Unmanned agricultural machinery navigation method, equipment, device and storage medium

Abstract

The present invention discloses a navigation method, equipment, device, and storage medium for an unmanned agricultural machine. The method comprises: determining the geodetic coordinates of characteristic corner points of graphic signage located on or on both sides of the unmanned agricultural machine's travel path; locating the geodetic coordinates of the unmanned agricultural machine while the machine is in motion; capturing an image of the surrounding environment; utilizing image edge detection technology to extract edge contours of patterns in the image of the surrounding environment, thereby identifying vertices of the pattern in the image of the surrounding environment; utilizing the vertices as characteristic corner points; and calculating the pixel coordinates of the characteristic corner points; and utilizing the geodetic coordinates and pixel coordinates of the characteristic corner points to correct and calibrate the geodetic coordinates of the unmanned agricultural machine. This method utilizes computer vision technology to correct the position coordinates of the unmanned agricultural machine without relying on vehicle-mounted RTK equipment and technology, achieving centimeter-level high-precision positioning of the unmanned agricultural machine, reducing equipment costs and improving positioning and navigation accuracy.

Description

Technical Field

[0001] The present invention relates to the fields of computer vision and assisted autonomous driving. More specifically, it relates to a navigation method, equipment, device, and storage medium for unmanned agricultural machinery. Background Art

[0002] my country is a major agricultural country with vast arable land, accounting for 7% of the world's arable land. However, compared with developed countries like Europe and the United States, our current level of agricultural mechanization remains relatively low, and industrial development needs to be improved. Furthermore, with the deepening of urbanization, an increasing number of young rural workers are leaving their hometowns for urban employment, resulting in a continuous decline in the rural labor force, which has had a significant impact on agricultural operations and the rural economy. Therefore, unmanned agricultural machinery has become a key application scenario for autonomous driving (unmanned driving) technology in recent years. Powered by autonomous driving technology, agricultural machinery can automate a variety of operations such as tilling, leveling, sowing, and harvesting, greatly offsetting the shortage of young workers and professional operators, and playing a significant role in improving agricultural efficiency and automation levels.

[0003] Due to the need for precise operations, unmanned agricultural machinery (UAVs) require very high navigation and positioning accuracy. This typically requires the use of real-time kinematic (RTK) carrier phase differential measurement equipment installed on UAVs. Using RTK technology, UAVs can achieve and maintain centimeter-level high-precision positioning capabilities, providing the foundation for operations such as driving, control, and path planning. However, the widespread application of RTK equipment in farming currently faces limitations. First, the cost of using RTK remains high. When using third-party commercial services, users must pay significant equipment and service fees, a cost that becomes even more significant when the number of agricultural machinery is large. Second, RTK service coverage is currently unavailable in many remote areas. Consequently, users must establish and maintain their own RTK base stations and communication infrastructure, which incurs increased labor and equipment costs. Therefore, obtaining an accurate, reliable, and cost-effective high-precision positioning reference has become a prominent pain point in the large-scale application of UAVs. Summary of the Invention

[0004] In order to solve the above technical problems or at least partially solve the above technical problems, the present disclosure provides an unmanned agricultural machinery navigation method, device and storage medium for achieving accurate and reliable navigation and positioning of unmanned agricultural machinery without relying on vehicle-mounted RTK equipment and technology.

[0005] An embodiment of the present invention provides a navigation method for an unmanned agricultural machinery, the method comprising: determining the geodetic coordinates of characteristic corner points of a graphic indicator set on or on both sides of a driving path of the unmanned agricultural machinery; locating the geodetic coordinates of the unmanned agricultural machinery during its driving process, capturing a surrounding environment image, and using image edge detection technology to extract the edge contour of a pattern in the surrounding environment image, thereby identifying the vertices of the pattern in the surrounding environment image, using the vertices as characteristic corner points, and calculating the pixel coordinates of the characteristic corner points; and using the geodetic coordinates and pixel coordinates of the characteristic corner points to correct and calibrate the geodetic coordinates of the unmanned agricultural machinery.

[0006] Preferably, determining the characteristic angle of the graphic sign set on the driving path of the unmanned agricultural machinery or on both sides of the driving path according to the method of claim 1, the geodetic coordinates of the point include: using an RTK high-precision receiver to statically measure and collect the coordinate position of the graphic sign within a period of time, taking the average of the coordinate positions of the graphic sign collected within the period of time, and calculating the geodetic coordinates of the characteristic corner point of the graphic sign.

[0007] Preferably, using image edge detection technology, extracting the edge contour of the pattern in the surrounding environment image includes calculating the gradient and angle of the pattern in the surrounding environment image; performing non-maximum suppression on the gradient; and connecting the edges of the pattern using double thresholds until the complete edge contour of the pattern is extracted.

[0008] Preferably, calculating the gradient and angle of the pattern in the surrounding environment image includes: calculating the gradient and angle of the pattern in the surrounding environment image according to the following formula: Wherein, x is the horizontal coordinate of the pixel point of the pattern in the surrounding environment image, y is the vertical coordinate of the pixel point of the pattern in the surrounding environment image, and f is the grayscale value of the pattern in the surrounding environment image. An angle matrix is ​​calculated for all pixel points of the pattern in the surrounding environment image.

[0009] Preferably, performing non-maximum suppression on the gradient includes: determining whether the grayscale value of the currently detected point C in the gradient is the largest within an 8-connected neighborhood; if it is the largest, continuing to check whether the grayscale values ​​of the first intersection point dTmp1 and the second corner point dTmp2 of the gradient direction in the gradient are greater than C; if C is greater than the grayscale values ​​of the first intersection point dTmp1 and the second corner point dTmp2, determining C as a maximum and setting the value of C to 1; otherwise, determining C as a non-maximum and setting the value of C to 0; traversing all points C in the gradient to find the local maximum of the pixel points in the gradient, and completing the non-maximum suppression of the gradient.

[0010] Preferably, using a double threshold to connect the edges of the pattern until the complete edge contour of the pattern is extracted includes selecting two thresholds, the two thresholds including a low threshold and a high threshold, considering points less than the low threshold as false edges and setting them to 0, and considering points greater than the high threshold as strong edges and setting them to 1; based on the high threshold points in the image, first connect them into contours, and when the breakpoint of the contour is reached, the algorithm will search for a point that meets the low threshold in the 8-threshold area of ​​the breakpoint, and then collect new edges based on this point until the entire image is closed.

[0011] Preferably, using the geodetic coordinates and pixel coordinates of the characteristic corner points to correct and calibrate the geodetic coordinates of the unmanned agricultural machine includes: calculating the corrected coordinates of the unmanned agricultural machine according to the following formula: Among them, p′ j is the geodetic coordinate of the feature corner point, p j is the corrected coordinate of the unmanned agricultural machine to be optimized, z′ ij For unmanned agricultural machinery in position T i Observe the characteristic corner point p′ j The resulting pixel measurement data, e ij is the error function, h(T i , pj) is the projection function from world coordinates to pixel coordinates.

[0012] On the other hand, an embodiment of the present invention provides an unmanned agricultural machinery navigation device, wherein the device includes: a feature corner point geodetic coordinate determination device, used to determine the geodetic coordinates of the feature corner points of the graphic signboards set on the driving path of the unmanned agricultural machinery or on both sides of the driving path; a feature corner point pixel coordinate calculation device, which locates the geodetic coordinates of the unmanned agricultural machinery during its driving process, collects the surrounding environment image, and uses image edge detection technology to extract the edge contour of the pattern in the surrounding environment image, thereby identifying the vertices of the pattern in the surrounding environment image, using the vertices as feature corner points, and calculating the pixel coordinates of the feature corner points; a coordinate calibration device, used to use the geodetic coordinates and pixel coordinates of the feature corner points to correct and calibrate the geodetic coordinates of the unmanned agricultural machinery.

[0013] On the other hand, an embodiment of the present invention also provides an unmanned agricultural machinery navigation device, which includes: a processor, a memory, and program instructions executable by the processor. When the program instructions are executed by the processor, the device for measuring the position coordinates of the indoor wireless signal transmission anchor point performs the following operations: determining the geodetic coordinates of the characteristic corner points of the graphic signboards set on the driving path of the unmanned agricultural machinery or on both sides of the driving path; locating the geodetic coordinates of the unmanned agricultural machinery during its driving process, collecting the surrounding environment image, and using image edge detection technology to extract the edge contour of the pattern in the surrounding environment image, thereby identifying the vertices of the pattern in the surrounding environment image, using the vertices as characteristic corner points, and calculating the pixel coordinates of the characteristic corner points; using the geodetic coordinates and pixel coordinates of the characteristic corner points to correct and calibrate the geodetic coordinates of the unmanned agricultural machinery.

[0014] On the other hand, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the method according to any one of claims 1 to 7 when executed by a processor.

[0015] Compared with the prior art, the technical solution provided by the embodiments of the present disclosure has the following advantages: the present disclosure sets graphic signs with characteristic corner points on both sides of the driving path of the unmanned agricultural machinery, and uses image edge detection technology during driving to identify the characteristic corner points of the graphic signs and calculate the pixel coordinates of the characteristic corner points. Then, the geodetic coordinates and pixel coordinates of the characteristic corner points are used to correct and calibrate the geodetic coordinates of the unmanned agricultural machinery. In this way, the position coordinates of the unmanned agricultural machinery can be corrected by using computer vision technology without relying on vehicle-mounted RTK equipment and technology, thereby achieving centimeter-level high-precision positioning of the unmanned agricultural machinery, meeting the high-precision operation requirements of the unmanned agricultural machinery, greatly reducing equipment costs, and improving positioning and navigation accuracy.

[0016] Other features and advantages of the embodiments of the present invention will be described in detail in the subsequent detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0018] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0019] Figure 1This is a flow chart of a navigation method for an unmanned agricultural machine according to an embodiment of the present disclosure;

[0020] Figure 2 An exemplary graphic sign according to an embodiment of the present disclosure;

[0021] Figure 3 This is a schematic diagram of the arrangement of the graphic signboard according to an embodiment of the present disclosure;

[0022] Figure 4 This is a schematic diagram of performing non-maximum suppression on a gradient according to an embodiment of the present disclosure;

[0023] Figure 5 This is a structural block diagram of an unmanned agricultural machinery navigation device according to an embodiment of the present disclosure;

[0024] Figure 6 This is a schematic diagram of an unmanned agricultural machinery navigation device according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0025] In order to enable those skilled in the art to better understand the technical solution of the present disclosure, the present disclosure is described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present disclosure are further described in detail below in conjunction with the accompanying drawings and specific embodiments, but are not intended to limit the present disclosure.

[0026] With the rapid development of my country's independent BeiDou satellite navigation system, the BeiDou-3 satellite navigation system has begun to provide open and free sub-meter-level Satellite-Based Augmentation System (SBAS) services and decimeter-level Precise Point Positioning (PPP) services to the industry and the public. In addition, unmanned agricultural machinery is generally equipped with or has been equipped with inertial navigation systems (INS), camera sensors and other equipment. Therefore, the inventors propose that by integrating BeiDou SBAS / PPP services with inertial, visual and other navigation methods, it is possible to enable unmanned agricultural machinery to obtain continuous, stable and reliable navigation and positioning results with absolute decimeter-level accuracy.

[0027] Based on this, Figure 1 As shown, the present invention provides a navigation method for unmanned agricultural machinery, the method comprising:

[0028] S101, determining the geodetic coordinates of characteristic corner points of graphic signs set on or on both sides of the driving path of the unmanned agricultural machine;

[0029] Among them, graphic signs with characteristic corner points that can be clearly identified by computer vision are set up on the driving path of the unmanned agricultural machinery or on both sides of the driving path.

[0030] In some embodiments, the graphic sign may be in the form of a square, rectangle, triangle, QR code, etc. Figure 2 shown.

[0031] When unmanned agricultural machinery works in the fields, it often adopts the A / B point operation mode, moving from point A to point B according to the preset driving path, such as Figure 3 Therefore, in some embodiments, the graphic signboard can be arranged on the driving path of the unmanned agricultural machine, such as at the end of a ridge; or it can be arranged on both sides of the driving path of the unmanned agricultural machine, such as on both sides of a ridge.

[0032] In a preferred embodiment, in order to accurately obtain the geodetic coordinates of the graphic sign and to correct the positioning results of subsequent unmanned agricultural machinery, the coordinate position of the graphic sign can be statically measured and collected in advance over a period of time using, for example, a high-precision satellite positioning receiver, and the coordinate position of the graphic sign collected over the period of time can be averaged to calculate the geodetic coordinates of the characteristic corner points of the graphic sign.

[0033] Furthermore, when acquiring the coordinates of the characteristic corner points, it is preferable to ensure that the phase center of the satellite positioning receiver antenna coincides with the characteristic corner points of the graphic sign as much as possible, thereby ensuring accurate measurement of the geodetic coordinates of the characteristic corner points of the graphic sign. For example, the satellite positioning receiver antenna can be placed at the vertices of a square or rectangular sign. If it is difficult to accurately align the phase center of the satellite positioning receiver antenna with the characteristic corner points, manual calibration and correction can be performed by the surveyor.

[0034] S102: While the unmanned agricultural machine is traveling, the geodetic coordinates of the unmanned agricultural machine are located, an image of the surrounding environment is captured, and edge contours of patterns in the surrounding environment image are extracted using image edge detection technology. Vertices of the patterns in the surrounding environment image are thereby identified as feature corner points, and pixel coordinates of the feature corner points are calculated.

[0035] In some embodiments, the two-dimensional pixel coordinates of the feature corner points can be obtained by the following processing: a camera sensor mounted on the unmanned agricultural machine captures an image of the surrounding environment, and for each frame of the image, image edge detection technology is used to extract the edge contours of the image or pattern, thereby identifying the vertices of the image or pattern and using them as feature corner points. The edge detection operator can be selected from the Canny operator, Roberts operator, Sobel operator, Marr-Hildreth operator, etc.

[0036] In order to find an optimal edge, for example, to identify as many actual edges in the image as possible; the identified edges should be as close as possible to the edges in the actual image; the edges in the image can only be identified once, and any image noise that may exist therein should not be identified as an edge. In a preferred embodiment, using image edge detection technology, extracting the edge contour of the pattern in the surrounding environment image may include the following steps:

[0037] S201: Calculate the gradient and angle of the pattern in the surrounding environment image.

[0038] S202, performing non-maximum suppression on the gradient;

[0039] S203 , connecting the edges of the pattern using a double threshold until a complete edge contour of the pattern is extracted.

[0040] In step S201, gradient is a very important concept in artificial intelligence, which is prevalent in the fields of machine learning and deep learning. The first-order differential of a one-dimensional function is defined as formula (1):

[0041]

[0042] Where f(x) is a differential function with respect to the unknown number x, x is the unknown number, and ε is a small variable with respect to the unknown number.

[0043] Image filtering is generally based on grayscale images, and the image is two-dimensional. Therefore, it is necessary to perform two-dimensional function differentiation, that is, formula (2) and formula (3):

[0044]

[0045]

[0046] Where f(x,y) is a two-dimensional function of unknowns x and y.

[0047] As can be seen from the above formula, the image gradient is the partial derivative of the current pixel with respect to the X-axis and Y-axis, so the gradient can also be understood as the rate of change of the pixel grayscale value in the field of image processing.

[0048] The modulus of the gradient represents the amount by which f(x,y) increases per unit distance in the direction of its maximum rate of change, that is:

[0049]

[0050] Among them, G is the modulus of the image gradient.

[0051] The calculation of the gradient angle is relatively simple, and its role is to provide a basis for the direction of non-maximum suppression. The calculation formula is as follows:

[0052]

[0053] Among them, x is the horizontal coordinate of the pixel in the image, y is the vertical coordinate of the pixel in the image, and f is the grayscale value of the image. Calculate the resulting angle matrix for all pixels.

[0054] In step S202, non-maximum suppression is performed on the gradient.

[0055] The gradient obtained in step S201 has problems such as thick and wide edges and weak edge interference. To solve this problem, non-maximum suppression can be used to find the local maximum of the pixel points and set the grayscale value corresponding to the non-maximum value to 0. In this way, a large number of non-edge pixels can be eliminated.

[0056] like Figure 4 As shown in Figure 1, C represents the point currently being tested, and g1-g4 are its 8-connected neighborhood points. In the figure, the oblique line represents the gradient direction of point C calculated in the previous step.

[0057] Determine whether the grayscale value of the currently detected point C in the gradient is the largest in the 8-connected neighborhood. If it is the largest, continue to check whether the grayscale values ​​of the first intersection point dTmp1 and the second corner point dTmp2 in the gradient direction are greater than C. If C is greater than the grayscale values ​​of the first intersection point dTmp1 and the second corner point dTmp2, then C is considered to be a maximum value and the value of C is set to 1. Otherwise, C is considered to be a non-maximum value and the value of C is set to 0. Traverse all points C in the gradient to find the local maximum of the pixel points in the gradient, and complete the non-maximum suppression of the gradient.

[0058] It should be noted that the intersection of the gradient directions does not necessarily fall at the positions of the 8 points in the 8-neighborhood. Therefore, in actual applications, dTmp1 and dTmp2 use the grayscale value formed by bilinear interpolation of two adjacent points.

[0059] In step S203, after processing in steps S201 and S202, the edge quality of the pattern is already very high, but many false edges still exist. Therefore, to remove the false edges, a dual threshold can be used to connect the pattern edges. Specifically, two thresholds are selected, including a low threshold and a high threshold. Points below the low threshold are considered to be false edges and set to 0, while points above the high threshold are considered to be strong edges and set to 1. Pixels in between require further inspection.

[0060] According to the high threshold points in the image, they are first connected into contours. When the breakpoint of the contour is reached, a point that meets the low threshold is searched in the 8-bit area of ​​the breakpoint, and new edges are collected based on this point until the entire image is closed.

[0061] S204: Identify and calculate the pixel coordinates of the feature corner points on the edge contour line of the pattern.

[0062] For a certain pixel point j on the edge contour line, its pixel coordinates are marked as (uj, vj), and the pixel coordinates of its left and right adjacent pixels are (uj-1, vj-1) and (uj+1, vj+1). is greater than a certain threshold Γ, that is: When , the pixel point j is the characteristic corner point of the pattern.

[0063] The threshold Γ should be reasonably determined based on the geometric shape of the graphic sign. For example, when the graphic sign is a square or rectangle, the threshold Γ should be set close to 90°; when the graphic sign is an equilateral triangle, the threshold Γ should be set close to 60°.

[0064] Further preferably, before step S201, the collected surrounding environment image may be subjected to denoising processing using Gaussian filtering. Gaussian filtering is a linear smoothing filter that can be used to eliminate Gaussian white noise and is widely used in noise reduction processing in image processing.

[0065] S103: Using the geodetic coordinates and pixel coordinates of the characteristic corner points, correct and calibrate the geodetic coordinates of the unmanned agricultural machine.

[0066] This is actually a process of building and solving a constrained BA (Bundle Adjustment) model. For a classic Simultaneous Location and Mapping (SLAM) problem, we start from a point p in the world coordinate system, take into account the internal and external parameters and distortion of the camera mounted on the unmanned agricultural machine, and finally project it into pixel coordinates. The following steps are required:

[0067] S301, convert the world coordinates to camera coordinates, where the camera extrinsic parameters (R, t) will be used:

[0068] P′=Rp+t=[X′, Y′, Z′] T (6)

[0069] Where P′ is the point in the world coordinate system that falls on the physical imaging plane O′-x′-y′ after being projected through the pinhole O, [X′, Y′, Z′] T is the coordinate of P′, R is the rotation matrix from world coordinates to camera coordinates, and t is the translation vector from world coordinates to camera coordinates.

[0070] S302, project P′ onto the normalized plane to obtain normalized coordinates:

[0071] P c =[u c , v c , 1] T =[X′ / Z′, Y′ / Z′, 1] T (7)

[0072] Among them, P c is the point P′ projected on the normalized pixel plane, [u c , v c , 1] T P c 's coordinates.

[0073] S303, considering the distortion of the normalized coordinates, obtain the original pixel coordinates [u′ c , v′ c ] T Here, only radial distortion is considered:

[0074]

[0075] Among them, k1, k2, r c Parameters of the distortion correction polynomial.

[0076] S304, calculate pixel coordinates based on the internal reference model:

[0077]

[0078] The above process can be abstractly recorded as formula (10):

[0079] z=h(x,y) (10)

[0080] The above gives the detailed parameterization process of BA processing. Specifically, x here refers to the camera pose at this time, that is, the external parameters R, t, and its corresponding Lie group is denoted by T. The landmark feature point y is the 3D point p here, and the observation data is the pixel coordinates Based on the principle of least squares method, the error equation of this observation can be listed:

[0081] e=zh(T,p) (11)

[0082] Then, we take the measurements at other times into account and add a subscript to the error. Let z ij For unmanned agricultural machinery in position T i Observe the road signs j The data generated, then the overall cost function is

[0083]

[0084] Solving formula (12) is equivalent to adjusting the posture of the unmanned agricultural machine and the landmark features in the environment at the same time, which is the so-called BA.

[0085] On this basis, for the method of the present invention, the identifiable characteristic corner point in the graphic sign is denoted as p′ j , z′ ij For unmanned agricultural machinery in position T i Observe the characteristic corner point p′ j The data generated is due to the feature corner point p′ j The geodetic coordinates of are known, which provides us with the conditions for correction and constraint. Therefore, we can add the equation about p′ on the basis of equation (6): j and z′ ij The constraint equation is , and the cost function is rewritten as formula (13):

[0086]

[0087] Formula (13) is not a linear function. By performing nonlinear optimization on it, the high-precision navigation and positioning result of the unmanned agricultural machinery based on the vision correction of the present invention can be finally obtained.

[0088] Therefore, the coordinates of the corrected unmanned agricultural machine are calculated according to formula (13):

[0089]

[0090] Among them, p′ j is the geodetic coordinate of the feature corner point, p j is the corrected coordinate of the unmanned agricultural machine to be optimized, z′ ij For unmanned agricultural machinery in position T i Observe the characteristic corner point p′ j The generated pixel measurement data, eij is the error function, h(T i , pj) is the projection function from world coordinates to pixel coordinates.

[0091] It's also worth noting that, because the graphic signs remain visible for a period of time while the unmanned agricultural machine is in motion, the effects of the aforementioned corrections and restrictions persist during this period. By appropriately increasing the number of graphic signs along the path of the unmanned agricultural machine, the method of the present invention can achieve even better performance and results.

[0092] Compared with the prior art, the technical solution provided by the embodiments of the present disclosure has the following advantages: the present disclosure sets graphic signs with characteristic corner points on both sides of the driving path of the unmanned agricultural machinery, and uses image edge detection technology during driving to identify the characteristic corner points of the graphic signs and calculate the pixel coordinates of the characteristic corner points. Then, the geodetic coordinates and pixel coordinates of the characteristic corner points are used to correct and calibrate the geodetic coordinates of the unmanned agricultural machinery. In this way, the position coordinates of the unmanned agricultural machinery can be corrected by using computer vision technology without relying on vehicle-mounted RTK equipment and technology, thereby achieving centimeter-level high-precision positioning of the unmanned agricultural machinery, meeting the high-precision operation requirements of the unmanned agricultural machinery, greatly reducing equipment costs, and improving positioning and navigation accuracy.

[0093] An embodiment of the present invention further provides an unmanned agricultural machinery navigation device, wherein the device comprises:

[0094] A device for determining the geodetic coordinates of characteristic corner points, used to determine the geodetic coordinates of characteristic corner points of graphic signboards arranged on or on both sides of the driving path of the unmanned agricultural machine;

[0095] A device for calculating pixel coordinates of characteristic corner points locates the geodetic coordinates of the unmanned agricultural machine while the unmanned agricultural machine is in motion, collects an image of the surrounding environment, and uses image edge detection technology to extract edge contours of patterns in the surrounding environment image, thereby identifying vertices of the patterns in the surrounding environment image, and using the vertices as characteristic corner points, and calculating the pixel coordinates of the characteristic corner points;

[0096] The coordinate calibration device is used to correct and calibrate the geodetic coordinates of the unmanned agricultural machine using the geodetic coordinates and pixel coordinates of the characteristic corner points.

[0097] On the other hand, the embodiment of the present invention also provides an unmanned agricultural machinery navigation device, such as Figure 5 As shown, the device includes:

[0098] Processor 601,

[0099] The memory 602 includes program instructions executable by the processor. When the program instructions are executed by the processor, the device for determining the position coordinates of the indoor wireless signal transmission anchor point performs the following operations:

[0100] Determine the geodetic coordinates of characteristic corner points of graphic signboards set on or on both sides of the driving path of the unmanned agricultural machinery;

[0101] During the driving process of the unmanned agricultural machine, the geodetic coordinates of the unmanned agricultural machine are obtained by locating the machine, an image of the surrounding environment is collected, and an edge contour of a pattern in the surrounding environment image is extracted using image edge detection technology, thereby identifying vertices of the pattern in the surrounding environment image, using the vertices as feature corner points, and calculating the pixel coordinates of the feature corner points;

[0102] The geodetic coordinates of the unmanned agricultural machine are corrected and calibrated using the geodetic coordinates and pixel coordinates of the characteristic corner points.

[0103] An embodiment of the present disclosure provides a computer-readable storage medium, characterized in that a computer program is stored on the computer-readable storage medium. When the computer program is executed by a processor, the various processes of the above-mentioned drone positioning and navigation method are implemented, and the same technical effects can be achieved. To avoid repetition, they will not be described here.

[0104] The terms "first", "second" and similar terms used in this disclosure do not indicate any order, quantity or importance, but are only used to distinguish different parts. The terms "include" or "comprises" and similar terms mean that the elements before the term include the elements listed after the term, and do not exclude the possibility of also including other elements. The terms "upper", "lower", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0105] All terms (including technical or scientific terms) used in this disclosure have the same meaning as those understood by one of ordinary skill in the art to which this disclosure belongs, unless otherwise specifically defined. It should also be understood that terms defined in, for example, general dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or highly formal sense, unless explicitly defined herein.

[0106] Technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

Claims

1. A navigation method for an unmanned agricultural machine, the method comprising: Determine the geodetic coordinates of characteristic corner points of graphic signboards set on or on both sides of the driving path of the unmanned agricultural machinery; During the driving process of the unmanned agricultural machine, the geodetic coordinates of the unmanned agricultural machine are obtained by locating the machine, an image of the surrounding environment is collected, and an edge contour of a pattern in the surrounding environment image is extracted using image edge detection technology, thereby identifying vertices of the pattern in the surrounding environment image, using the vertices as feature corner points, and calculating the pixel coordinates of the feature corner points; The geodetic coordinates and pixel coordinates of the characteristic corner points are used to correct and calibrate the geodetic coordinates of the unmanned agricultural machine; wherein the coordinates of the corrected unmanned agricultural machine are calculated according to the following formula: Among them, p j ′ is the geodetic coordinate of the feature corner point, p j is the corrected coordinate of the unmanned agricultural machine to be optimized, z ′ i j is the posture of the unmanned agricultural machine in T i Observe the characteristic corner point p at j ′ The resulting pixel measurement data, e ij is the error function, h(T i , p j ) is the projection function from world coordinates to pixel coordinates.

2. The unmanned agricultural machine navigation method according to claim 1, wherein: Determine the characteristic angles of graphic signs set on or on both sides of the path of unmanned agricultural machinery According to the method of claim 1, the geodetic coordinates of the point include: The coordinate position of the graphic sign is statically measured and collected by using an RTK high-precision receiver over a period of time, the coordinate position of the graphic sign collected over the period of time is averaged, and the geodetic coordinates of the characteristic corner point of the graphic sign are calculated.

3. The unmanned agricultural machine navigation method according to claim 1, wherein: Extracting the edge contour of the pattern in the surrounding environment image using image edge detection technology includes: Calculating the gradient and angle of the pattern in the surrounding environment image; performing non-maximum suppression on the gradient; Double thresholds are used to connect the edges of the pattern until the complete edge contour of the pattern is extracted.

4. The unmanned agricultural machine navigation method according to claim 3, wherein: Calculating the gradient and angle of the pattern in the surrounding environment image includes: The gradient and angle of the pattern in the surrounding environment image are calculated according to the following formula: Wherein, x is the horizontal coordinate of the pixel point of the pattern in the surrounding environment image, y is the vertical coordinate of the pixel point of the pattern in the surrounding environment image, and f is the grayscale value of the pattern in the surrounding environment image. An angle matrix is ​​calculated for all pixel points of the pattern in the surrounding environment image.

5. The unmanned agricultural machine navigation method according to claim 3, wherein: Performing non-maximum suppression on the gradient includes: Determine whether the grayscale value of the currently detected point C in the gradient is the largest in the 8-connected neighborhood. If it is the largest, continue to check whether the grayscale values ​​of the first intersection point dTmp1 and the second corner point dTmp2 in the gradient direction are greater than C. If C is greater than the grayscale values ​​of the first intersection point dTmp1 and the second corner point dTmp2, then C is considered to be a maximum value and the value of C is set to 1. Otherwise, C is considered to be a non-maximum value and the value of C is set to 0. Traverse all points C in the gradient to find the local maximum of the pixel points in the gradient, and complete the non-maximum suppression of the gradient.

6. The unmanned agricultural machine navigation method according to claim 3, wherein: Use double thresholds to connect the edges of the pattern until the complete edge contour of the pattern is extracted, including: Select two thresholds, including a low threshold and a high threshold, consider points smaller than the low threshold as false edges and set to 0, and consider points larger than the high threshold as strong edges and set to 1; According to the high threshold points in the image, they are first connected into contours. When the breakpoint of the contour is reached, the algorithm will search for a point that meets the low threshold in the 8-bit area of ​​the breakpoint, and then collect new edges based on this point until the entire image is closed.

7. An unmanned agricultural machinery navigation device, wherein: The device includes: A device for determining the geodetic coordinates of characteristic corner points, used to determine the geodetic coordinates of characteristic corner points of graphic signboards arranged on or on both sides of the driving path of the unmanned agricultural machine; A device for calculating pixel coordinates of characteristic corner points locates the geodetic coordinates of the unmanned agricultural machine while the unmanned agricultural machine is in motion, collects an image of the surrounding environment, and uses image edge detection technology to extract edge contours of patterns in the surrounding environment image, thereby identifying vertices of the patterns in the surrounding environment image, using the vertices as characteristic corner points, and calculating the pixel coordinates of the characteristic corner points; The coordinate calibration device is used to correct and calibrate the geodetic coordinates of the unmanned agricultural machine using the geodetic coordinates and pixel coordinates of the characteristic corner points; wherein the coordinates of the corrected unmanned agricultural machine are calculated according to the following formula: Among them, p j ′ is the geodetic coordinate of the feature corner point, p j is the corrected coordinate of the unmanned agricultural machine to be optimized, z ′ i j is the posture of the unmanned agricultural machine in T i Observe the characteristic corner point p at j ′ The resulting pixel measurement data, e ij is the error function, h(T i , p j ) is the projection function from world coordinates to pixel coordinates.

8. A navigation device for an unmanned agricultural machine, comprising: processor, The memory includes program instructions executable by the processor, and when the program instructions are executed by the processor, the following operations are performed: Determine the geodetic coordinates of characteristic corner points of graphic signboards set on or on both sides of the driving path of the unmanned agricultural machinery; During the driving process of the unmanned agricultural machine, the geodetic coordinates of the unmanned agricultural machine are obtained by locating the machine, an image of the surrounding environment is collected, and an edge contour of a pattern in the surrounding environment image is extracted using image edge detection technology, thereby identifying vertices of the pattern in the surrounding environment image, using the vertices as feature corner points, and calculating the pixel coordinates of the feature corner points; The geodetic coordinates and pixel coordinates of the characteristic corner points are used to correct and calibrate the geodetic coordinates of the unmanned agricultural machine; wherein the coordinates of the corrected unmanned agricultural machine are calculated according to the following formula: Among them, p j ′ is the geodetic coordinate of the feature corner point, p j is the corrected coordinate of the unmanned agricultural machine to be optimized, z ′ i j is the posture of the unmanned agricultural machine in T i Observe the characteristic corner point p at j ′ The resulting pixel measurement data, e ij is the error function, h(T i , p j ) is the projection function from world coordinates to pixel coordinates.

9. A 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 6 is implemented.

Images