized light projector A and reflected back from the target object. The analyzer 4 is disposed in the incident light path of the receiver B and is used to select light rays with a specific polarization direction.

[0075] In this example of the application, the three-dimensional sensing system also integrates a receiving part adapted to the projection part, which includes a receiver B and an analyzer 4.

[0076] The receiver B's main function is to capture structured light patterns reflected from the surface of the target object, such as a first structured light pattern or a second structured light pattern.

[0077] The analyzer 4 is located at the incident end of the receiver B and functions as an optical filter, selectively transmitting light with a specific polarization direction. This design addresses the potential limitations of highly reflective surfaces (such as glass, mirrors, and smooth metals) in three-dimensional sensing. A key solution to the problem of failure. Its working principle and function are as follows: (1) Suppressing specular reflection interference: Highly reflective surfaces produce strong specular reflection light with a polarization direction consistent with the incident light. This strong light can easily overexpose the photosensitive chip 7 of the receiver B, obscuring the diffuse reflection details of the object's surface. By setting the transmission axis of the analyzer 4 to be orthogonal to the main polarization direction of the projected light, this part of the specular reflection light can be effectively blocked, thereby significantly reducing the risk of overexposure and preserving the precious diffuse reflection signal.

[0078] (2) Improved signal-to-noise ratio: In complex optical environments, specular reflection is considered "noise". The analyzer 4 suppresses this noise directly from the optical level through polarization filtering, making the useful diffuse reflection signal more prominent, thereby significantly improving the system signal-to-noise ratio.

[0079] (3) Collaborative Mechanism: The combination of receiver B and analyzer 4 is not a simple superposition, but rather constitutes a powerful optical detection front end. Analyzer 4 first preprocesses the incident light to filter out harmful polarization noise; subsequently, receiver B is responsible for converting the filtered, high-quality optical signal into an electrical signal (image). This collaborative working mode, especially with the cooperation of the zoom polarization projector A emitting orthogonally polarized light, provides a fundamental guarantee for achieving stable and reliable 3D perception in highly challenging scenarios (such as home environments with a large amount of glass and metal).

[0080] In some examples of this application, the analyzer 4 is a rotatable linear polarizer or a liquid crystal polarization modulation device.

[0081] In the example provided in this application, the analyzer 4 can be implemented using a rotatable linear polarizer or a voltage-controlled liquid crystal polarization modulation device (such as a liquid crystal film). Both of these solutions can achieve polarization direction switching, providing different hardware options for the system to suppress specular reflection.

[0082] A rotatable linear polarizer is a mechanical implementation that uses a micro-motor to drive the polarizer to physically rotate, thereby continuously changing the orientation angle of the transmission axis. The advantages of this approach are its mature structure, relatively low cost, and suitability for applications where switching speed requirements are not high.

[0083] Voltage-controlled liquid crystal polarization modulation devices (such as liquid crystal films) are an electronic solution that utilizes the electrically controlled birefringence properties of liquid crystal materials. By changing the applied voltage, the alignment direction of liquid crystal molecules is adjusted, thereby achieving rapid switching of the transmission polarization direction. The advantages of this solution are that it has no moving parts, fast response speed, and high reliability, making it suitable for high-frequency acquisition or systems with requirements for size and integration.

[0084] The two implementation methods mentioned above each have their own focus in terms of cost and performance, providing a flexible and reliable technical choice for 3D sensing systems under different application requirements.

[0085] This example demonstrates the feasibility of achieving switchable polarization by showcasing two techniques. It highlights design flexibility: choosing a rotatable linear polarizer prioritizes cost-effectiveness and high performance, while selecting a liquid crystal polarization modulator (such as a liquid crystal film) prioritizes speed, reliability, and integration (no moving parts, rapid switching). Both approaches provide robust and optional hardware support for achieving anti-specular reflection functionality in the entire 3D sensing system.

[0086] In some examples of this application, the receiver B is configured to: capture the structured light pattern reflected back by the target object in frames when the zoom collimating lens group 2 is at the first focal length f1 or the second focal length f2; The three-dimensional sensing system also includes a processor connected to the receiver B and configured to perform the following operations: The analyzer 4 is controlled to have its transmission axis parallel to the first polarization direction during the first frame acquisition, so that the receiver B can acquire the first polarization direction light 01 and form a first image. The analyzer 4 is controlled to have its transmission axis parallel to the second polarization direction during the second frame acquisition, so that the receiver B can acquire the second polarization direction light 02 and form a second image. The first image and the second image are fused together, and depth calculation is performed based on the fused image to reconstruct the three-dimensional information of the target object.

[0087] In a specific example, see Figure 7 and Figure 8 The receiver B has a rotatable linear polarizer as an analyzer 4 at its incident end. This linear polarizer can be designed to be driven by a micro-motor to achieve rapid switching of the transmission axis direction. Its operation is as follows: First, control the micro motor to rotate the analyzer 4 (such as a linear polarizer) to the 0° position, see [link to relevant documentation]. Figure 7At this time, the transmission axis of the analyzer 4 is parallel to the first polarization direction (such as 90° linearly polarized light), allowing light in this direction to pass through efficiently. The photosensitive chip 7 in the receiver B then acquires a frame image, which mainly records the light intensity information in the 90° polarization direction of the scene, forming the first image.

[0088] Subsequently, the micro motor was immediately controlled to switch the analyzer 4 to a 90° orientation, see [link / reference]. Figure 8 At this point, the transmission axis of the analyzer 4 rotates to be parallel to the second polarization direction (e.g., 0° linearly polarized light). The receiver B acquires the next frame image, which mainly records the light intensity information in the 0° polarization direction, forming the second image.

[0089] Finally, the processor can perform fusion calculations on the two image frames. Since specular reflection light is significantly suppressed in each frame because its polarization direction is orthogonal to the analyzer's transmission axis, the effective diffuse reflection signal carrying the object's three-dimensional shape information is complementaryly preserved in both frames. Through image fusion, the effective signal is enhanced, and the overall signal-to-noise ratio and depth calculation accuracy in challenging scenarios (such as highly reflective surfaces) are significantly improved.

[0090] In another specific example, see Figure 9 The receiver B has a liquid crystal polarization modulation device (such as a liquid crystal film) as an analyzer 4 at its incident end. This liquid crystal polarization modulation device utilizes the electrically controlled anisotropy of liquid crystal molecules, applying different modes of driving voltage to control the molecular alignment, thereby achieving selective transmission of light with a specific polarization direction. Its specific working process is as follows: First, a first driving voltage (e.g., a low-frequency voltage) is applied to the liquid crystal polarization modulation device, causing the liquid crystal molecules inside to align in a first preset direction (e.g., horizontal alignment). In this state, the liquid crystal film selectively transmits light in the first polarization direction (e.g., 90° linearly polarized light), and the photosensitive chip 7 in the receiver B then captures a frame image to form the first image.

[0091] Subsequently, the driving voltage is switched to a second driving voltage (e.g., a high-frequency voltage), causing the alignment direction of the liquid crystal molecules to change to a second preset direction orthogonal to the first direction (e.g., vertical alignment). At this time, the liquid crystal polarization modulation device selectively transmits light rays with a second polarization direction orthogonal to the first polarization direction (e.g., 0° linearly polarized light), and the photosensitive chip 7 acquires the next frame image to form the second image.

[0092] Finally, the processor can perform fusion processing on the first and second images. This scheme switches the polarization detection state electronically, without any moving mechanical parts, and has significant advantages such as fast response speed, long service life, good vibration resistance, and easier system integration.

[0093] The technical solution described in this application, through processor-controlled analyzer 4 for frame-by-frame polarization acquisition and subsequent image fusion processing, cleverly solves the industry problem of strong reflection interference from both optical and algorithmic perspectives, thereby achieving high-precision three-dimensional sensing. By fusing two frames of images, the intensity of the effective diffuse reflection signal is significantly improved, while specular reflection is suppressed in each frame, which can improve the signal-to-noise ratio and measurement accuracy in low-reflectivity areas.

[0094] See some examples in this application. Figure 7 and Figure 8 The receiver B also includes a lens group 5, a narrowband filter 6, and a photosensitive chip 7 arranged sequentially along the optical path.

[0095] The receiver B includes an analyzer 4, a lens group 5, a narrowband filter 6, and a photosensitive chip 7 arranged sequentially along the optical path. See below. Figure 7 and Figure 8 Each component has a clearly defined function and works in sequence.

[0096] Analyzer 4: Located at the front end of the optical path, it is used to switch the polarization direction of transmission according to control commands, selectively suppress specular reflection light in frames and transmit useful diffuse reflection signals, and is a key component for realizing anti-interference function.

[0097] Lens group 5: Located after the analyzer 4, it is used to focus the light passing through the analyzer 4 and clearly image it onto the photosensitive chip 7. Its imaging quality directly affects the clarity and accuracy of the captured pattern.

[0098] Narrowband filter 6: disposed between the lens group 5 and the photosensitive chip 7, used to transmit only light with the same wavelength as the projected light source (such as 940nm infrared light), effectively suppressing ambient light interference and improving the signal-to-noise ratio.

[0099] Photosensitive chip 7: As the core photoelectric conversion element, it converts the received light signal into a digital image signal and outputs it to the processing module. Its resolution, sensitivity and dynamic range directly affect the image details and system performance.

[0100] The receiver B, through the orderly coordination of the above components in the optical path, achieves high-precision, high-signal-to-noise ratio acquisition of structured light patterns with specific polarization states and wavelengths, providing reliable image input for subsequent depth calculations.

[0101] According to another embodiment of this application, a three-dimensional sensing method is provided, the three-dimensional sensing method comprising the following steps S1 to S3: Step S1: Projection Step: The zoom polarized light projector A is controlled to project a structured light pattern onto the target object by adjusting the focal length of its zoom collimating lens group 2 to the target focal length and illuminating the light-emitting area corresponding to the target focal length; wherein the target focal length is a first focal length f1 or a second focal length f2, and the structured light pattern includes two types of light rays with a first polarization direction and a second polarization direction that are orthogonal to each other. Step S2: Data collection steps: The receiver B, in conjunction with the analyzer 4 located on its incident light path and the processor connected thereto, performs frame-by-frame acquisition of the structured light pattern reflected from the target object, including: The processor controls the analyzer 4 so that its transmission axis is parallel to the first polarization direction during the first frame acquisition, so that the receiver B acquires the first polarization direction light ray O1 to form a first image; and The processor controls the analyzer 4 so that its transmission axis is parallel to the second polarization direction during the second frame acquisition, so that the receiver B can acquire the second polarization direction light 02 and form a second image; Step S3: Processing steps: The processor performs a fusion process on the first image and the second image, and performs depth calculations based on the fused image to reconstruct the three-dimensional information of the target object.

[0102] The three-dimensional sensing method provided in this application achieves multi-depth, high-precision three-dimensional detection in a single light projector system by coordinating and controlling the zoom operation of the projection end, the switching of the polarization light source, and the polarization filtering of the receiving end, and significantly improves the ability to resist specular reflection interference.

[0103] In step S1, the zoom polarized light projector A is controlled to perform two operations: Adjusting the focal length: Adjust the focal length of the zoom collimating lens group 2 to the desired target focal length (e.g., f1 or f2). And, Illuminate the corresponding light-emitting area: Simultaneously illuminate the light-emitting area (first light-emitting area 11 or second light-emitting area 12) that is pre-matched to the focal length of the target.

[0104] Through the aforementioned coordinated control, the zoom polarization projector A projects a structured light pattern containing orthogonal polarization information onto the target object.

[0105] Step S2 is the information capture and preprocessing stage, which can solve the problem of specular reflection interference.

[0106] The core of step S2 is that the processor controls the analyzer 4 to perform frame-by-frame polarization acquisition: Acquiring the first image: Control the analyzer 4 so that its transmission axis is parallel to the first polarization direction. At this time, receiver B mainly receives and records the light ray 01 in the first polarization direction, forming the first image. In this frame, specular reflection light in the second polarization direction is effectively suppressed.

[0107] Acquiring the second image: The analyzer 4 is controlled to switch its transmission axis to be parallel to the second polarization direction. Receiver B mainly receives and records the light ray O2 in the second polarization direction, forming the second image. This frame suppresses specular reflection light in the first polarization direction.

[0108] This time-division polarization filtering separates two types of light rays with orthogonal polarization directions mixed in the reflected light into two independent images. This significantly weakens the strong specular reflection light caused by highly reflective surfaces that maintains its original polarization direction in any frame, while the diffuse reflection light carrying information about the object's three-dimensional shape and whose polarization state is randomized is preserved in both frames.

[0109] Step S3 is image fusion: the processor performs fusion processing on the acquired first and second images. The purpose of image fusion is complementarity and enhancement. It superimposes and enhances the useful diffuse reflection signals in the two frames, while suppressing specular reflection noise in each frame, resulting in a significant improvement in the overall signal-to-noise ratio of the fused image.

[0110] Ultimately, based on this clear image that eliminates most of the interference from specular reflections, depth calculation algorithms (such as triangulation) can accurately and reliably reconstruct the three-dimensional information of the target object, maintaining high accuracy even in challenging scenarios with highly reflective objects such as glass and metal.

[0111] According to yet another embodiment of this application, a robotic vacuum cleaner is provided, which includes the three-dimensional sensing system described above.

[0112] The specific implementation methods of the three-dimensional sensing method and the sweeping robot in this application can refer to the various embodiments of the three-dimensional sensing system described above. Therefore, they have at least all the beneficial effects brought about by the technical solutions of the above embodiments, and will not be described in detail here.

[0113] The above embodiments mainly describe the differences between the various embodiments. As long as the different optimization features between the various embodiments are not contradictory, they can be combined to form a better embodiment. For the sake of brevity, they will not be elaborated here.

[0114] While specific embodiments of this application have been described in detail by way of examples, those skilled in the art should understand that the above examples are for illustrative purposes only and are not intended to limit the scope of this application. Those skilled in the art should understand that modifications can be made to the above embodiments without departing from the scope and spirit of this application. The scope of this application is defined by the appended claims.

Claims

1. A three-dimensional sensing system, characterized in that, Including a zoom polarized light projector (A); The zoom polarized light projector (A) includes: The light source module (1) includes a first light-emitting area (11) and a second light-emitting area (12) that can be controlled independently. Each light-emitting area includes a first light-emitting aperture (a) that emits light rays (01) with a first polarization direction and a second light-emitting aperture (b) that emits light rays (02) with a second polarization direction, wherein the first polarization direction and the second polarization direction are orthogonal to each other. The zoom collimating lens group (2) is set on the light-emitting side of the light source module (1) to collimate the light emitted from the light source module (1) and change the projection depth of field by adjusting the focal length; A single diffractive optical element (3) is disposed on the light-emitting side of the zoom collimating lens group (2) to diffract the collimated beam into a structured light pattern. The zoom polarized light projector (A) is configured as follows: When the first depth of field projection is performed, the zoom collimating lens group (2) is adjusted to the first focal length f1 and the first light-emitting area (11) is lit to project the first structured light pattern. When the second depth of field projection is performed, the zoom collimating lens group (2) is adjusted to the second focal length f2 and the second light-emitting area (12) is lit to project the second structured light pattern; wherein the first structured light pattern and the second structured light pattern have the same field of view and speckle density.

2. The three-dimensional sensing system according to claim 1, characterized in that, The first light-emitting area (11) and the second light-emitting area (12) are arranged in a superimposed or parallel arrangement in space.

3. The three-dimensional sensing system according to claim 1, characterized in that, The zoom collimating lens group (2) is one of a motor-driven lens group, a liquid lens, or a liquid crystal lens.

4. The three-dimensional sensing system according to claim 1, characterized in that, The minimum transverse period Ph and minimum vertical period Pv of the diffractive optical element (3) satisfy the following relationship: Ph=λ / 2sin(artan(L1 / (2*f1))); Pv=λ / 2sin(artan(H1 / (2*f1))); Wherein, λ is the wavelength of the light emitted from the light source module (1), L1 and H1 are the length and width of the first light-emitting area (11) respectively, and f1 is the first focal length.

5. The three-dimensional sensing system according to claim 4, characterized in that, The length L2 and width H2 of the second light-emitting area (12) satisfy the following relationship: L2=2*f2*tan(arsin(λ / 2*Ph)); H2=2*f2*tan(arsin(λ / 2*Pv)); Where f2 is the second focal length, and satisfies f1 < f2 and H1*L1 < H2*L2.

6. The three-dimensional sensing system according to any one of claims 1-5, characterized in that, The three-dimensional sensing system also includes: Receiver (B) for receiving a structured light pattern emitted from the zoom polarized light projector (A) and reflected back by the target object; and, The analyzer (4) is disposed on the incident light path of the receiver (B) and is used to select light rays that pass through a specific polarization direction.

7. The three-dimensional sensing system according to claim 6, characterized in that, The analyzer (4) is a rotatable linear polarizer or a liquid crystal polarization modulation device.

8. The three-dimensional sensing system according to claim 6, characterized in that, The receiver (B) is configured to: when the zoom collimating lens group (2) is at the first focal length f1 or the second focal length f2, to perform frame-by-frame acquisition of the structured light pattern reflected back by the target object; The three-dimensional sensing system also includes a processor connected to the receiver (B) and configured to perform the following operations: The analyzer (4) is controlled so that its transmission axis is parallel to the first polarization direction during the first frame acquisition, so that the receiver (B) can acquire the light rays (01) of the first polarization direction and form a first image; The analyzer (4) is controlled so that its transmission axis is parallel to the second polarization direction during the second frame acquisition, so that the receiver (B) can acquire the light rays (02) of the second polarization direction and form a second image; The first image and the second image are fused together, and depth calculation is performed based on the fused image to reconstruct the three-dimensional information of the target object.

9. The three-dimensional sensing system according to claim 8, characterized in that, The receiver (B) also includes a lens group (5), a narrowband filter (6) and a photosensitive chip (7) arranged sequentially along the optical path.

10. A three-dimensional sensing method, characterized in that, include: Control the zoom polarized light projector (A) to project a structured light pattern onto the target object by adjusting the focal length of its zoom collimating lens group (2) to the target focal length and illuminating the light-emitting area corresponding to the target focal length; wherein the target focal length is a first focal length f1 or a second focal length f2, and the structured light pattern includes two types of light rays with a first polarization direction and a second polarization direction that are orthogonal to each other. The structured light pattern reflected from the target object is captured in frames using a receiver (B) and a polarizer (4) located on its incident light path, along with a processor connected thereto. This includes: The processor controls the analyzer (4) so ​​that its transmission axis is parallel to the first polarization direction during the first frame acquisition, so that the receiver (B) acquires the light rays (01) in the first polarization direction to form a first image; and The processor controls the analyzer (4) so ​​that its transmission axis is parallel to the second polarization direction when the second frame is acquired, so that the receiver (B) can acquire the light rays (02) of the second polarization direction and form a second image; The processor performs a fusion process on the first image and the second image, and performs depth calculations based on the fused image to reconstruct the three-dimensional information of the target object.

11. A robotic vacuum cleaner, characterized in that, include: The three-dimensional sensing system according to any one of claims 1-9.

Images