Visual navigation positioning method and device, electronic equipment and storage medium
By continuously acquiring environmental images with a monocular camera and performing point-of-interest matching and domino-like image processing, the problem of positioning inaccuracy caused by satellite signal obstruction and inertial navigation errors was solved, achieving continuous and accurate navigation and positioning.
Patent Information
- Application Number
- CN202511124146.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-12
- Publication Date
- 2025-10-21
AI Technical Summary
In applications such as autonomous vehicles, drones, and robots, existing technologies suffer from positioning inaccuracies due to satellite signal blockage and accumulated errors in inertial navigation.
By continuously acquiring environmental images with a monocular camera, detecting and matching points of interest, and using epipolar geometric constraints to calculate the motion information of the monocular camera and the three-dimensional spatial coordinates of the points of interest, domino-like processing is performed to achieve continuous navigation and positioning.
In the absence of satellite signal obstruction, continuous and accurate navigation and positioning were achieved, improving positioning accuracy.
Smart Images
Figure CN120820144A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of navigation and positioning technology, and in particular to a visual navigation and positioning method, device, electronic device and storage medium. Background Art
[0002] Navigation and positioning are crucial in applications such as self-driving cars, drones, and robots, helping devices perceive their surroundings and plan their paths. Currently, related technologies primarily rely on a combination of satellite navigation and inertial navigation. This approach works well when satellite signals are stable, but can be blocked in places like tunnels and buildings. Furthermore, the accumulated errors from inertial navigation can increase, leading to inaccurate positioning. Summary of the Invention
[0003] In view of the above-mentioned defects or deficiencies in the related art, it is desired to provide a visual navigation and positioning method, device, electronic device and storage medium that can continuously navigate and position with improved accuracy.
[0004] In a first aspect, the present application provides a visual navigation positioning method, the visual navigation positioning method comprising:
[0005] Obtain at least two environment images continuously captured by a monocular camera;
[0006] respectively detecting points of interest in the two environmental images, and matching the points of interest in the two environmental images to obtain a matching combination;
[0007] determining the first motion information of the monocular camera and the three-dimensional spatial coordinates of the point of interest according to the matching combination, wherein the three-dimensional spatial coordinates of the point of interest are obtained based on the first motion information of the monocular camera;
[0008] A domino folding process is performed on the environmental image captured at a moment after the at least two environmental images according to the three-dimensional spatial coordinates of the point of interest to obtain second motion information of the monocular camera.
[0009] Optionally, in some embodiments of the present application, determining the first motion information of the monocular camera according to the matching combination includes:
[0010] The two-dimensional pixel coordinates of each interest point in the matching combination in the corresponding environment image are obtained, and first motion information of the monocular camera is calculated based on the two-dimensional pixel coordinates and epipolar geometry constraints.
[0011] Optionally, in some embodiments of the present application, the first motion information of the monocular camera is obtained by the following formula:
[0012]
[0013] In the above formula, E represents the essential matrix, and its value is t 12 ^R 12 , t 12 Represents the translation matrix of the monocular camera from the first position to the second position, R 12 represents the rotation matrix of the monocular camera moving from the first position to the second position, ^ represents the outer product operation; F represents the basic matrix, whose value is A -T t 12 ^R 12 A -1 , A represents the intrinsic parameter matrix of the monocular camera, T represents the transpose; x1 = A -1 p1, x2 = A -1 p2, p1 and p2 represent the two-dimensional pixel coordinates of each interest point in the corresponding environment image.
[0014] Optionally, in some embodiments of the present application, the depth factor s1 in the three-dimensional space coordinates of the interest point is obtained by the following formula:
[0015] s1x2^R 12 x1+x2^t 12 =0.
[0016] Optionally, in some embodiments of the present application, respectively detecting points of interest in the two environment images includes:
[0017] A corner point response value of each pixel point in the environment image is calculated, and if the corner point response value is greater than a preset threshold, the pixel point corresponding to the corner point response value is used as the interest point.
[0018] Optionally, in some embodiments of the present application, the corner point response value R is obtained by the following formula:
[0019] R=detM-k(traceM) 2 ;
[0020] In the above formula, detM=αβ-γ 2 , traceM=α+β, k represents a constant ranging from 0.04 to 0.06; ω(x,y) represents the Gaussian weighted function centered at pixel (x,y) in the environment image I(x,y), I x Indicates the area obtained in the x direction, I y Indicates the area obtained in the y direction.
[0021] Optionally, in some embodiments of the present application, matching the points of interest in the two environment images to obtain a matching combination includes:
[0022] The descriptors of the interest points in the environment image are calculated, and the descriptors of the interest points are matched using a brute force matching algorithm to obtain the matching combination.
[0023] In a second aspect, the present application provides a visual navigation and positioning device, the visual navigation and positioning device comprising:
[0024] An acquisition module is used to acquire at least two environment images continuously captured by a monocular camera;
[0025] a matching module, configured to detect the points of interest of the two environmental images respectively, and match the points of interest of the two environmental images to obtain a matching combination;
[0026] a determination module, configured to determine the first motion information of the monocular camera and the three-dimensional spatial coordinates of the point of interest according to the matching combination, wherein the three-dimensional spatial coordinates of the point of interest are obtained based on the first motion information of the monocular camera;
[0027] The domino-falling processing module is used to perform domino-falling processing on the environmental image collected at a time after the at least two environmental images according to the three-dimensional spatial coordinates of the point of interest, so as to obtain the second motion information of the monocular camera.
[0028] In a third aspect, the present application provides an electronic device comprising a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the instruction, the program, the code set or the instruction set is loaded and executed by the processor to implement the steps of the visual navigation positioning method described in any one of the first aspects.
[0029] In a fourth aspect, the present application provides a computer-readable storage medium, which stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the visual navigation positioning method described in any one of the first aspects.
[0030] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:
[0031] The embodiments of the present application provide a visual navigation and positioning method, device, electronic device and storage medium, which continuously capture at least two environmental images through a monocular camera without using satellite signals, and then use the environmental images as a processing basis to detect points of interest in the two environmental images respectively, and match the points of interest in the two environmental images to obtain matching combinations, so that the first motion information of the monocular camera and the three-dimensional spatial coordinates of the points of interest can be determined based on these matching combinations, and thereby the environmental images captured at the time after the at least two environmental images are subjected to domino-reversing processing to obtain the second motion information of the monocular camera, thereby realizing continuous navigation and positioning without satellite signal obstruction, and effectively improving accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0033] Figure 1 A flowchart of a visual navigation positioning method provided in an embodiment of the present application;
[0034] Figure 2 A schematic diagram of an epipolar geometry constraint provided in an embodiment of the present application;
[0035] Figure 3 A schematic diagram of the principle of a domino flipping process provided in an embodiment of the present application;
[0036] Figure 4 A structural block diagram of a visual navigation and positioning device provided in an embodiment of the present application;
[0037] Figure 5 This is a structural block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0038] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.
[0039] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the accompanying drawings and specific implementation methods.
[0040] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of this application can be combined with each other. Figures 1 to 5 The visual navigation positioning method, device, electronic device and storage medium provided in the embodiments of the present application are described in detail.
[0041] Please refer to Figure 1 , which is a flow chart of a visual navigation positioning method provided in an embodiment of the present application, the method specifically includes the following steps:
[0042] S101, obtaining at least two environment images continuously captured by a monocular camera.
[0043] Exemplarily, in the embodiment of the present application, continuous collection refers to uninterrupted collection, for example, collection once every 1 second, thereby ensuring the continuity of the surrounding environment.
[0044] S102 , detecting interest points of the two environment images respectively, and matching the interest points of the two environment images to obtain a matching combination.
[0045] In some embodiments of the present application, the Harris corner detection algorithm can be used to calculate the corner response value of each pixel in the environment image. If the corner response value is greater than a preset threshold, the pixel corresponding to the corner response value is regarded as the point of interest. For example, the autocorrelation function E(m,n) is set as:
[0046] E(m,n)=∑ω(x,y)[I(x+m,y+n)-I(x,y)] 2 (1)
[0047] In formula (1), ω(x,y) represents the Gaussian weighted function centered at the pixel point (x,y) in the environment image I(x,y), m represents the offset in the x direction, and n represents the offset in the y direction.
[0048] Expand I(x+m,y+n)-I(x,y) to get:
[0049] [I(x+m,y+n)-I(x,y)]=[I(x,y)+I x m+I y n+o(m 2 ,n 2 )-I(x,y)](2) then the autocorrelation function is updated as:
[0050] E(m,n)=∑ω(x,y)[I x m+I y n] 2 (3)
[0051] Further, it can be written as:
[0052]
[0053]
[0054] In formula (5), I x Indicates the area obtained in the x direction, I y Indicates the area obtained in the y direction.
[0055] Furthermore, the autocorrelation function is simplified into a quadratic function, namely:
[0056] E(m,n)=αm+βn 2 +2γmn (6)
[0057] The corner point response value R is obtained by formula (7):
[0058] R=detM-k(traceM) 2 (7)
[0059] In formula (7), det M = αβ - γ 2 , traceM=α+β, k represents a constant ranging from 0.04 to 0.06.
[0060] In some embodiments of the present application, the descriptors of each point of interest in the environmental image can be calculated, and the descriptors of the points of interest can be matched using a brute force matching algorithm to obtain a matching combination. It should be noted that a descriptor is a vector used to describe local features in an image, which can capture image information around the point of interest so as to match and identify between different images, while the brute force (BF) matching algorithm matches each possible descriptor. For example, the environmental image is divided into 4*4 equal intervals, and then non-maximum suppression processing is performed on each segmented area. The feature matching adopts a BRIEF-like description algorithm, selects a 7*7 neighborhood, and takes the near circle points 1-16, and squares and normalizes the pixel values of points 1-16 and point p to limit the range of all pixel values to the [0,1] interval. The normalized pixel values of points 1-16 are subtracted from the pixel value of the center point p to obtain a 16-dimensional normalized difference vector desc
[16] , and a threshold is set for brute force matching to obtain a matching combination.
[0061] S103 : determining first motion information of the monocular camera and three-dimensional spatial coordinates of the point of interest according to the matching combination, where the three-dimensional spatial coordinates of the point of interest are obtained based on the first motion information of the monocular camera.
[0062] In some embodiments of the present application, the two-dimensional pixel coordinates of each point of interest in the corresponding environment image can be obtained by matching and combining, and the first motion information of the monocular camera can be calculated based on the two-dimensional pixel coordinates and the epipolar geometry constraint. Figure 2 As shown in the schematic diagram of epipolar geometry constraint, the monocular camera first captures the first environment image I1 at the first position O1, and then moves to the second position O2 to capture the second environment image I2. The corresponding points of interest of a point P in space on the two environment images are p1 and p2 respectively. The plane formed by the three points P, O1 and O2 intersects the environment images I1 and I2 at the poles e1 and e2 respectively. The straight line formed by the pole e1 and the interest point p1 is the epipolar line l1, and the straight line formed by the pole e2 and the interest point p2 is the epipolar line l2. The epipolar geometry constraint means that the interest point p2 must be located on the projection line of the O1p1 ray on the second environment image I2.
[0063] Furthermore, the coordinate system of the monocular camera when taking the first environmental image I1 is set to the world coordinate system, and the coordinates of point P in the monocular camera coordinate system are set to:
[0064] P c =P w =[XYZ] T (8)
[0065] The relationship between the coordinates of point P and the two two-dimensional pixel coordinates is:
[0066] s1p1=AP c1 (9)
[0067] s2p2=AP c2 (10)
[0068] P c2 =R 12 P c1 +t 12 (11)
[0069] In equations (9) to (11), s1 and s2 represent the depth factors corresponding to each environment image, p1 and p2 represent the two-dimensional pixel coordinates of each interest point in the corresponding environment image, and A represents the intrinsic parameter matrix of the monocular camera; 12 Represents the translation matrix of the monocular camera from the first position O1 to the second position O2, R 12 Represents the rotation matrix of the monocular camera moving from the first position O1 to the second position O2.
[0070] Furthermore, set x1 = A -1 p1, x2 = A -1 p2, then:
[0071] x2=R 12 x1+t 12 (12)
[0072] In formula (12), x1 and x2 represent the coordinates of each interest point on the normalized plane. This equality is called equality in the sense of scale. 12 Do the outer product, ^ represents the outer product operation, and multiply on the left have to:
[0073]
[0074] Set x1 = A -1 p1, x2 = A -1 Substituting p2 into formula (13), we get:
[0075]
[0076] The physical meanings of Equations (13) and (14) are both epipolar geometry constraints. Let E = t 12 ^R 12 , F=A -T t 12 ^R 12 A -1 , E represents the essential matrix, F represents the basic matrix, T represents the transpose, and E and F are respectively substituted into formula (13) and formula (14), and we get:
[0077]
[0078] That is to say, the essential matrix E contains all the motion information of the monocular camera, namely translation and rotation. Then, the first motion information of the monocular camera is obtained by substituting the two-dimensional pixel coordinates of each point of interest into equations (15) and (16) to calculate the essential matrix E. The translation matrix t can be obtained by decomposing the essential matrix E. 12 and the rotation matrix R 12 .
[0079] In some embodiments of the present application, triangulation is a process of restoring the three-dimensional coordinates of an object based on environmental images captured at different angles. From the epipolar geometry constraint formula, we know that:
[0080] s2x2=s1R 12 x1+t 12 (17)
[0081] Furthermore, by multiplying both sides of equation (17) by x^2, we can obtain:
[0082] s2x2^x2=0=s1x2^R 12 x1+x2^t 12 (18)
[0083] The right side of equation (18) can be regarded as the equation of the depth factor s1. According to this equation, the depth factor s1 can be directly calculated, and then the three-dimensional spatial coordinates of the point of interest in the monocular camera coordinate system can be obtained.
[0084] S104 , performing a domino folding process on the environmental image captured at a time after at least two environmental images according to the three-dimensional spatial coordinates of the point of interest, to obtain second motion information of the monocular camera.
[0085] In some embodiments of the present application, Figure 3 The schematic diagram of the domino-tumbling process is shown. From the figure, we can see that there is a similar relationship between the triangles. O represents the optical center of the monocular camera, that is:
[0086] ΔOab∽ΔOAB(19)
[0087] ΔOac∽ΔOAC(20)
[0088] ΔObc∽ΔOBC(21)Furthermore, according to the cosine theorem, we can obtain:
[0089] OA 2 +OB 2 -2OA·OB cos<a,b> =AB 2 (twenty two)
[0090] OA 2 +OC 2 -2OA·OC cos<a,c> =AC 2 (twenty three)
[0091] OB 2 +OC 2 -2OB·OC cos<b,c> =BC 2 (twenty four)
[0092] Divide both sides of equations (22) to (24) by OC 2 ,make but:
[0093] x 2 +y 2 -2xycos<a,b> -v=0 (25)
[0094] x 2 +1-2xcos<a,c> -wv=0 (26)
[0095] y 2 +1-2ycos<b,c> -uv=0 (27)
[0096] Substituting v in formula (25) into formula (26) and formula (27), we get:
[0097] (1-u)y 2 -ux 2 -cos<b,c> y+2uxycos<a,b> +1=0 (28)
[0098] (1-w)x 2 -wy 2 -cos<a,c> x+2wxycos<a,b> +1=0 (29)
[0099] In equations (28) and (29), after knowing the three-dimensional spatial coordinates of A, B, and C and the two-dimensional pixel coordinates of a, b, and c, u, w, cos<a,b> 、cos<a,c> and cos<b,c> All are known quantities, only x and y are unknown quantities. Therefore, the two unknown quantities x and y can be calculated through equations (28) and (29), and then the second motion information of the monocular camera, namely the translation matrix and rotation matrix, can be obtained.
[0100] The visual navigation and positioning method provided in the embodiment of the present application continuously captures at least two environmental images through a monocular camera without using satellite signals. Then, based on the environmental images, the points of interest of the two environmental images are detected respectively, and the points of interest of the two environmental images are matched to obtain matching combinations. Therefore, the first motion information of the monocular camera and the three-dimensional spatial coordinates of the points of interest can be determined based on these matching combinations. The environmental images captured at the time after the at least two environmental images are then subjected to a domino-reversing process to obtain the second motion information of the monocular camera, thereby realizing continuous navigation and positioning without satellite signal obstruction and effectively improving accuracy.
[0101] Based on the above embodiments, the present application provides a visual navigation and positioning device. The visual navigation and positioning device 100 can be applied to Figures 1 to 3 In the visual navigation positioning method of the corresponding embodiment. Please refer to Figure 4 , the visual navigation and positioning device 100 includes:
[0102] An acquisition module 101 is configured to acquire at least two environment images continuously captured by a monocular camera;
[0103] The matching module 102 is used to detect the points of interest of the two environmental images respectively, and match the points of interest of the two environmental images to obtain a matching combination;
[0104] a determination module 103 for determining first motion information of the monocular camera and three-dimensional spatial coordinates of a point of interest according to the matching combination, wherein the three-dimensional spatial coordinates of the point of interest are obtained based on the first motion information of the monocular camera;
[0105] The domino-falling processing module 104 is configured to perform domino-falling processing on the environmental image captured at a time after at least two environmental images according to the three-dimensional spatial coordinates of the point of interest, so as to obtain second motion information of the monocular camera.
[0106] Optionally, in some embodiments of the present application, the determination module 103 is specifically used to obtain the two-dimensional pixel coordinates of each interest point in the matching combination in the corresponding environment image, and calculate the first motion information of the monocular camera based on the two-dimensional pixel coordinates and the epipolar geometric constraints.
[0107] Optionally, in some embodiments of the present application, the first motion information of the monocular camera is obtained by the following formula:
[0108]
[0109] In the above formula, E represents the essential matrix, and its value is t 12 ^R 12 , t 12 Represents the translation matrix of the monocular camera from the first position to the second position, R 12 Represents the rotation matrix of the monocular camera moving from the first position to the second position, ^ represents the outer product operation; F represents the basic matrix, whose value is A -T t 12 ^R 12 A -1 , A represents the intrinsic parameter matrix of the monocular camera, T represents the transpose; x1 = A -1 p1, x2 = A -1 p2, p1 and p2 represent the two-dimensional pixel coordinates of each interest point in the corresponding environment image.
[0110] Optionally, in some embodiments of the present application, the depth factor s1 in the three-dimensional space coordinates of the point of interest is obtained by the following formula:
[0111] s1x2^R 12 x1+x2^t 12 =0.
[0112] Optionally, in some embodiments of the present application, the matching module 102 is specifically used to calculate the corner point response value of each pixel point in the environment image. If the corner point response value is greater than a preset threshold, the pixel point corresponding to the corner point response value is used as the point of interest.
[0113] Optionally, in some embodiments of the present application, the corner point response value R is obtained by the following formula:
[0114] R=detM-k(traceM) 2 ;
[0115] In the above formula, det M = αβ-γ 2, traceM=α+β, k represents a constant ranging from 0.04 to 0.06; ω(x,y) represents the Gaussian weighted function centered at pixel (x,y) in the environment image I(x,y), I x Indicates the area obtained in the x direction, I y Indicates the area obtained in the y direction.
[0116] Optionally, in some embodiments of the present application, the matching module 102 is further specifically configured to calculate the descriptors of each interest point in the environment image, and match the descriptors of the interest points using a brute force matching algorithm to obtain a matching combination.
[0117] It should be noted that, for the description of the same steps and contents in this embodiment as those in other embodiments, reference can be made to the description in other embodiments and will not be repeated here.
[0118] The visual navigation and positioning device provided in the embodiment of the present application continuously captures at least two environmental images through a monocular camera without using satellite signals. Then, based on the environmental images as a processing reference, the points of interest of the two environmental images are respectively detected, and the points of interest of the two environmental images are matched to obtain matching combinations, so that the first motion information of the monocular camera and the three-dimensional spatial coordinates of the points of interest can be determined based on these matching combinations, and thereby a domino-reversing process is performed on the environmental images captured at the time after the at least two environmental images to obtain the second motion information of the monocular camera, thereby realizing continuous navigation and positioning without satellite signal obstruction, and effectively improving accuracy.
[0119] Based on the above embodiments, the present application provides an electronic device. Figure 5 The electronic device 200 may include a processor 201 and a memory 202. The memory 202 stores at least one instruction, at least one program, code set or instruction set, which is loaded and executed by the processor 201 to implement Figures 1 to 3 The steps of the visual navigation positioning method of the corresponding embodiment.
[0120] As another aspect, the present invention provides a computer-readable storage medium for storing program code for executing the aforementioned Figures 1 to 3 Any implementation of the visual navigation positioning method of the corresponding embodiment.
[0121] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and modules described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0122] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the modules is merely a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. On the other hand, the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms. The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the scheme of this embodiment.
[0123] In addition, the functional modules in the various embodiments of the present application may be integrated into a single processing unit, or each module may exist physically separately, or two or more units may be integrated into a single module. The aforementioned integrated units may be implemented in the form of hardware or in the form of software functional units. If the integrated units are implemented in the form of software functional units and sold or used as independent products, they may be stored in a computer-readable storage medium.
[0124] Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes a number of instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the visual navigation and positioning method of each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program code.
[0125] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0126] This document uses specific examples to illustrate the principles and implementation methods of this application. The description of the above examples is only intended to help understand the method and core concept of this application. At the same time, for those skilled in the art, based on the concept of this application, there may be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting this application.
Claims
1. A visual navigation positioning method, characterized in that: The visual navigation positioning method comprises: Obtain at least two environment images continuously captured by a monocular camera; respectively detecting points of interest in the two environmental images, and matching the points of interest in the two environmental images to obtain a matching combination; determining the first motion information of the monocular camera and the three-dimensional spatial coordinates of the point of interest according to the matching combination, wherein the three-dimensional spatial coordinates of the point of interest are obtained based on the first motion information of the monocular camera; A domino folding process is performed on the environmental image captured at a moment after the at least two environmental images according to the three-dimensional spatial coordinates of the point of interest to obtain second motion information of the monocular camera.
2. The visual navigation positioning method according to claim 1, characterized in that: The determining the first motion information of the monocular camera according to the matching combination includes: The two-dimensional pixel coordinates of each interest point in the matching combination in the corresponding environment image are obtained, and first motion information of the monocular camera is calculated based on the two-dimensional pixel coordinates and epipolar geometry constraints.
3. The visual navigation positioning method according to claim 2, characterized in that: The first motion information of the monocular camera is obtained by the following formula: In the above formula, E represents the essential matrix, and its value is t 12 ^R 12 , t 12 Represents the translation matrix of the monocular camera from the first position to the second position, R 12 Represents the rotation matrix of the monocular camera moving from the first position to the second position, ^ represents the outer product operation; F represents the basic matrix, whose value is A -T t 12 ^R 12 A -1 , A represents the intrinsic parameter matrix of the monocular camera, T represents the transpose; x1 = A -1 p1, x2 = A - 1 p2, p1 and p2 represent the two-dimensional pixel coordinates of each interest point in the corresponding environment image.
4. The visual navigation positioning method according to claim 3, characterized in that: The depth factor s1 in the three-dimensional space coordinates of the point of interest is obtained by the following formula: s1x2^R 12 x1+x2^t 12 =0。 5. The visual navigation positioning method according to any one of claims 1 to 4, characterized in that: The detecting the points of interest in the two environment images respectively includes: A corner point response value of each pixel point in the environment image is calculated, and if the corner point response value is greater than a preset threshold, the pixel point corresponding to the corner point response value is used as the interest point.
6. The visual navigation positioning method according to claim 5, characterized in that: The corner point response value R is obtained by the following formula: R=detM-k(traceM) 2 ; In the above formula, detM=αβ-γ 2 , traceM=α+β, k represents a constant ranging from 0.04 to 0.06; ω(x,y) represents the Gaussian weighted function centered at pixel (x,y) in the environment image I(x,y), I x Indicates the area obtained in the x direction, I y Indicates the area obtained in the y direction.
7. The visual navigation positioning method according to claim 5, characterized in that: The matching of the points of interest of the two environment images to obtain a matching combination includes: The descriptors of the interest points in the environment image are calculated, and the descriptors of the interest points are matched using a brute force matching algorithm to obtain the matching combination.
8. A visual navigation and positioning device, characterized in that: The visual navigation and positioning device comprises: An acquisition module, configured to acquire at least two environment images continuously captured by a monocular camera; a matching module, configured to detect points of interest in the two environment images respectively, and match the points of interest in the two environment images to obtain a matching combination; a determination module, configured to determine the first motion information of the monocular camera and the three-dimensional spatial coordinates of the point of interest according to the matching combination, wherein the three-dimensional spatial coordinates of the point of interest are obtained based on the first motion information of the monocular camera; The domino-falling processing module is used to perform domino-falling processing on the environmental image collected at a time after the at least two environmental images according to the three-dimensional spatial coordinates of the point of interest, so as to obtain the second motion information of the monocular camera.
9. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory stores at least one instruction, at least one program, code set or instruction set, and the instruction, the program, the code set or the instruction set is loaded and executed by the processor to implement the steps of the visual navigation positioning method described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the steps of the visual navigation positioning method according to any one of claims 1 to 7.