Sampling method and system of optical sensor, electronic equipment and storage medium

Abstract

The invention provides a sampling method and system of an optical sensor, electronic equipment and a storage medium. The sampling method comprises the following steps: determining an observation vector set of all sampling points in an initial sampling point set relative to a known observation object; traversing the observation vector set based on each aperture of the optical sensor, determining all sampling points capable of covering the aperture, determining an observer subset corresponding to the aperture according to all the sampling points, and determining a total observer set according to the observer subsets corresponding to all the apertures; determining a coverage matrix about the aperture and the total observer set; determining a coverage contribution value of the sampling point to the aperture based on the coverage matrix; determining a target sampling point according to the coverage contribution value so as to update the target sampling point to a target sampling point set; and updating the coverage matrix, performing multiple iterations until a preset condition is met, and updating the target sampling points determined by each iteration to the target sampling point set to obtain a final target sampling point set.

Description

Technical Field

[0001] The present invention relates to the technical field of optical sensor sampling, and in particular to a sampling method and system for an optical sensor, an electronic device, and a storage medium. Background Art

[0002] Optical sensors (such as stellar sensors, solar sensors, polarization sensors, and optical flow sensors) are core sensors for high-precision attitude navigation and control in aerospace vehicles and unmanned systems. Optical sensors capture ambient optical information (such as stellar, solar, or polarization signals) and combine it with algorithms to calculate the vehicle's attitude. Their accuracy directly depends on the accuracy of the sensor's raw measurements. However, measurement errors primarily originate from the sensor's optical and imaging systems, impacting measurement accuracy.

[0003] Currently, optical sensors are typically used for multi-angle sampling based on high-precision turntables. For example, a high-precision turntable can rotate along multiple axes to point the optical sensor to a specific spatial orientation, thereby collecting relevant optical information about a known object.

[0004] However, the inventors of the present invention have found that the current sampling method of optical sensors based on high-precision turntables still has problems such as nonlinear imaging, the number of samples is limited by the field of view, and the sampling time is limited by the dynamic performance of the turntable. As a result, the current conventional sampling method has the problem of low sampling efficiency.

[0005] The contents of the background technology section are merely technologies known to the public and do not necessarily represent the existing technologies in this field. Summary of the Invention

[0006] According to one aspect of the present invention, a sampling method for an optical sensor is provided, comprising: determining an observation vector set for all sampling points in an initial sampling point set relative to a known observation object; traversing the observation vector set based on each aperture of the optical sensor to determine all sampling points that can cover the aperture, thereby determining a subset of observers corresponding to the aperture based on all sampling points, and determining a total set of observers based on the subsets of observers corresponding to all apertures; determining a covering matrix for the aperture and the total set of observers; determining a coverage contribution value of the sampling point to the aperture based on the covering matrix; determining a target sampling point based on the coverage contribution value, thereby updating the target sampling point to the target sampling point set; updating the covering matrix, iterating multiple times until a preset condition is satisfied, and updating the target sampling point determined in each iteration to the target sampling point set to obtain a final target sampling point set.

[0007] According to another aspect of the present invention, a sampling system for an optical sensor is provided. The sampling system includes a sampling point set processing module and a target sampling point determination module. The sampling point set processing module determines an initial sampling point set based on an initial step size, determines a set of observation vectors for all sampling points in the initial sampling point set relative to a known observation object, and traverses the set of observation vectors based on each aperture of the optical sensor to determine all sampling points that can cover the aperture. The module then determines a subset of observers corresponding to the aperture based on all sampling points, and a total set of observers based on the subsets of observers corresponding to all apertures. The target sampling point determination module determines a covering matrix for the aperture and the total set of observers, determines a coverage contribution value of the sampling point to the aperture based on the covering matrix, determines a target sampling point based on the coverage contribution value, updates the target sampling point to the target sampling point set, and iterates the covering matrix multiple times until a preset condition is satisfied. The target sampling point determined in each iteration is updated to the target sampling point set to obtain a final target sampling point set.

[0008] According to yet another aspect of the present invention, an electronic device is provided. The electronic device includes: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, enables the one or more processors to implement the method described above.

[0009] According to another aspect of the present invention, a non-volatile computer-readable storage medium is provided, wherein a computer program is stored on the storage medium, and when the computer program is executed by a processor, the method described above can be implemented.

[0010] According to another aspect of the present invention, a computer program product is provided, which includes a computer program stored on a computer-readable storage medium; the computer program includes program instructions, which, when executed by a computer, cause the computer to execute the method described above.

[0011] Beneficial effects

[0012] The present invention can select a minimum subset from the initial sampling point set to determine that each aperture of the optical sensor can be covered by sufficient sampling points based on the target azimuth angle, thereby maximizing the coverage of the observation. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0014] Figure 1 A schematic diagram showing a flow chart of a sampling method according to an embodiment of the present invention;

[0015] Figure 2 Another schematic flow chart showing the sampling method according to an embodiment of the present invention;

[0016] Figure 3 Another schematic flow chart showing the sampling method according to an embodiment of the present invention;

[0017] Figure 4 A schematic diagram showing the distribution of all sampling points in the total set of observers according to an embodiment of the present invention is shown;

[0018] Figure 5 A schematic diagram showing the distribution of all sampling points in the target sampling point set according to an embodiment of the present invention is shown;

[0019] Figure 6 A schematic diagram showing a sampling sequence according to an embodiment of the present invention;

[0020] Figure 7 A schematic structural diagram of a sampling system according to an embodiment of the present invention is shown.

[0021] Description of reference numerals:

[0022] Sampling system 1; sampling point set processing module 10; target sampling point determination module 20. DETAILED DESCRIPTION

[0023] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be embodied in many forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete and will fully convey the concepts of the example embodiments to those skilled in the art. Like reference numerals in the drawings represent like or similar parts, and thus repeated description thereof will be omitted.

[0024] The described features, structures or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, many specific details are provided to provide a full understanding of the embodiments of the present disclosure. However, those skilled in the art will appreciate that the technical solutions of the present disclosure may be practiced without one or more of these specific details, or other methods, components, materials, devices, etc. may be employed. In these cases, well-known structures, methods, devices, implementations, materials or operations will not be shown or described in detail.

[0025] Furthermore, the terms "include," "comprise," and "have," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or elements is not limited to the listed steps or elements, but may optionally include steps or elements not listed, or may optionally include other steps or elements inherent to the process, method, product, or apparatus.

[0026] The terms "first", "second" and the like in the specification, claims and drawings of the present invention are used to distinguish different objects rather than to describe a specific order.

[0027] The following is a clear and complete description of the technical solutions of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the embodiments described are part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of the present invention.

[0028] Each axis of the high-precision turntable can rotate independently. The combination of these axes orients the optical sensor mounted on the turntable to a specific spatial orientation. Each specific spatial orientation corresponds to a sampling angle, and each sampling angle can be considered a sampling point. Therefore, to ensure sampling efficiency and accuracy, the sampling points must fully cover different spatial orientations and angles.

[0029] However, the inventors have discovered that the current optical sensor sampling method based on a high-precision turntable still has the following problems:

[0030] 1. Nonlinear imaging:

[0031] The optical system of an optical sensor is mostly based on the principle of pinhole imaging. The angle of rotation of the incident light and the displacement of the image are nonlinear and approximately tangent (for example, the non-uniform displacement of the light spot when a flashlight is rotated). For example, when the incident light is rotated from 59° to 60° and when the incident light is rotated from 0° to 1°, the position of the image is different.

[0032] 2. The number of samples is limited by the field of view:

[0033] The field of view of the optical sensor is limited. If the step length of the turntable is too large, the number of sampling points will be too small, and key areas may be easily missed. If the step length is too small, the number of sampling points will increase sharply (usually thousands to tens of thousands of points are required), and the overlap rate of adjacent fields of view will be too high, resulting in serious data redundancy.

[0034] 3. Sampling time is limited by the dynamic performance of the turntable:

[0035] High-precision turntables require high rotational stability. Each angle adjustment requires a cycle of acceleration, constant speed, deceleration, and stabilization, taking several to tens of seconds. Current sampling methods use point-by-point scanning with uniform steps, resulting in a total sampling time of several or even dozens of hours. This long sampling cycle also increases wear and tear on the equipment and energy consumption due to the long operation of the turntable.

[0036] Therefore, the current optical sensor sampling method based on a high-precision turntable still has problems such as insufficient coverage, high redundancy, and long sampling period, which lead to low sampling efficiency.

[0037] According to one aspect of the present invention, a sampling method for an optical sensor is provided. Figure 1 A flow chart of a sampling method according to an embodiment of the present invention is shown. Figure 1 As shown, the sampling method may include steps S100 to S700. Exemplarily, the sampling method may be executed by a sampling system with computing capabilities.

[0038] According to an example embodiment, the optical sensor may be an optical sensor based on the pinhole imaging principle, and the optical system of the optical sensor includes multiple apertures (e.g., N apertures). The optical sensor is disposed on a turntable (a mechanical platform that can rotate about a fixed axis). For example, the turntable may be a high-precision turntable (including a three-axis turntable with rotatable A-axis, B-axis, and C-axis). The following description uses a high-precision turntable as an example.

[0039] According to an example embodiment, in step S100 , the sampling system determines an initial sampling point set according to an initial step size.

[0040] For example, the initial step length Δa is the angle of each rotation or movement of the turntable, and the initial step length Δa is a positive real number in degrees. For example, the initial step length Δa may be in the range of [0.1°, 10°].

[0041] The sampling system traverses the rotation angle a of the A axis within a preset angle range (e.g., -90° to 90°) based on the initial step size Δa. For each rotation angle a, the sampling system traverses the rotation angle b of the B axis within a preset angle range (e.g., -90° to 90°) based on the initial step size Δa.

[0042] The rotation angle c of the C-axis is set to 0, and the sampling system determines the angle of the sampling point to be defined as (a, b, 0). For each rotation angle a, the sampling system traverses the rotation angle b of the B-axis within a preset angle range (e.g., -90° to 90°) based on the initial step size Δa, and sets the rotation angle c of the C-axis to 0, thereby determining the initial sampling point set S1 = {(a, b, 0)}.

[0043] As an embodiment, taking Δa=1 as an example, for each rotation angle a, all rotation angles b are traversed. For example, when the rotation angle a=-90°, S1={(-90°, -90°, 0), (-90°, -89°, 0) ... (90°, 89°, 0), (90°, 90°, 0), and the same applies to other rotation angles a.

[0044] In step S200 , the sampling system determines a set of observation vectors of all sampling points in the initial sampling point set relative to a known observation object.

[0045] For example, the sampling system can determine the observation vector corresponding to each sampling point in the initial sampling point set based on the turntable's attitude matrix at each sampling point and the position unit vector of the known observed object. The sampling system can then aggregate the observation vectors corresponding to all sampling points to form an observation vector set. That is, in this observation vector set, each sampling point corresponds to an observation vector. For example, the known observed object is the object observed by the optical sensor.

[0046] Figure 2 Another flow chart of the sampling method according to an embodiment of the present invention is shown. Figure 2 As shown, step S200 may further include steps S210-230.

[0047] Alternatively, as Figure 2 As shown, in step S210, the sampling system determines the posture matrix of the turntable at each sampling point.

[0048] For example, the sampling system can determine the attitude matrix D corresponding to the turntable at each sampling point (a, b, 0) based on the Euler rotation formula. The attitude matrix D can be a 3*3 matrix representing rotation. The attitude matrix D can represent the rotational attitude of the turntable's coordinate system relative to the coordinate system of the known observed object.

[0049] In step S220, the sampling system determines the position unit vector of the known observed object based on the observation vector of the posture matrix according to the posture matrix.

[0050] In step S230 , the sampling system determines an observation vector set based on the observation vectors corresponding to all sampling points.

[0051] For example, the position unit vector represents the direction of the known observed object in its coordinate system and is expressed as a unit vector. The observation vector represents the direction vector of the known observed object in the optical sensor's own coordinate system.

[0052] For example, the sampling system can calculate the observation vector ω based on the product of the posture matrix D and the position unit vector ν0, such as: ω = D*ν0.

[0053] The sampling system collects the observation vectors corresponding to each sampling point to obtain the observation vector set S2 = {ω}.

[0054] As an embodiment, a sampling point P is (-50°, 20°, 0). Then, on a turntable with ZXZ Euler rotation, the attitude matrix D can be:

[0055]

[0056] And, the position unit vector of the known observation object is ν0 = [0, 1, 0] T , then the corresponding observation vector ω=[0.76604,-0.60402,0.219846] T .

[0057] Through the above embodiments, the present invention calculates the attitude matrix of the turntable at each sampling point in real time, accurately converts the position unit vector of the known observed object into an observation vector based on the turntable coordinate system, and can minimize the accumulation of observation errors caused by attitude changes.

[0058] In step S300 , the sampling system traverses the set of observation vectors based on each aperture of the optical sensor, determines all sampling points that can cover the aperture, and determines a subset of observers corresponding to the aperture based on all sampling points.

[0059] For example, suppose the optical system of the optical sensor has N apertures, where the coordinates of the kth aperture are [X k , Y k , H k ].

[0060] For each aperture, the sampling system traverses the observation vector set S2 and calculates the ideal two-dimensional imaging coordinates (x k ,y k ) and the coordinates of the kth aperture [X k , Y k , H k ], according to the three-dimensional vector [x k -X k ,y k -Y k , -H k ]T The property of being parallel to the observation vector ω determines the ideal two-dimensional imaging coordinates (x k ,y k ) can fall within the imaging range of the optical sensor detector. If it can fall within, it means that the sampling point (a, b, 0) corresponding to the observation vector ω can cover the aperture, and the sampling system adds the sampling point (a, b, 0) corresponding to the observation vector ω to the observer subset Θ corresponding to the aperture. k middle.

[0061] According to an example embodiment, in step S300 , the sampling system determines a total set of observers based on subsets of observers corresponding to all apertures.

[0062] For example, the sampling system traverses all N apertures, thereby obtaining N observer subsets corresponding to the N apertures. The sampling system aggregates the N observer subsets, thereby obtaining the total observer set Θ = {Θ1, Θ2, Θ3...Θ k ...Θ N}.

[0063] As an example, assume that the optical system of the optical sensor has 900 apertures, the coordinates of the first aperture are [X1, Y1, H1] = [-8000, -6000, 4000] (unit is micrometer, the same below), the coordinates of the 200th aperture are [X 200 , Y 200 , H 200 ]=[-3500, 6200, 4000], the detector imaging range of the optical sensor is a rectangular range of plus or minus 2800 microns in the x direction and plus or minus 2100 microns in the y direction.

[0064] For a certain sampling point P, the observation vector ω=[0.76604,-0.60402,0.219846] T , we can calculate that the ideal two-dimensional imaging coordinates (x1, y1) of the observation vector ω through the first aperture are (-3472.6, -13159.7). This ideal two-dimensional imaging coordinate exceeds the imaging range of the above detector, so the sampling point P cannot cover the first aperture. For a certain sampling point P, the observation vector ω = [0.76604, -0.60402, 0.219846] T , the ideal two-dimensional imaging coordinates (x 200 ,y 200 ) is (1027.4, -959.7). This ideal two-dimensional imaging coordinate falls within the detector imaging range, so the sampling point P can cover the 200th aperture. Therefore, the sampling point P is added to the observer subset corresponding to the 200th aperture.

[0065] Figure 3 Another flow chart of the sampling method according to an embodiment of the present invention is shown. Figure 3 As shown, after step S300, the sampling method may further include steps S400a-S400c.

[0066] Optionally, in step S400a, the sampling system determines the number of elements in each observer subset in the total observer set.

[0067] In step S400b, the sampling system adjusts the initial step size to a modified step size when the minimum number of elements in all observer subsets is less than a preset threshold.

[0068] For example, Card(·) is the number of elements in the observer subset, and the sampling system takes Card(Θ1), Card(Θ2), ..., Card(Θ N ) and other minimum values ​​of the number of elements Mincard. When the minimum value Mincard is less than the preset threshold Q, the sampling system modifies the initial step size Δa based on the adjustment coefficient γ, such as the modified step size The calculation formula can be: Among them, 0<γ<1.

[0069] For example, the preset threshold Q may be the minimum number of observations that each aperture is expected to be covered by sampling points.

[0070] As an example, the initial step size Δa is set to 1, Q is set to 30, and the calculated minimum value Mincard is set to 17. In this case, the initial step size needs to be readjusted. Set γ to 0.75, and the corrected step size is

[0071] In step S400c, the sampling system determines the total set of observers based on the modified step size.

[0072] For example, when the initial step length Δa is adjusted to the modified step length Afterwards, the sampling system is based on the modified step size Steps S100 to S300 are executed again to obtain an updated total set of observers. With this arrangement, the present invention can ensure that there are enough elements in the observer subset corresponding to each aperture, so that the aperture can be fully observed.

[0073] Figure 4 FIG. 1 shows a distribution diagram of all sampling points in the total set of observers according to an embodiment of the present invention. For example, after steps S100-S300, the sampling system can obtain the following: Figure 4The total set of observers shown includes multiple sampling points (approximately 13,385 sampling points).

[0074] In step S400 , the sampling system determines a covering matrix for the total set of apertures and observers.

[0075] For example, the coverage matrix Φ is a two-dimensional matrix about whether the aperture can be covered by the sampling points in the total set of observers. The two-dimensional matrix is ​​an N*M matrix, where N is the total number of apertures and M is the number of sampling points in the total set of observers.

[0076] Exemplarily, the elements in the two-dimensional matrix are binary values ​​(such as 0 or 1). For example, if the element (i, j) in the two-dimensional matrix is ​​1, it means that the i-th aperture can be covered by the j-th sampling point; if the element (i, j) in the two-dimensional matrix is ​​0, it means that the i-th aperture cannot be covered by the j-th sampling point.

[0077] As an embodiment, when the number of apertures N is 900 and M=Card(Θ)=13385, the covering matrix Φ is a two-dimensional matrix with 900 rows and 13385 columns.

[0078] In step S500 , the sampling system determines the coverage contribution value of the sampling point to the aperture based on the coverage matrix.

[0079] For example, the sampling system determines that the aperture is not adequately covered.

[0080] For example, the sampling system creates a state vector of length N, which is used to record the number of target sampling points Θ for each aperture. * The current number of coverages, the target sampling point set Θ * Initialized to an empty set.

[0081] For each aperture, the sampling system considers the current coverage number to be less than the target coverage number K and the maximum possible coverage number T k The minimum value (i.e. min(K, T k )) of the aperture, marked as insufficiently covered aperture.

[0082] The target coverage times K can be customized according to actual needs, and the maximum possible coverage times T k is the maximum possible coverage number of the aperture among all sampling points.

[0083] As an embodiment, at the initial iteration, since the target sampling point set Θ * It is an empty set. Taking the 200th aperture as an example, its current coverage number is 0, so in the initial iteration, the aperture is temporarily marked as an insufficiently covered aperture.

[0084] The sampling system determines the coverage contribution of each sampling point to the under-covered aperture.

[0085] For example, for all unselected sampling points, the sampling system calculates the potential contribution of each sampling point to the insufficiently covered apertures. The potential contribution may represent the number of insufficiently covered apertures that may be newly covered by the sampling point.

[0086] Exemplarily, the sampling system can calculate the coverage contribution value of the sampling by summing the insufficiently covered rows of the corresponding column in the coverage matrix.

[0087] In step S600 , the sampling system determines a target sampling point according to the coverage contribution value, so as to update the target sampling point to a target sampling point set.

[0088] For example, the sampling system determines the sampling point with the largest coverage contribution value among all unselected sampling points as the target sampling point, and updates the target sampling point to the target sampling point set Θ * middle.

[0089] For example, if there are multiple sampling points with the same coverage contribution value, the target sampling point can be determined by random selection or based on a preset rule (such as the one with the smallest number). The preset rule can be customized based on user needs.

[0090] As an embodiment, in the first iteration, since the sum of the 3675th column is the largest, the coverage contribution value of this column is the largest, so the sampling system selects the sampling point of the 3675th column as the target sampling point and updates the target sampling point to the target sampling point set Θ * middle.

[0091] In step S700, the sampling system updates the coverage matrix, iterates multiple times until the preset conditions are met, and updates the target sampling points determined in each iteration to the target sampling point set to obtain the final target sampling point set.

[0092] For example, the sampling system updates the current coverage times and coverage matrix Φ of all apertures and continues multiple iterations (repeatedly executing steps S400-S600) to determine a target sampling point in each iteration process and update the target sampling point to the target sampling point set Θ * .

[0093] When the preset conditions are met, the sampling system stops iterating to obtain the final target sampling point set Θ * The preset conditions can be customized according to user needs.

[0094] Illustratively, the preset condition may include at least one of the following:

[0095] 1. There is no insufficiently covered aperture, that is, the current coverage times of all apertures are not less than min(K, T k );

[0096] 2. There are no remaining unselected sampling points to choose from;

[0097] 3. There are no unselected sampling points that can further increase the coverage of any aperture, that is, there are no sampling points that can provide potential contributions.

[0098] According to example embodiments, the sampling system may ultimately output a target sampling point set including a plurality of target sampling points.

[0099] Figure 5 FIG. 4 shows a distribution diagram of all sampling points in the target sampling point set according to an embodiment of the present invention. For example, after steps S400-S700, the sampling system can obtain the following: Figure 5 The target sampling point set shown includes multiple target sampling points (about 2230 sampling points).

[0100] Optionally, after step S700 , the sampling system may further reorder the target sampling point set to obtain a sampling sequence of the target sampling points, so that the optical sensor performs sampling based on the sampling sequence.

[0101] For example, the sampling system can perform dual sorting based on the rotation angle a and the rotation angle b. For example, the sampling system can first perform primary sorting based on the size of the rotation angle a, and then perform secondary sorting based on the size of the rotation angle b within the sampling points with the same rotation angle a, and finally sort the target sampling point set Θ * Reorder the target sampling points in .

[0102] The sampling system can also divide the target sampling points into several segments, the rotation angles a of the target sampling points in adjacent segments are monotonically sorted, the target sampling points in each segment have the same rotation angle a, and the rotation angles b are also monotonically sorted.

[0103] Optionally, the sampling system traverses the reordered target sampling point set Θ in segmented order * If jumping to a new segment, the target sampling points in the new segment are reversely sorted according to the rotation angle b (such as changing the ascending sort to the descending sort, or changing the descending sort to the ascending sort).

[0104] Figure 6 A schematic diagram showing the sampling sequence of an embodiment of the present invention is shown. Figure 6As shown in the above embodiment, the present invention reorders the target sampling points so that all target sampling points can be sampled in an orderly manner based on a certain rotation angle. In this way, the present invention can minimize the time consumption caused by switching sampling points, thereby reducing sampling time and improving sampling efficiency.

[0105] Through the above embodiments, the present invention can select a minimum subset from the initial sampling point set to determine that each aperture of the optical sensor can be covered by sufficient sampling points based on the target azimuth angle, thereby maximizing the coverage of the observation.

[0106] The present invention can also fully consider the sequential rotation characteristics of the turntable by reordering the target sampling point set, thereby maximizing the avoidance of the additional control system stabilization time penalty caused by disordered rotation, reducing the sampling time and improving the sampling efficiency.

[0107] According to exemplary embodiments, the sampling method provided by the present invention can also be used in high-precision turntable correction scenarios. For example, when the optical sensor's correction range is large, the field of view is small, or high precision is required, the present invention uses coverage as a constraint to optimize the sampling strategy. Compared to conventional methods, this method can significantly shorten the total sampling time, effectively improving sampling efficiency and the overall utilization rate of the turntable. This is not a limitation of the present invention.

[0108] According to yet another aspect of the present invention, a sampling system for an optical sensor is provided. Figure 7 FIG. 1 is a schematic diagram showing the structure of a sampling system according to an embodiment of the present invention. Figure 7 As shown, the sampling system 1 includes a sampling point set processing module 10 and a target sampling point determination module 20 .

[0109] According to an example embodiment, the sampling point set processing module 10 determines an initial sampling point set according to an initial step size.

[0110] For example, the initial step length Δa is the angle of each rotation or movement of the turntable, and the initial step length Δa is a positive real number in degrees. For example, the initial step length Δa may be in the range of [0.1°, 10°].

[0111] The sampling point set processing module 10 traverses the rotation angle a of the A axis within a preset angle range (e.g., -90° to 90°) based on the initial step size Δa. For each rotation angle a, the sampling point set processing module 10 traverses the rotation angle b of the B axis within a preset angle range (e.g., -90° to 90°) based on the initial step size Δa.

[0112] The C-axis rotation angle c is set to 0, and the sampling point set processing module 10 determines the sampling point angle definition as (a, b, 0). For each rotation angle a, the sampling point set processing module 10 traverses the B-axis rotation angle b within a preset angle range (e.g., -90° to 90°) based on the initial step size Δa, and sets the C-axis rotation angle c to 0, thereby determining the initial sampling point set S1 = {(a, b, 0)}.

[0113] As an embodiment, taking Δa=1 as an example, for each rotation angle a, all rotation angles b are traversed. For example, when the rotation angle a=-90°, S1={(-90°, -90°, 0), (-90°, -89°, 0) ... (90°, 89°, 0), (90°, 90°, 0), and the same applies to other rotation angles a.

[0114] The sampling point set processing module 10 determines a set of observation vectors of all sampling points in the initial sampling point set relative to the known observation object.

[0115] For example, the sampling point set processing module 10 can determine the observation vector corresponding to each sampling point in the initial sampling point set based on the turntable's posture matrix at each sampling point and the position unit vector of the known observed object. The sampling point set processing module 10 can then aggregate the observation vectors corresponding to all sampling points to form an observation vector set. That is, in this observation vector set, each sampling point corresponds to an observation vector. For example, the known observed object is the object observed by the optical sensor.

[0116] Optionally, the sampling point set processing module 10 determines the attitude matrix of the turntable at each sampling point.

[0117] For example, the sampling point set processing module 10 can determine the attitude matrix D corresponding to the turntable at each sampling point (a, b, 0) based on the Euler rotation formula. The attitude matrix D can be a 3*3 matrix representing rotation. The attitude matrix D can represent the rotational attitude of the coordinate system of the turntable relative to the coordinate system of the known observed object.

[0118] The sampling point set processing module 10 determines the position unit vector of the known observed object based on the observation vector of the posture matrix according to the posture matrix.

[0119] The sampling point set processing module 10 determines an observation vector set according to the observation vectors corresponding to all sampling points.

[0120] For example, the position unit vector represents the direction of the known observed object in its coordinate system and is expressed as a unit vector. The observation vector represents the direction vector of the known observed object in the optical sensor's own coordinate system.

[0121] Exemplarily, the sampling point set processing module 10 can calculate the observation vector ω according to the product of the posture matrix D and the position unit vector ν0, such as: ω = D*ν0.

[0122] The sampling point set processing module 10 collects the observation vectors corresponding to each sampling point to obtain an observation vector set S2 = {ω}.

[0123] As an embodiment, a sampling point P is (-50°, 20°, 0). Then, on a turntable with ZXZ Euler rotation, the attitude matrix D can be:

[0124]

[0125] And, the position unit vector of the known observation object is ν0 = [0, 1, 0] T , then the corresponding observation vector ω=[0.76604,-0.60402,0.219846] T .

[0126] Through the above embodiments, the present invention calculates the attitude matrix of the turntable at each sampling point in real time, accurately converts the position unit vector of the known observed object into an observation vector based on the turntable coordinate system, and can minimize the accumulation of observation errors caused by attitude changes.

[0127] The sampling point set processing module 10 traverses the observation vector set based on each aperture of the optical sensor, determines all sampling points that can cover the aperture, and determines the observer subset corresponding to the aperture based on all sampling points.

[0128] For example, suppose the optical system of the optical sensor has N apertures, where the coordinates of the kth aperture are [X k , Y k , H k ].

[0129] For each aperture, the sampling point set processing module 10 traverses the observation vector set S2 and calculates the ideal two-dimensional imaging coordinates (x k ,y k ) and the coordinates of the kth aperture [X k , Y k , H k ], according to the three-dimensional vector [x k -X k ,y k -Y k , -H k ] T The property of being parallel to the observation vector ω determines the ideal two-dimensional imaging coordinates (x k ,y k) can fall within the imaging range of the optical sensor detector. If it can fall within, it means that the sampling point (a, b, 0) corresponding to the observation vector ω can cover the aperture, and the sampling point set processing module 10 adds the sampling point (a, b, 0) corresponding to the observation vector ω to the observer subset Θ corresponding to the aperture. k middle.

[0130] According to an example embodiment, the sampling point set processing module 10 determines a total set of observers based on the subsets of observers corresponding to all apertures.

[0131] For example, the sampling point set processing module 10 traverses all N apertures to obtain N observer subsets corresponding to the N apertures. The sampling point set processing module 10 aggregates the N observer subsets to obtain the total observer set:

[0132] Θ={Θ1,Θ2,Θ3...Θ k ...Θ N}.

[0133] As an example, assume that the optical system of the optical sensor has 900 apertures, the coordinates of the first aperture are [X1, Y1, H1] = [-8000, -6000, 4000] (unit is micrometer, the same below), the coordinates of the 200th aperture are [X 200 , Y 200 , H 200 ]=[-3500, 6200, 4000], the detector imaging range of the optical sensor is a rectangular range of plus or minus 2800 microns in the x direction and plus or minus 2100 microns in the y direction.

[0134] For a certain sampling point P, the observation vector ω=[0.76604,-0.60402,0.219846] T , we can calculate that the ideal two-dimensional imaging coordinates (x1, y1) of the observation vector ω through the first aperture are (-3472.6, -13159.7). This ideal two-dimensional imaging coordinate exceeds the imaging range of the above detector, so the sampling point P cannot cover the first aperture. For a certain sampling point P, the observation vector ω = [0.76604, -0.60402, 0.219846] T , the ideal two-dimensional imaging coordinates (x 200 ,y 200 ) is (1027.4, -959.7). This ideal two-dimensional imaging coordinate falls within the detector imaging range, so the sampling point P can cover the 200th aperture. Therefore, the sampling point P is added to the observer subset corresponding to the 200th aperture.

[0135] Optionally, the sampling point set processing module 10 determines the number of elements in each observer subset in the total observer set.

[0136] The sampling point set processing module 10 adjusts the initial step size to a modified step size when the minimum number of elements in all observer subsets is less than a preset threshold.

[0137] For example, Card(·) is the number of elements in the observer subset, and the sampling point set processing module 10 takes Card(Θ1), Card(Θ2), ..., Card(Θ N ) and the minimum value Mincard of the number of elements. When the minimum value Mincard is less than the preset threshold Q, the sampling point set processing module 10 modifies the initial step length Δa based on the adjustment coefficient γ, such as the modified step length The calculation formula can be: Among them, 0<γ<1.

[0138] For example, the preset threshold Q may be the minimum number of observations that each aperture is expected to be covered by sampling points.

[0139] As an example, the initial step size Δa is set to 1, Q is set to 30, and the calculated minimum value Mincard is set to 17. In this case, the initial step size needs to be readjusted. Set γ to 0.75, and the corrected step size is

[0140] The sampling point set processing module 10 determines the total set of observers based on the modified step size.

[0141] For example, when the initial step length Δa is adjusted to the modified step length Afterwards, the sampling point set processing module 10 processes the sample point set based on the modified step size. The above steps are repeated to obtain an updated total set of observers. With this arrangement, the present invention can ensure that there are enough elements in the observer subset corresponding to each aperture so that the aperture can be fully observed.

[0142] The target sampling point determination module 20 determines a coverage matrix related to the aperture and the total set of observers.

[0143] For example, the coverage matrix Φ is a two-dimensional matrix about whether the aperture can be covered by the sampling points in the total set of observers. The two-dimensional matrix is ​​an N*M matrix, where N is the total number of apertures and M is the number of sampling points in the total set of observers.

[0144] Exemplarily, the elements in the two-dimensional matrix are binary values ​​(such as 0 or 1). For example, if the element (i, j) in the two-dimensional matrix is ​​1, it means that the i-th aperture can be covered by the j-th sampling point; if the element (i, j) in the two-dimensional matrix is ​​0, it means that the i-th aperture cannot be covered by the j-th sampling point.

[0145] As an embodiment, when the number of apertures N is 900 and M=Card(Θ)=13385, the covering matrix Φ is a two-dimensional matrix with 900 rows and 13385 columns.

[0146] The target sampling point determination module 20 determines the coverage contribution value of the sampling point to the aperture based on the coverage matrix.

[0147] For example, the target sampling point determination module 20 determines that the aperture is not adequately covered.

[0148] Exemplarily, the target sampling point determination module 20 creates a state vector of length N, which is used to record the number of target sampling points set Θ for each aperture. * The current number of coverages, the target sampling point set Θ * Initialized to an empty set.

[0149] For each aperture, the target sampling point determination module 20 determines whether the current coverage number is less than the target coverage number K and the maximum possible coverage number T k The minimum value (i.e. min(K, T k )) of the aperture, marked as insufficiently covered aperture.

[0150] The target coverage times K can be customized according to actual needs, and the maximum possible coverage times T k is the maximum possible coverage number of the aperture among all sampling points.

[0151] As an embodiment, at the initial iteration, since the target sampling point set Θ * It is an empty set. Taking the 200th aperture as an example, its current coverage number is 0, so in the initial iteration, the aperture is temporarily marked as an insufficiently covered aperture.

[0152] The target sampling point determination module 20 determines the coverage contribution value of each sampling point to the insufficiently covered aperture.

[0153] For example, for all unselected sampling points, the target sampling point determination module 20 calculates the potential contribution of each sampling point to the insufficiently covered aperture. The potential contribution may represent the number of insufficiently covered apertures that may be newly covered by the sampling point.

[0154] Exemplarily, the target sampling point determination module 20 may calculate the coverage contribution value of the sampling by summing the insufficiently covered rows of the corresponding column in the coverage matrix.

[0155] The target sampling point determination module 20 determines the target sampling point according to the coverage contribution value, so as to update the target sampling point to the target sampling point set.

[0156] For example, the target sampling point determination module 20 determines the sampling point with the largest coverage contribution value among all unselected sampling points as the target sampling point, and updates the target sampling point to the target sampling point set Θ * middle.

[0157] For example, if there are multiple sampling points with the same coverage contribution value, the target sampling point determination module 20 may select the target sampling point randomly or based on a preset rule (such as the one with the smallest number). The preset rule may be customized based on user needs.

[0158] As an embodiment, in the first iteration, since the sum of the 3675th column is the largest, the coverage contribution value of this column is the largest, and the target sampling point determination module 20 selects the sampling point of the 3675th column as the target sampling point, and updates the target sampling point to the target sampling point set Θ * middle.

[0159] The target sampling point determination module 20 updates the coverage matrix, iterates multiple times until the preset conditions are met, and updates the target sampling points determined in each iteration to the target sampling point set to obtain the final target sampling point set.

[0160] For example, the target sampling point determination module 20 updates the current coverage times and coverage matrix Φ of all apertures, and continues multiple iterations (repeatedly performing the above steps) to determine a target sampling point in each iteration process, and updates the target sampling point to the target sampling point set Θ * .

[0161] When the preset conditions are met, the target sampling point determination module 20 stops iterating to obtain the final target sampling point set Θ * The preset conditions can be customized according to user needs.

[0162] Illustratively, the preset condition may include at least one of the following:

[0163] 1. There is no insufficiently covered aperture, that is, the current coverage times of all apertures are not less than min(K, T k );

[0164] 2. There are no remaining unselected sampling points to choose from;

[0165] 3. There are no unselected sampling points that can further increase the coverage of any aperture, that is, there are no sampling points that can provide potential contributions.

[0166] According to an example embodiment, the target sampling point determination module 20 may eventually output a target sampling point set including a plurality of target sampling points.

[0167] Optionally, the target sampling point determination module 20 may further reorder the target sampling point set to obtain a sampling sequence of the target sampling points, so that the optical sensor performs sampling based on the sampling sequence.

[0168] For example, the target sampling point determination module 20 may perform dual sorting based on the rotation angle a and the rotation angle b. For example, the target sampling point determination module 20 may first perform primary sorting based on the size of the rotation angle a, and then perform secondary sorting based on the size of the rotation angle b within the sampling points with the same rotation angle a, and finally sort the target sampling point set θ. * Reorder the target sampling points in .

[0169] The target sampling point determination module 20 may further divide the target sampling points into several segments, wherein the rotation angles a of the target sampling points in adjacent segments are monotonically sorted, the target sampling points in each segment have the same rotation angle a, and the rotation angles b are also monotonically sorted.

[0170] Optionally, the target sampling point determination module 20 traverses the reordered target sampling point set Θ in a segmented order. * If jumping to a new segment, the target sampling points in the new segment are reversely sorted according to the rotation angle b (such as changing the ascending sort to the descending sort, or changing the descending sort to the ascending sort).

[0171] Through the above embodiment, the present invention reorders the target sampling points so that all target sampling points can be sampled in an orderly manner based on a certain rotation angle. In this way, the present invention can minimize the time consumption caused by switching sampling points, thereby reducing sampling time and improving sampling efficiency.

[0172] Through the above embodiments, the present invention can select a minimum subset from the initial sampling point set to determine that each aperture of the optical sensor can be covered by sufficient sampling points based on the target azimuth angle, thereby maximizing the coverage of the observation.

[0173] The present invention can also fully consider the sequential rotation characteristics of the turntable by reordering the target sampling point set, thereby maximizing the avoidance of the additional control system stabilization time penalty caused by disordered rotation, reducing the sampling time and improving the sampling efficiency.

[0174] According to exemplary embodiments, the sampling system provided by the present invention can also be used in high-precision turntable correction scenarios. For example, when the optical sensor's correction range is large, the field of view is small, or high precision is required, the present invention uses coverage as a constraint to optimize the sampling strategy. Compared to conventional methods, this can significantly shorten the total sampling time, effectively improving sampling efficiency and the overall utilization rate of the turntable. This is not a limitation of the present invention.

[0175] According to yet another aspect of the present invention, an electronic device is provided. The electronic device includes: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, enables the one or more processors to implement the method described above.

[0176] According to another aspect of the present invention, a non-volatile computer-readable storage medium is provided, wherein a computer program is stored on the storage medium, and when the computer program is executed by a processor, the method described above can be implemented.

[0177] According to another aspect of the present invention, a computer program product is provided, which includes a computer program stored on a computer-readable storage medium; the computer program includes program instructions, which, when executed by a computer, cause the computer to execute the method described above.

[0178] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art will be able to modify the technical solutions of the aforementioned embodiments or substitute equivalents for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention shall be included within the scope of protection of the present invention.

Claims

1. A sampling method for an optical sensor, characterized in that: include: Determine the initial sampling point set according to the initial step size; Determine a set of observation vectors of all sampling points in the initial sampling point set relative to a known observation object; Traversing the set of observation vectors based on each aperture of the optical sensor, determining all sampling points that can cover the aperture, determining a subset of observers corresponding to the aperture based on all sampling points, and determining a total set of observers based on the subsets of observers corresponding to all apertures; Determining a covering matrix for the aperture and the total set of observers; Determining coverage contribution values ​​of sampling points to the aperture based on the coverage matrix; Determine a target sampling point according to the coverage contribution value, so as to update the target sampling point to a target sampling point set; The covering matrix is ​​updated, and it is iterated multiple times until a preset condition is satisfied, and the target sampling points determined in each iteration are updated to the target sampling point set to obtain a final target sampling point set.

2. The sampling method according to claim 1, characterized in that Determining a set of observation vectors of all sampling points in the initial sampling point set relative to a known observation object includes: Determine the attitude matrix of the turntable at each sampling point; Determining, based on the attitude matrix, an observation vector of a position unit vector of the known observed object based on the attitude matrix; The observation vector set is determined according to the observation vectors corresponding to all the sampling points.

3. The sampling method according to claim 1, characterized in that After determining the total set of observers based on the subsets of observers corresponding to all the apertures, the sampling method further includes: Determining the number of elements in each observer subset of the total observer set; When the minimum number of elements in all observer subsets is less than a preset threshold, adjusting the initial step size to a revised step size; The total set of observers is re-determined based on the revised step size.

4. The sampling method according to claim 1, characterized in that After obtaining the final target sampling point set, the sampling method further includes: The target sampling point set is reordered to obtain a sampling sequence of target sampling points, so that the optical sensor performs sampling based on the sampling sequence.

5. A sampling system for an optical sensor, characterized in that: include: a sampling point set processing module, which determines an initial sampling point set based on an initial step size, determines an observation vector set of all sampling points in the initial sampling point set relative to a known observation object, and traverses the observation vector set based on each aperture of the optical sensor to determine all sampling points that can cover the aperture, thereby determining an observer subset corresponding to the aperture based on all sampling points, and further determining a total observer set based on the observer subsets corresponding to all apertures; A target sampling point determination module determines a coverage matrix between the aperture and the total set of observers, determines a coverage contribution value of a sampling point to the aperture based on the coverage matrix, determines a target sampling point based on the coverage contribution value, updates the target sampling point to the target sampling point set, and updates the coverage matrix multiple times until a preset condition is satisfied. The target sampling point determined in each iteration is updated to the target sampling point set to obtain a final target sampling point set.

6. The sampling system according to claim 5, characterized in that The sampling point set processing module determines the attitude matrix of the turntable at each sampling point; The sampling point set processing module determines, based on the posture matrix, an observation vector of the position unit vector of the known observed object based on the posture matrix; The sampling point set processing module determines the observation vector set according to the observation vectors corresponding to all the sampling points.

7. The sampling system according to claim 5, characterized in that The sampling point set processing module determines the number of elements in each observer subset in the total observer set; The sampling point set processing module adjusts the initial step size to a modified step size when the minimum number of elements in all observer subsets is less than a preset threshold; The sampling point set processing module re-determines the total set of observers based on the revised step size.

8. The sampling system according to claim 5, characterized in that The target sampling point determination module reorders the target sampling point set to obtain a sampling sequence of target sampling points, so that the optical sensor performs sampling based on the sampling sequence.

9. An electronic device, characterized in that: include: one or more processors; a storage device for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 4.

10. A non-volatile computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images