Precision measuring method of self-propelled vehicle, calculation device and self-propelled vehicle

Abstract

The invention discloses a precision measuring method of a self-propelled vehicle, an arithmetic device and the self-propelled vehicle. The precision measurement method of the self-propelled vehicle comprises a distance calculation step, a regression center calculation step and an average calculation step. The distance calculation step comprises a control step, a light beam emission step, an image pickup step, an image analysis step and a conversion step. The regression center calculation step is executed after the distance calculation step is repeatedly executed for at least two times. According to the precision measuring method of the self-propelled vehicle, the X-axis offset, the Y-axis offset and the angle deflection of the self-propelled vehicle in the X-axis direction and the Y-axis direction can be measured. In the distance calculation step, the Apriltag on the marking piece is matched with the light spot on the marking piece to calculate the real distance of the X axis and the real distance of the Y axis.

Description

Technical Field

[0001] The present invention relates to a precision measurement method, a computing device capable of executing the precision measurement method, and a vehicle capable of executing the precision measurement method, and more particularly to a precision measurement method for a self-driving vehicle, a computing device, and a self-driving vehicle comprising the computing device. Background Art

[0002] Existing technology lacks equipment for measuring the motion accuracy of autonomous mobile robots (AMRs) and automated guided vehicles (AGVs) used in various factory settings. After purchasing an AMR or AGV, personnel can only learn about the vehicle's motion accuracy from the manufacturer's manuals, but are unable to verify the vehicle's actual motion accuracy. Summary of the Invention

[0003] The present invention discloses a method for measuring the accuracy of a self-driving vehicle, a computing device, and a self-driving vehicle, which are mainly used to solve the problem in the prior art that the accuracy of a self-driving vehicle cannot be measured.

[0004] One embodiment of the present invention discloses a method for measuring the accuracy of a self-driving vehicle, which can measure an X-axis offset and a Y-axis offset of a self-driving vehicle in the X-axis direction and the Y-axis direction. The method can be executed by a computing device and includes: repeatedly performing a distance calculation step a predetermined number of times, the predetermined number of times being no less than 2 times, the distance calculation step including: a control step of controlling the self-driving vehicle to move to a predetermined position; a light beam emitting step of controlling a light beam device to emit at least one light beam toward a marking member to form at least one light spot on the marking member; the marking member has an April ag or ArUco, and in the light beam emitting step, the light spot is away from the Apriltag or ArUco; wherein the light beam device is arranged around the predetermined position, and the marking member is arranged on the self-propelled vehicle, or the marking member is arranged around the predetermined position, and the light beam device is arranged on the self-propelled vehicle; an image pickup step: controlling an image pickup device to pick up the image of the marking member to form an image to be analyzed, the image to be analyzed including a label image corresponding to the Apriltag or ArUco, and a light spot image corresponding to the light spot; an image analysis step: parsing the label image in the image to be analyzed to obtain the corresponding The four corner coordinates of the Apriltag or ArUco are used to calculate the X-axis distance and the Y-axis distance of a reference point coordinate of the tag image and the center coordinate of a light spot in the light spot image; wherein the reference point coordinate is the coordinate corresponding to one of the four corner coordinates, or the reference point is the coordinate of a tag center point of the tag image calculated using the four corner coordinates; a conversion step: using the real length of one side of the Apriltag or ArUco and the straight line distance calculated using the two corner coordinates located on the same axis, convert the X The X-axis distance and the Y-axis distance are converted into an X-axis true distance and a Y-axis true distance, and a measurement coordinate is generated accordingly; the measurement coordinate includes the X-axis true distance and the Y-axis true distance; a regression center calculation step: using multiple measurement coordinates, a regression operation is performed to calculate a regression center coordinate; an average calculation step: using the multiple measurement coordinates and the regression center coordinate, the X-axis offset and the Y-axis offset of the self-propelled vehicle are calculated; the X-axis offset is the average of the distances between the regression center coordinate and each measurement coordinate in the X-axis direction, and the Y-axis offset is the average of the distances between the regression center coordinate and each measurement coordinate in the Y-axis direction.

[0005] Preferably, the color of the Apriltag or ArUco on the marking piece is composed of black or gray and white, and the length and width of the Apriltag or ArUco are not less than 10 mm.

[0006] Preferably, the total number of pixels respectively included in the length and width of the label image is at least 5 times the total number of pixels included in the diameter of the light spot image.

[0007] Preferably, in the light beam emitting step, the light spot does not overlap with the Apriltag or ArUco on the marking element.

[0008] Preferably, if the maximum acceptable X-axis offset and the maximum acceptable Y-axis offset of the self-driving vehicle in the X-axis direction and the Y-axis direction are respectively limited to an X-axis offset limit value and a Y-axis offset limit value, and when the self-driving vehicle moves to a predetermined position and there is no deviation of the self-driving vehicle in the X-axis direction and the Y-axis direction, in the light beam emitting step, the shortest straight-line distance between the April tag or ArUco on the marking element and the light spot in the X-axis direction is greater than the X-axis offset limit value, and the shortest straight-line distance between the April tag or ArUco on the marking element and the light spot in the Y-axis direction is greater than the Y-axis offset limit value.

[0009] Preferably, in the light beam emission step, the light beam device is controlled to emit two light beams to the marking member to form two non-overlapping light spots on the marking member; in the image analysis step, a deflection angle is calculated using the image to be analyzed; the deflection angle is the angle between a reference line of the label image and a line connecting the centers of the two light spot images; the reference line passes through the reference point and one of the corner coordinates that is not the reference point; in the average calculation step, an average deflection angle is also calculated, and the average deflection angle is the average of multiple deflection angles.

[0010] Preferably, the marking member is a component independent of the self-propelled vehicle, and the marking member is detachably provided on the self-propelled vehicle; or, the marking member is a part of the self-propelled vehicle.

[0011] Preferably, the image analysis step includes the following steps: a judgment step: parsing the four corner coordinates corresponding to the Apriltag or ArUco, and using the four corner coordinates to calculate the lengths of the four side lines corresponding to the Apriltag or ArUco, and judging whether the difference between the lengths of the four side lines exceeds a preset difference; if the difference between the lengths of the four side lines does not exceed the preset difference, executing a calculation step: calculating the X-axis distance and Y-axis distance in the X-axis direction and the Y-axis direction of the reference point coordinates and the center coordinates of the light spot; if the difference between the lengths of the four side lines exceeds the preset difference, first executing an image correction step, and then executing the calculation step; the image correction step is: correcting the image to be analyzed so that the difference between the lengths of the four side lines does not exceed the preset difference.

[0012] One embodiment of the present invention discloses a computing device for installation on a self-driving vehicle. The computing device can execute the method for measuring the accuracy of a self-driving vehicle according to any one of claims 1 to 8. The computing device can be connected to a processing device of the self-driving vehicle, the processing device can control the self-driving vehicle to move to a predetermined position, the computing device can be connected to an image pickup device, the computing device can control the movement of the image pickup device, and the computing device can receive images to be analyzed transmitted by the image pickup device.

[0013] Preferably, the computing device can be connected to an external electronic device, and the computing device can obtain the actual length and the predetermined number of times from the external electronic device, or the computing device can receive a modification message transmitted by the external electronic device to modify at least one of the pre-stored actual length and the predetermined number of times.

[0014] Preferably, after the computing device performs the averaging step, the computing device further includes an output step of outputting the X-axis offset and the Y-axis offset to an external electronic device.

[0015] Preferably, the computing device can control the light beam device so that the light beam device emits light beams of different wavelengths to form light spots of different colors on the marking element.

[0016] One embodiment of the present invention discloses a self-propelled vehicle, comprising: a computing device and a processing device according to claim 9; a detection device for detecting the environment surrounding the self-propelled vehicle to generate detection information; the detection device capable of transmitting the detection information to the processing device; and a driving device electrically connected to the processing device. The processing device is capable of controlling the operation of the driving device based on the detection information to move the self-propelled vehicle to a predetermined position.

[0017] Preferably, the self-propelled vehicle further comprises a marking element, a light beam device and an image pickup device.

[0018] Preferably, after executing the accuracy measurement method for the self-driving vehicle, the computing device stores the X-axis offset and the Y-axis offset as an X-axis correction value and a Y-axis correction value for the self-driving vehicle. After the computing device stores the X-axis correction value and the Y-axis correction value, when the computing device receives movement information, the computing device controls the self-driving vehicle to move to a specified position corresponding to the movement information based on the movement information, the X-axis correction value, and the Y-axis correction value.

[0019] Preferably, the computing device can switch between a mobile positioning mode and a calibration mode. When the computing device executes the mobile positioning mode, the computing device controls the self-driving vehicle to move to a predetermined position based on the movement information, the X-axis correction amount, and the Y-axis correction amount, and controls the image pickup device to pick up the Apriltag or ArUco around the predetermined position to obtain corresponding identification data; when the computing device determines that the identification data obtained after the self-driving vehicle moves to the predetermined position is the same as the default identification data included in the movement information, the computing device determines that the self-driving vehicle has moved to the specified position; when the computing device is in the calibration mode, the computing device will execute the accuracy measurement method of the self-driving vehicle.

[0020] Preferably, when the computing device receives movement information, the computing device will first determine whether the self-driving car has stored therein the X-axis correction amount and the Y-axis correction amount. If the computing device determines that the self-driving car does not have stored therein the X-axis correction amount and the Y-axis correction amount, the computing device can issue a calibration warning message. If the computing device receives a calibration request message after issuing the calibration warning message, the computing device will execute the accuracy measurement method for the self-driving car and store the X-axis offset and Y-axis offset obtained after executing the accuracy measurement method as the X-axis correction amount and Y-axis correction amount of the self-driving car.

[0021] Preferably, the computing device can control the light beam device so that the light beam device emits light beams of different wavelengths to form light spots of different colors on the marking element.

[0022] In summary, the self-driving vehicle accuracy measurement method, computing device, and self-driving vehicle of the present invention can be used to measure the X-axis offset and Y-axis offset of the self-driving vehicle in the X-axis and Y-axis directions, thereby allowing relevant personnel to know the accuracy of the self-driving vehicle.

[0023] To further understand the features and technical contents of the present invention, please refer to the following detailed description and drawings of the present invention. However, such description and drawings are only used to illustrate the present invention and are not intended to limit the scope of protection of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 Schematic diagram of a self-propelled vehicle of the present invention.

[0025] Figure 2 FIG. 4 is a flow chart of a first embodiment of a method for measuring the accuracy of a self-driving vehicle according to the present invention.

[0026] Figure 3 It is a top view schematic diagram of the self-propelled vehicle of the present invention when it moves to a predetermined position.

[0027] Figure 4 Schematic diagram of an image to be analyzed according to the first embodiment of the method for measuring the accuracy of a self-driving vehicle of the present invention.

[0028] Figure 5 It is a top view schematic diagram of the self-propelled vehicle of the present invention when it moves to a predetermined position and has no deviation in the X-axis and Y-axis directions.

[0029] Figure 6 FIG. 4 is a flow chart of a second embodiment of the method for measuring the accuracy of a self-driving vehicle according to the present invention.

[0030] Figure 7 FIG. 4 is a flow chart of a third embodiment of the method for measuring the accuracy of a self-driving vehicle according to the present invention.

[0031] Figure 8 1 is a top view of a third embodiment of the computing device, the self-driving vehicle, the light beam device, and the image pickup device of the present invention.

[0032] Figure 9 FIG. 1 is a schematic diagram of an image to be analyzed according to a third embodiment of the method for measuring the accuracy of a self-driving vehicle of the present invention. DETAILED DESCRIPTION

[0033] In the following description, if it is indicated to refer to a specific figure or as shown in a specific figure, it is only used to emphasize that most of the related content described in the subsequent description appears in the specific figure, but it does not limit the subsequent description to only refer to the specific figure.

[0034] Please refer to Figure 1 , which shows a schematic diagram of the self-propelled vehicle of the present invention. The self-propelled vehicle A of the present invention comprises a marking part 1, a vehicle body A1, a computing device A2, a processing device A3, a detection device A4 and a driving device A5. The computing device A2, the processing device A3, the detection device A4 and the driving device A5 are arranged on the vehicle body A1. The computing device A2 can execute the precision measurement method of the self-propelled vehicle of the present invention, and the computing device A2 is electrically connected to the processing device A3. The computing device A2 is, for example, various industrial computers, computer devices including various processing chips, etc., which are not limited here. The self-propelled vehicle A of the present invention can be, for example, a self-propelled vehicle with an autonomous path planning function applied as an AMR (Autonomous Mobile Robot), an AGV (Automated Guided Vehicle), a sweeping robot, an autonomous driving car, etc.

[0035] Detection device A4 is used to detect the environment surrounding autonomous vehicle A and generate detection information, which is then transmitted to processing device A3. Drive device A5 is electrically connected to processing device A3. Based on the detection information, processing device A3 controls the operation of drive device A5 to move autonomous vehicle A to a predetermined location. To facilitate understanding, a predetermined location is indicated by a frame B in the figures of this embodiment. The specific operation of processing device A3, detection device A4, and drive device A5 is similar to that of conventional AMRs (Autonomous Mobile Robots) and will not be further described here.

[0036] When performing precision measurement on self-driving vehicle A, personnel may, for example, first install a light beam device 2 around a predetermined location, install an image pickup device 3 on the upper surface of self-driving vehicle A, and place a marking member 1 on the upper surface of self-driving vehicle A. The light beam device 2 may, for example, include a support frame 21 and a light beam generator 22 , with the light beam generator 22 being installed around the predetermined location via the support frame 21 . The image pickup device 3 may, for example, include an image pickup device 31 and a support frame 32 , with the image pickup device 31 being installed on self-driving vehicle A via the support frame 32 .

[0037] The computing device A2 can be connected to the image pickup device 3 via a wireless or wired connection. The computing device A2 can control the operation of the image pickup device 3 so that the image pickup device 3 captures an image of the marking element 1 mounted on the self-propelled vehicle A. The computing device A2 can also receive the image 33 to be analyzed transmitted by the image pickup device 3. In various embodiments, the computing device A2 can also control the activation and deactivation of the light beam device 2.

[0038] It is worth noting that, in various embodiments, the computing device A2 of the present invention may be sold, implemented, and manufactured independently, and is not necessarily sold or manufactured together with the autonomous vehicle A. In one embodiment where the computing device A2 is sold independently, a purchaser may install the computing device A2 in an autonomous vehicle (e.g., an AMR, AVG, etc.) and connect the computing device A2 to the processing device A3 in the autonomous vehicle.

[0039] In one embodiment where the computing device A2 is sold independently, the computing device A2 may serve as a remote server, or the computing device A2 may serve as a cloud server in a factory. In one embodiment where the computing device A2 is sold independently, the computing device A2 may be integrated with the light beam device 2 or the image pickup device 3 into the same housing.

[0040] Please refer to Figures 1 to 4, Figure 2 The figure shows a flow chart of the precision measurement method of the self-driving vehicle of the present invention. Figure 3 Schematic diagram of a top view of the self-propelled vehicle of the present invention when it moves to a predetermined position. Figure 4 Schematic diagram of an image to be analyzed according to the first embodiment of the method for measuring the accuracy of a self-driving vehicle of the present invention.

[0041] The accuracy measurement method of a self-driving vehicle of the present invention can measure an X-axis offset and a Y-axis offset of the self-driving vehicle in the X-axis direction and the Y-axis direction. The accuracy measurement method of the self-driving vehicle includes:

[0042] The distance calculation step S1 is first repeated a predetermined number of times, followed by a center return calculation step S2 and an average calculation step S3. The distance calculation step S1 includes a control step S11, a light beam emission step S12, an image pickup step S13, an image analysis step S14, and a conversion step S15.

[0043] The control step S11 is to control the self-propelled vehicle A to move to a predetermined position (such as Figure 3 The marked box B);

[0044] The light beam emitting step S12 includes controlling the light beam device 2 to emit at least one light beam toward the marking element 1 to form at least one light spot 12 on the marking element 1. The marking element 1 has an April tag 4, and in the light beam emitting step S12, the light spot 12 is away from the April tag 4.

[0045] The image pickup step S13 is to control the image pickup device 3 to pick up the image of the label 1 to form an image 33 to be analyzed. The image 33 to be analyzed includes a label image 331 corresponding to the April tag 4 and a light spot image 332 corresponding to the light spot 12.

[0046] The image analysis step S14 is as follows: parsing the tag image 331 in the image to be analyzed 33 to obtain the four corner coordinates corresponding to Apriltag 4, and calculating the reference point coordinates of a reference point of the tag image 331 and the center point coordinates of a light spot of the light spot image 332, an X-axis distance Δx and a Y-axis distance Δy in the X-axis direction and the Y-axis direction.

[0047] At Figure 4 In the example, the upper left corner point 331A of the label image 331 is used as the reference point, but the reference point is not limited to this. The reference point coordinates can be the coordinates corresponding to any one of the four corner points 331A, 331B, 331C, and 331D of the label image 331, or the reference point can also be the coordinates of a label center point of the label image 331 calculated using the four corner coordinates (i.e., Figure 4 The coordinates corresponding to the label center point 331E in ).

[0048] The conversion step S15 is to convert the X-axis distance and the Y-axis distance into an X-axis real distance and a Y-axis real distance using a real length of one side of the April tag 4 and a straight line distance D calculated using the coordinates of two corners located on the same axis, and generate a measurement coordinate based on the converted X-axis real distance and the Y-axis real distance. The measurement coordinate includes the X-axis real distance and the Y-axis real distance, and the measurement coordinate is generated based on the converted X-axis real distance and the Y-axis real distance. That is, the measurement coordinate is (Δx, Δy).

[0049] In the above description, Figure 4 The linear distance D is calculated using two corner points 331A and 331B in the image to be analyzed, but this is not limiting. In different embodiments, the corresponding linear distances may be calculated using two corner points 331A and 331C, or two corner points 331B and 331D, or two corner points 331C and 331D. The actual length of Apriltag 4 in the conversion step S15 may be pre-stored in the computing device A2 or the processing device A3. In the conversion step S15, the linear distance D in the image to be analyzed and its corresponding actual length are primarily used to convert the X-axis distance and the Y-axis distance, thereby obtaining the X-axis actual distance and the Y-axis actual distance.

[0050] It should be particularly emphasized that the above-mentioned image analysis step S14 and conversion step S15 are mainly used to convert the X-axis distance and the Y-axis distance into the real X-axis distance and the real Y-axis distance. Therefore, after referring to the content of the above-mentioned conversion step S15, relevant technical personnel can easily think of taking the length of any two points of Apriltag 4 (for example, the straight-line distance from the center point of Apriltag 4 to one of the corners) as the real length according to actual needs, and use the four corner coordinates to calculate the length corresponding to the real length in the image to be analyzed.

[0051] In actual applications, the image length D of the image 33 to be analyzed may be different from the actual length due to various factors (such as the placement position of the image pickup device 3 relative to the marking element 1 and the magnification of the image pickup device 3 when picking up the marking element 1). Therefore, after the image analysis step S14, the design of the conversion step S15 can effectively improve or even solve the above problem, thereby obtaining the actual X-axis distance and Y-axis distance that are close to the actual value.

[0052] The regression center calculation step S2 is to use multiple measured coordinates to calculate a regression center coordinate through a regression operation. In practical applications, the regression operation can be, for example, to sum and average the X-axis coordinates of all measured coordinates to calculate the X-axis coordinate of the regression center coordinate; similarly, the regression operation can be, for example, to sum and average the Y-axis coordinates of all measured coordinates to calculate the Y-axis coordinate of the regression center coordinate.

[0053] The average calculation step S3 is as follows: using multiple measured coordinates and the regression center coordinates, the X-axis offset and Y-axis offset of the self-driving vehicle A are calculated; the X-axis offset is the average of the distances in the X-axis direction between the regression center coordinate and each measured coordinate, and the Y-axis offset is the average of the distances in the Y-axis direction between the regression center coordinate and each measured coordinate.

[0054] It should be noted that in order to ensure that the X-axis distance Δx and Y-axis distance Δy calculated each time have reference value, each time the distance calculation step S1 is executed, the reference point image corresponding to the same reference point in the reference pattern in the image to be analyzed 33 must be used together with the light point image 332 to calculate the X-axis distance Δx and Y-axis distance Δy. In this way, it can be ensured that the X-axis distance Δx and Y-axis distance Δy calculated after each execution of the distance calculation step S1 have reference value.

[0055] The predetermined position in each execution of the control step S11 can be the same or different according to the requirements. Of course, if the predetermined position in each execution of the control step S11 is different, it is necessary to set the light beam device 2 or the image pickup device 3 at each predetermined position.

[0056] The predetermined number of times is no less than two, but is not limited thereto. In an embodiment where autonomous vehicle A is employed as an AMR (Autonomous Mobile Robot), the parking position of autonomous vehicle A may vary slightly each time it moves to the same predetermined location due to various factors. Therefore, by repeating distance calculation step S1 a predetermined number of times, and then performing regression center calculation step S2 and average calculation step S3, the resulting X-axis offset and Y-axis offset values can more closely approximate the actual offset values of autonomous vehicle A in the X-axis and Y-axis directions each time it moves to the predetermined location. In the example of the autonomous vehicle accuracy measurement method of the present invention being used in accordance with a general AMR testing standard (e.g., AMRA-201 Mobile Robot - General Requirements and Testing Methods), the predetermined number of times is no less than 30. This ensures that the resulting X-axis offset and Y-axis offset values sufficiently represent the actual offset values of the AMR in the X-axis and Y-axis directions.

[0057] In practical applications, the light beam device 2 may be, for example, a laser device capable of emitting at least one laser beam, but is not limited thereto. As long as the light beam device 2 can form a light spot image 332 on the marking element 1 that is recognizable by common image recognition software (e.g., image recognition software using the OpenCV database), the light beam device 2 may be capable of emitting various light beams.

[0058] In this embodiment, the light beam device 2 is set around the predetermined position and the marking member 1 is set on the self-propelled vehicle A for example, but the present invention is not limited to this. In different embodiments, the light beam device 2 can also be set on the self-propelled vehicle A, and the marking member 1 can be set around the predetermined position, and the image pickup device 3 can be set around the self-propelled vehicle A or the predetermined position, as long as the image pickup device 3 can correctly pick up the image of the marking member 1 (including the April tag and the light spot formed on the marking member 1 by the light beam device 2).

[0059] In an embodiment in which the light beam device 2 is disposed on the self-propelled vehicle A and the marking element 1 is disposed around a predetermined position, the image pickup device 3 may be an electronic device possessed by the self-propelled vehicle A itself; that is, if the image pickup device possessed by the self-propelled vehicle A itself can pick up the April tag on the marking element 1 located around the predetermined position and the light spot formed on the marking element 1 by the light beam device 2, then there is no need to additionally provide other image pickup devices on the self-propelled vehicle A.

[0060] The marking member 1 may be, for example, a component independent of the self-driving vehicle A, and the marking member 1 may be detachably arranged on the self-driving vehicle A. For example, the marking member 1 may be, for example, a sticker, and one side of the marking member 1 may be printed with Apriltag 4, and the Apriltag 4 is composed of black (or gray) and white, and the other side of the marking member 1 may have a release film. After the user tears off the release film, the marking member 1 can be attached to the upper surface of the self-driving vehicle A through the adhesive layer on the marking member 1. The adhesive layer may be, for example, various adhesives, preferably, reusable adhesives. The specific form of the marking member 1 is not limited to the above-mentioned sticker form. In different embodiments, a magnetic layer may also be provided on one side of the marking member 1, and the marking member 1 can be adsorbed on the upper surface of the self-driving vehicle A using the magnetic layer. It should be noted that in the drawings of this embodiment, the April tag TAG36H11 is taken as an example, but in actual application, any family of April tags, such as TAG16H5, TAG25H7, TAG25H9, TAG36H10, etc., can be selected according to actual needs.

[0061] In practice, in the image analysis step S14, the computing device A2 may, for example, run a tag recognition program. The tag recognition program may be written in a programming language such as Python. The tag recognition program may, for example, read / input the image to be analyzed and call a corresponding function in the Apriltag library in OpenCV to identify the Apriltag in the image to be analyzed. If the Apriltag in the image to be analyzed is successfully recognized, the tag recognition program will obtain the coordinates of the four corners corresponding to the Apriltag (and the coordinates of the tag's center point). The Apriltag library and its functions specifically called by the tag recognition program are conventional technology and will not be further described here.

[0062] In the image analysis step S14, after executing the label recognition process, an image processing process may be executed to perform at least one image processing process on the image to be analyzed, thereby obtaining the center point coordinates corresponding to the light spot image. For example, the image processing process may first perform binarization processing on the image to be analyzed using the color of the light spot on the marking element to generate a first processed image. Erosion and dilation processing and edge detection processing are then performed on the first processed image to identify the edges of the light spot in the first processed image. Finally, the findContours() method in the OpenCV vision library is used to obtain the center point coordinates of the light spot image located in the first processed image. Of course, the method for performing image analysis on the image to be analyzed to obtain the center point coordinates corresponding to the light spot image is not limited to the above description.

[0063] Continuing from the above, the computing device A2 executes the above-mentioned label recognition program and image processing program, and after obtaining the four corner coordinates and the coordinates of the center point of the light spot, it can subsequently execute a calculation program to use one of the corner coordinates (or the label center coordinates) as a reference point to calculate the X-axis distance and Y-axis distance between the reference point coordinates and the light spot center coordinates in the X-axis direction and Y-axis direction respectively.

[0064] It is worth noting that in preferred applications, during the light beam emission step S12, the light spot does not overlap with the April tag 4 on the marking element 1. This effectively avoids the tag recognition process from being unable to correctly identify the April tag or correctly resolve the coordinates of the light spot center corresponding to the light spot image during the image analysis step S14. Of course, in embodiments where the light spot overlaps with the April tag 4 on the marking element 1, the brightness of the light spot can be adjusted to ensure that the four corner coordinates and the light spot center coordinate can be effectively resolved during the image analysis step S14.

[0065] Please refer to Figure 2 and Figure 5 , Figure 5 The diagram shows a top view of the self-driving vehicle of the present invention when it moves to a predetermined position and the self-driving vehicle has no deviation in the X-axis and Y-axis directions. If the maximum acceptable X-axis offset and the maximum acceptable Y-axis offset of the self-driving vehicle A in the X-axis and Y-axis directions are respectively limited to an X-axis offset limit value and a Y-axis offset limit value, and when the self-driving vehicle A moves to the predetermined position and the self-driving vehicle A has no deviation in the X-axis and Y-axis directions (e.g. Figure 5 As shown), in the light beam emission step S12, the shortest straight-line distance Δx1 between the April tag 4 on the marking element 1 and the light spot 12 in the X-axis direction is greater than the X-axis offset limit value LX, and the shortest straight-line distance Δy1 between the April tag 4 on the marking element 1 and the light spot 12 in the Y-axis direction is greater than the Y-axis offset limit value LY.

[0066] In other words, in practice, when determining the position of the Apriltag 4 on the marking element 1, the light beam device 2 can be first placed around the default position, and the self-propelled vehicle A equipped with the marking element 1 can be placed at the predetermined position, and the self-propelled vehicle A located at the predetermined position is not deviated in the X-axis direction and the Y-axis direction. Then, the light beam device 2 is used to form a light spot 12 on the marking element 1. Then, the light spot 12 is used in conjunction with the X-axis offset limit value LX and the Y-axis offset limit value LY to draw a clearance area E on the marking element 1. Finally, the Apriltag 4 is placed outside the clearance area E. For example, if the X-axis offset limit value LX and the Y-axis offset limit value LY are ±3 and ±5 cm respectively, the clearance area E can be a 6*10 cm area, the light spot is located at the center of the clearance area E, and the April tag 4 is set outside the clearance area E. This design ensures that when the light beam emission step S12 is subsequently executed, the light spot 12 will not be formed on the April tag 4, that is, the light spot 12 does not overlap with the April tag 4.

[0067] As described above, the self-driving vehicle accuracy measurement method of the present invention, the computing device capable of executing the self-driving vehicle accuracy measurement method of the present invention, and the self-driving vehicle incorporating the computing device of the present invention can calculate the X-axis offset and Y-axis offset of the self-driving vehicle when it moves to a predetermined position. This allows interested parties to determine whether the X-axis offset and Y-axis offset of their purchased self-driving vehicle when it moves to a predetermined position are within the accuracy error range claimed by the manufacturer. Furthermore, by understanding the X-axis offset and Y-axis offset of each self-driving vehicle, interested parties can deploy the vehicle in appropriate locations, significantly reducing the risk of collisions caused by the vehicle's own offset within the location.

[0068] For example, if the self-driving car A moves to the predetermined position, it will be Figure 3 In the state, the X-axis offset calculated by computing device A2 is -5 cm and the Y-axis offset is +5 cm. Moreover, the manual of the self-driving vehicle A states that its accuracy error in the X-axis and Y-axis directions is ±1 cm. Therefore, the relevant personnel can know that the self-driving vehicle A does not meet the accuracy error of ±1 cm claimed by the industry.

[0069] In the prior art, there are no tools, devices, or methods for measuring the accuracy of self-driving vehicles (i.e., X-axis offset and Y-axis offset). Consequently, personnel are forced to rely solely on the accuracy claims of manufacturers. Only when the self-driving vehicle collides in a field (e.g., a factory) do they discover that the vehicle's accuracy may differ from the manufacturer's claims, causing significant frustration. The present invention's method for measuring the accuracy of a self-driving vehicle, a computing device capable of executing the method, and a self-driving vehicle incorporating the computing device are designed to alleviate this problem.

[0070] Please refer back Figure 1 It is worth mentioning that in one embodiment, the computing device A2 can also be connected to an external electronic device C (for example, via various wireless methods), and the computing device A2 can obtain the actual length and the predetermined number of times from the external electronic device C. Alternatively, the computing device A2 can receive a modification message transmitted by the external electronic device C to modify at least one of the pre-stored actual length and the predetermined number of times. For example, the external electronic device C can be a smartphone, tablet computer, laptop computer, desktop computer, remote server, etc. When the external electronic device C runs a predetermined application and is connected to the computing device A2, the external electronic device C can send the actual length, the predetermined number of times, or the modification message to the computing device A2.

[0071] Continuing from the above, preferably, after the computing device A2 performs the aforementioned average calculation step S3, it may further include an output step: outputting the X-axis offset and the Y-axis offset to the external electronic device C. In other words, after the computing device A2 executes the self-driving vehicle accuracy measurement method of the present invention, it transmits the final calculated X-axis offset and Y-axis offset to the external electronic device C. Relevant personnel can then view the external electronic device C to obtain the self-driving vehicle's X-axis offset and Y-axis offset.

[0072] Please refer back Figure 1 In actual applications, the computing device A2 is electrically connected to the processing device A3, and the computing device A2 may, for example, send a relevant inquiry signal to the processing device A3 to know that the self-driving vehicle A has currently moved to the predetermined position; or, when the processing device A3 moves to the computing device A2, the processing device A3 may automatically transmit a relevant signal to the computing device A2 to notify the computing device A2. After the computing device A2 knows that the self-driving vehicle A has currently moved to the predetermined position, the computing device A2 may start to execute the light beam emission step and other related steps.

[0073] In an embodiment where the computing device A2 is independent of the self-driving vehicle A, relevant personnel can start the computing device A2, the light beam device 2, and the image pickup device 3 when the self-driving vehicle A moves to a predetermined position to execute the aforementioned precision measurement method of the self-driving vehicle of the present invention.

[0074] In summary, the self-driving vehicle precision measurement method of the present invention, the computing device capable of executing the self-driving vehicle precision measurement method of the present invention, and the design of the self-driving vehicle incorporating the computing device of the present invention allow relevant personnel to clearly understand the offset of the self-driving vehicle in the X-axis and Y-axis directions when the self-driving vehicle moves to a predetermined position. This allows relevant personnel to better plan the movement range of the self-driving vehicle in the field, thereby significantly reducing the probability of collision when the self-driving vehicle moves independently in the field.

[0075] Furthermore, the present invention's method for measuring the accuracy of a self-driving vehicle, a computing device capable of executing the method, and a self-driving vehicle design incorporating the computing device can assist researchers in calibrating the vehicle's software and hardware, enabling the vehicle to more accurately move to a predetermined location. Prior art self-driving vehicle developers had to rely on manual methods to determine whether the vehicle had correctly reached its predetermined location during the development of its autonomous driving software and hardware, which caused significant frustration.

[0076] Please refer back Figure 1It is worth noting that in one embodiment of the self-driving vehicle of the present invention, after executing the self-driving vehicle accuracy measurement method, computing device A2 may store the X-axis offset and the Y-axis offset in a storage device (e.g., various types of memory) associated with self-driving vehicle A as an X-axis correction value and a Y-axis correction value for the self-driving vehicle. After computing device A2 stores the X-axis correction value and the Y-axis correction value, if computing device A2 receives movement information transmitted by an external electronic device C, computing device A2 may control self-driving vehicle A to move to a specified position corresponding to the movement information based on the movement information, the X-axis correction value, and the Y-axis correction value.

[0077] Continuing from the above, in actual applications, computing device A2 can, for example, switch between a mobile positioning mode and a calibration mode. When computing device A2 is in mobile positioning mode, computing device A2 controls autonomous vehicle A to move to a predetermined location based on the movement information, the X-axis calibration amount, and the Y-axis calibration amount. It also controls image pickup device 3 to capture the April tag 4 near the predetermined location to obtain corresponding identification data (e.g., the April tag ID). If the computing device determines that the identification data obtained after the autonomous vehicle has moved to the predetermined location is the same as the default identification data (e.g., the April tag ID) included in the movement information, computing device A2 determines that autonomous vehicle A has moved to the designated location.

[0078] Continuing with the above description, for example, if the X-axis correction amount and the Y-axis correction amount stored in computing device A2 are +1.5 cm and -1.2 cm, respectively, and the movement information received by computing device A2 includes the coordinates of the designated location and the ID of the April tag set around the designated location, which are (10, 20) and 0, respectively, then when computing device A2 controls autonomous vehicle A to move to the designated location based on the movement information, computing device A2 actually controls autonomous vehicle A to move to the coordinates (11.5, 18.8). After computing device A2 controls autonomous vehicle A to move to the coordinates (11.5, 18.8), computing device A2 controls the image pickup device to pick up April tag 4 located around the designated location and determines whether the ID corresponding to the April tag is 0. If the ID is 0, computing device A2 determines that autonomous vehicle A has moved to the designated location corresponding to the movement information.

[0079] It should be emphasized that the aforementioned use of an image pickup device to capture an April tag image and parse the April tag's corresponding ID to determine whether the self-driving vehicle has moved to a designated location is a common application of April tags. This application is clearly different from the use of April tags in the self-driving vehicle accuracy measurement method described above. In other words, existing technologies related to self-driving vehicle movement only utilize the April tag's ID (i.e., the identification data) to determine whether the self-driving vehicle has reached a designated location. These technologies do not utilize the corner coordinates in the April tag to calculate the vehicle's offset.

[0080] Furthermore, the present invention's self-driving vehicle accuracy measurement method utilizes AprilTags in conjunction with a light beam device to calculate the vehicle's X- and Y-axis offsets. However, the present invention's self-driving vehicle accuracy measurement method cannot simply utilize AprilTags to calculate X- and Y-axis offsets. Therefore, even if a skilled artisan were to consult conventional techniques regarding the use of AprilTags for self-driving vehicle positioning, they would not be able to readily conceive of the present invention's accuracy measurement method.

[0081] In one embodiment of the present invention, when computing device A2 receives movement information, computing device A2 will first determine whether the self-driving car A has stored X-axis and Y-axis calibration values. If computing device A2 determines that the self-driving car A does not have stored X-axis and Y-axis calibration values, computing device A2 can issue a calibration warning message. In practice, computing device A2 transmits the calibration warning message to an external electronic device C, for example, via various wireless communication methods, and the relevant personnel can view relevant prompt text such as "Self-driving car number 00123, has not been calibrated yet, please confirm whether to perform the calibration operation first" through external electronic device C. When the relevant personnel know through external electronic device C that self-driving car A has not been calibrated, the relevant personnel can, for example, transmit a calibration request message to computing device A2 through external electronic device C. When the computing device A2 receives the calibration request information, the computing device A2 executes the self-driving vehicle precision measurement method and stores the X-axis offset and Y-axis offset obtained after executing the self-driving vehicle precision measurement method as the X-axis calibration value and Y-axis calibration value of the self-driving vehicle.

[0082] More specifically, before self-driving car A leaves the factory, its computing device or processing unit may first execute the self-driving car accuracy measurement method to confirm that the X-axis offset and Y-axis offset of self-driving car A meet factory requirements. When self-driving car A leaves the factory, no X-axis and Y-axis calibration values may be stored in self-driving car A, or the X-axis and Y-axis calibration values may be stored as zero. When a person purchases self-driving car A and starts it for the first time, attempting to control it to move to a predetermined location in the factory, the person will see a calibration warning message on the external electronic device C because the X-axis and Y-axis calibration values are not stored in self-driving car A, or the X-axis and Y-axis calibration values are zero. The person can then use the external electronic device C to issue a calibration request message, causing the computing device to automatically execute the self-driving car accuracy measurement method, thereby completing the calibration of self-driving car A at the purchaser's site.

[0083] In other words, although self-driving car A is calibrated for offset in the X-axis and Y-axis directions before leaving the factory, after self-driving car A is purchased by a relevant person, self-driving car A may still experience offset in the X-axis or Y-axis directions due to various factors (such as different floor materials at the site). Therefore, in one practical application, after self-driving car A is purchased, the aforementioned calibration process can be performed again (i.e., a computing device or processing device executes the self-driving car accuracy measurement method of the present invention) so that the corresponding X-axis and Y-axis correction values are stored in self-driving car A.

[0084] As described above, before self-driving vehicle A leaves the factory, its computing device or processing unit has already executed the self-driving vehicle accuracy measurement method, and self-driving vehicle A has already completed the offset correction for the X-axis and Y-axis directions. Therefore, in various embodiments, after the relevant personnel purchase self-driving vehicle A, the computing device can directly control the self-driving vehicle to move to a specified location based on the movement information.

[0085] Please refer to Figure 6 , which is a flow chart of a second embodiment of the method for measuring the accuracy of a self-driving vehicle according to the present invention. The biggest difference between this embodiment and the previous embodiment is that the image analysis step S14 includes the following steps:

[0086] A determination step S141: parsing the coordinates of the four corners corresponding to the April tag, calculating the lengths of the four sides corresponding to the April tag using the four corner coordinates, and determining whether the difference between the lengths of the four sides exceeds a predetermined difference;

[0087] If the differences between the lengths of the four side lines do not exceed the preset difference, a calculation step S142 is performed to calculate the X-axis distance and Y-axis distance of the reference point coordinates and the light spot center coordinates in the X-axis direction and the Y-axis direction.

[0088] If the difference between the lengths of the four edges exceeds a predetermined value, an image correction step S14X is first performed, followed by calculation step S142. Image correction step S14X corrects the image to be analyzed so that the difference between the lengths of the four edges does not exceed the predetermined value. In practical applications, image correction step S14X can utilize OpenCV functions such as warpPerspective() and getPerspectiveTransform() to correct the image to be analyzed. Of course, in practice, any function capable of performing keystone correction on an image can be used.

[0089] Specifically, in determination step S141, if the difference between the lengths of the four side lines exceeds a preset difference, it indicates that the image to be analyzed may have undergone trapezoidal deformation due to various factors. Therefore, if the deformed image to be analyzed is directly used for subsequent calculations, errors may occur. Therefore, if the difference between the lengths of the four side lines exceeds the preset difference, the image correction step S14X is first performed before executing the calculation step S142 to ensure the accuracy of the ultimately calculated X-axis distance and Y-axis distance.

[0090] Please refer to Figures 7 to 9 , Figure 7 FIG. 1 is a flow chart of a third embodiment of the method for measuring the accuracy of a self-driving vehicle according to the present invention. Figure 8 FIG2 is a top view of a third embodiment of a computing device, a self-propelled vehicle, a light beam device, and an image pickup device according to the present invention. Figure 9 FIG. 1 is a schematic diagram of an image to be analyzed according to a third embodiment of the method for measuring the accuracy of a self-driving vehicle of the present invention.

[0091] The major difference between the light beam emitting step S12A of this embodiment and the light beam emitting step S12 of the aforementioned embodiment is that in the light beam emitting step S12A, the light beam device 2 is controlled to emit two light beams toward the marking element 1, thereby forming two non-overlapping light spots 12 on the marking element 1. The image analysis step S14A of this embodiment differs from the image analysis step S14 of the aforementioned embodiment in that the image analysis step S14A of this embodiment further analyzes the image to be analyzed 35 to calculate a deflection angle θ.

[0092] The deflection angle θ is the angle between a reference line L1 of the label image 351 and a line L2 connecting the centers of the two light spot images 352 and 353. Reference line L1 passes through corner point 351D, which serves as a reference point, and reference line L2 also passes through one of the corner points 351C, which is not a reference point. Of course, in different embodiments, reference line L1 may alternatively pass through corner points 351A and 351B, or through corner points 351A and 351C, or through corner points 351B and 351D.

[0093] In contrast, the average calculation step S3A of this embodiment differs from the average calculation step S3 of the aforementioned embodiment in that, in the average calculation step S3A of this embodiment, multiple deflection angles θ are further used to calculate an average deflection angle, which is the average of the multiple deflection angles θ.

[0094] It should be noted that in each image analysis step S14A, the X-axis distance Δx and Y-axis distance Δy are calculated using the same light spot image 353 and the same corner coordinates or label center coordinates corresponding to the label image 351. In this embodiment, the light beam device 2 forms two identical light spots 12 on the marking element 1. However, in different embodiments, the light beam device 2 may form two light spots 12 of different shapes on the marking element 1.

[0095] As described above, the accuracy measurement method of the self-driving vehicle of this embodiment can not only calculate the X-axis offset and the Y-axis offset of the self-driving vehicle, but also calculate the average deflection angle of the self-driving vehicle.

[0096] During the process of inventing the present invention, the applicant attempted to use a light beam device to form a light spot on a marking member, and to use an image pickup device to pick up an image containing the light spot and a colored reference point located on the marking member to form an image to be analyzed. Finally, the image to be analyzed was analyzed through a related program, and the center of the analyzed light spot image and the center of the reference point image were used to calculate the offset of the self-driving vehicle in the X-axis or Y-axis direction. Although this method can also calculate the offset of the self-driving vehicle in the X-axis or Y-axis direction, in practice, the applicant found that when performing image analysis, at least one of the light spot image and the reference point image in the image to be analyzed often cannot be correctly parsed. If either the light spot image or the reference point image is not correctly parsed, the offset of the self-driving vehicle in the X-axis or Y-axis direction cannot be calculated subsequently.

[0097] The applicant has carefully examined the causes of the above problems and found that they may be caused by the fact that when the image pickup device picks up the marking piece, the reference point is affected by the halo of the light spot, the reference point is affected by the ambient light, and the light spot is affected by the ambient light. As a result, the outline of the reference point or the outline of the light spot in the image to be analyzed cannot be correctly parsed.

[0098] Continuing from the above, the applicant has also discovered that, since the image to be analyzed is pre-processed during analysis, such as binarization, if the colored reference points are affected by the laser light, the image to be analyzed may completely disappear after the binarization. If the images corresponding to the colored reference points in the image to be analyzed completely disappear after the binarization, it will be impossible to calculate the offset of the autonomous vehicle in the X-axis or Y-axis direction in subsequent steps.

[0099] Furthermore, in practice, the reference points on the marking parts are prone to damage or contamination due to various factors. In such cases, the relevant program may not be able to correctly resolve the reference points when analyzing the image to be analyzed. Ultimately, it may be impossible to calculate the offset of the autonomous vehicle in the X-axis or Y-axis direction.

[0100] In contrast, the self-driving vehicle precision measurement method of the present invention utilizes the AprilTag and the light beam emitted by the light beam device to calculate the vehicle's X-axis and Y-axis offsets. Therefore, the self-driving vehicle precision measurement method of the present invention can significantly improve the aforementioned issues arising from colored reference points. Specifically, the AprilTag is essentially composed of black (or gray) and white. Therefore, during the image analysis step, the self-driving vehicle precision measurement method of the present invention only needs to ensure that the corresponding light point center coordinates in the image to be analyzed can be resolved and that the AprilTag can be properly read. The aforementioned issue of reference point or light point disappearance is essentially avoided in the self-driving vehicle precision measurement method of the present invention. Even if the self-driving vehicle precision measurement method of the present invention encounters the issue of light point disappearance during the image analysis step, personnel can quickly resolve the issue by changing the color of the light beam emitted by the light beam device.

[0101] As described above, in a preferred embodiment of the computing device and self-driving vehicle of the present invention, computing device A2 can be capable of controlling light beam device 2 to emit light beams of different wavelengths, thereby forming light spots 12 of different colors on marking element 1. With this design, when computing device A2 executes the precision measurement method for a self-driving vehicle of the present invention and the light spot disappears during the image analysis step, a person can use an input device (e.g., a touch panel, mouse, keyboard, etc. connected to an external electronic device C, buttons on self-driving vehicle A, or buttons on light beam device 2) to change the wavelength of the light beam emitted by light beam device 2, thereby causing light beam device 2 to form light spots of different colors on marking element 1. This can quickly resolve the light spot disappearance issue during the image analysis step. Of course, in a more preferred embodiment, computing device A2 can automatically control the light beam device to emit light beams of different wavelengths if, after executing the image analysis step, computing device A2 determines that the image to be analyzed cannot be resolved into a light spot image.

[0102] It should be emphasized that the Apriltag mentioned above can be replaced with ArUco according to actual needs. In the self-driving vehicle precision measurement method described above, the use of Apriltag, compared to ArUco, has the following technical benefits: higher-precision positioning, adaptability to more complex environments, support for a wider range of angular offsets, and support for longer distances. However, the computing resources required for the use of Apriltag are slightly higher than those required for the use of ArUco. In addition, in terms of current program development resources, the support for related program development resources for Apriltag is higher than that for related program development resources for ArUco.

[0103] It should be emphasized that the precision measurement method for the self-driving vehicle of the present invention, whether using Apriltag or ArUco, is far superior to that of conventional two-dimensional barcodes (such as QR codes). Specifically, the positioning accuracy limits achievable using Apriltag, ArUco, and QR codes are sub-millimeter, millimeter, and centimeter levels, respectively. Therefore, using Apriltag and ArUco is clearly superior to QR codes.

[0104] In any of the above embodiments, in actual applications, in order to obtain optimal measurement results, the length and width of the Apriltag or ArUco are not less than 10 mm, and the total number of pixels contained in the length and width of the tag image, respectively, is at least 5 times the total number of pixels contained in the diameter of the light spot image. In other words, the diameter of the light spot image is smaller than the length or width of the tag image. In this way, the accuracy of the final measurement result can be effectively improved. Of course, in practice, the specific total number of pixels contained in the tag image and the light spot image is determined by the resolution of the image pickup device, the lens magnification used during pickup, and other designs, and is not limited here. If the diameter of the light spot image is greater than or equal to the length or width of the tag image, then in the image analysis step, the X-axis distance and the Y-axis distance cannot be calculated using the reference point and the center of the light spot image.

[0105] In summary, the self-driving vehicle accuracy measurement method, computing device, and self-driving vehicle of the present invention, through the design of distance calculation steps, regression center calculation steps, and average calculation steps, combined with the application of Apriltag (or ArUco), can more accurately and quickly measure the X-axis offset and Y-axis offset of the self-driving vehicle in the X-axis and Y-axis directions, thereby allowing relevant personnel to understand the accuracy of the self-driving vehicle.

[0106] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. Therefore, all equivalent technical changes made by applying the description and drawings of the present invention are included in the protection scope of the present invention.

Claims

1. A method for measuring the accuracy of a self-driving vehicle, characterized in that: The self-driving vehicle accuracy measurement method can measure an X-axis offset and a Y-axis offset of a self-driving vehicle in the X-axis direction and the Y-axis direction. The self-driving vehicle accuracy measurement method can be executed by a computing device. The self-driving vehicle accuracy measurement method includes: Repeating a distance calculation step a predetermined number of times, the predetermined number of times being no less than 2 times, the distance calculation step comprising: A control step: controlling the self-propelled vehicle to move to a predetermined position; a light beam emitting step of controlling a light beam device to emit at least one light beam toward a marking member to form at least one light spot on the marking member; the marking member has an April tag or ArUco, and during the light beam emitting step, the light spot is away from the April tag or ArUco; wherein the light beam device is disposed around the predetermined position and the marking member is disposed on the self-propelled vehicle, or the marking member is disposed around the predetermined position and the light beam device is disposed on the self-propelled vehicle; An image pickup step comprises controlling an image pickup device to pick up an image of the marking element to form an image to be analyzed, wherein the image to be analyzed includes a tag image corresponding to the April tag or ArUco and a light spot image corresponding to the light spot. An image analysis step comprises parsing the tag image in the image to be analyzed to obtain the coordinates of the four corners corresponding to the April tag or ArUco, and calculating an X-axis distance and a Y-axis distance between a reference point coordinate of the tag image and a center coordinate of a light spot of the light spot image in the X-axis direction and the Y-axis direction. The reference point coordinates are coordinates corresponding to one of the four corner coordinates, or the reference point coordinates are coordinates of a center point of the label image calculated using the four corner coordinates; a conversion step: using a real length of one side of the Apriltag or ArUco and a straight line distance calculated using the coordinates of two of the corners located on the same axis, converting the X-axis distance and the Y-axis distance into a real X-axis distance and a real Y-axis distance, and generating a measurement coordinate accordingly; The measured coordinates include the X-axis true distance and the Y-axis true distance; a regression center calculation step: using the plurality of measured coordinates to calculate a regression center coordinate through a regression operation; An average calculation step: using the multiple measured coordinates and the regression center coordinates, calculating the X-axis offset and the Y-axis offset of the self-propelled vehicle; the X-axis offset is the average value of the distance between the regression center coordinate and each of the measured coordinates in the X-axis direction, and the Y-axis offset is the average value of the distance between the regression center coordinate and each of the measured coordinates in the Y-axis direction.

2. The method for measuring the accuracy of a self-propelled vehicle according to claim 1, wherein: The color of the April tag or ArUco on the marking piece is composed of black or gray and white, and the length and width of the April tag or ArUco are not less than 10 mm.

3. The method for measuring the accuracy of a self-propelled vehicle according to claim 2, wherein: The total number of pixels included in the length and width of the label image is at least 5 times the total number of pixels included in the diameter of the light spot image.

4. The method for measuring the accuracy of a self-propelled vehicle according to claim 1, wherein: In the light beam emitting step, the light spot does not overlap with the Apriltag or ArUco on the marking element.

5. The method for measuring the accuracy of a self-propelled vehicle according to claim 1, wherein: If the maximum acceptable X-axis offset and the maximum acceptable Y-axis offset of the self-propelled vehicle in the X-axis direction and the Y-axis direction are respectively limited to an X-axis offset limit value and a Y-axis offset limit value, and when the self-propelled vehicle moves to the predetermined position and the self-propelled vehicle has no deviation in the X-axis direction and the Y-axis direction, in the light beam emitting step, the shortest straight-line distance between the April tag or ArUco on the marking piece and the light spot in the X-axis direction is greater than the X-axis offset limit value, and the shortest straight-line distance between the April tag or ArUco on the marking piece and the light spot in the Y-axis direction is greater than the Y-axis offset limit value.

6. The method for measuring the accuracy of a self-propelled vehicle according to claim 1, wherein: In the light beam emission step, the light beam device is controlled to emit two light beams toward the marking member to form two non-overlapping light spots on the marking member; in the image analysis step, a deflection angle is also calculated using the image to be analyzed; the deflection angle is the angle between a reference line of the label image and a center line of the two light spot images; the reference line passes through the reference point and one of the corner coordinates that is not the reference point; in the average calculation step, an average deflection angle is also calculated, and the average deflection angle is the average value of the deflection angles of multiple strokes.

7. The method for measuring the accuracy of a self-propelled vehicle according to claim 1, wherein: The marking member is a component independent of the self-propelled vehicle, and the marking member is detachably arranged on the self-propelled vehicle; or, the marking member is a part of the self-propelled vehicle.

8. The method for measuring the accuracy of a self-driving vehicle according to claim 1, wherein: The image analysis step comprises the following steps: A determination step: parsing the four corner coordinates corresponding to the April tag or ArUco, calculating the lengths of four edges corresponding to the April tag or ArUco using the four corner coordinates, and determining whether a difference between the lengths of the four edges exceeds a preset difference; If the difference between the lengths of the four side lines does not exceed the preset difference, a calculation step is performed: calculating the coordinates of the reference point and the coordinates of the center of the light spot, The X-axis distance and the Y-axis distance in the X-axis direction and the Y-axis direction; if the difference between the lengths of the four side lines exceeds the preset difference, an image correction step is first performed, and then the calculation step is performed; the image correction step is: correcting the image to be analyzed so that the difference between the lengths of the four side lines does not exceed the preset difference.

9. A computing device, characterized in that: The computing device is used to be installed on a self-propelled vehicle. The computing device can execute the precision measurement method for the self-propelled vehicle described in any one of claims 1 to 8, and the computing device can be connected to a processing device of the self-propelled vehicle. The processing device can control the self-propelled vehicle to move to the predetermined position. The computing device can be connected to the image pickup device, and the computing device can control the operation of the image pickup device, and the computing device can receive the image to be analyzed transmitted by the image pickup device.

10. The computing device according to claim 9, wherein: The computing device can be connected to an external electronic device, and the computing device can obtain the actual length and the predetermined number of times from the external electronic device, or the computing device can receive a modification message transmitted by the external electronic device to modify at least one of the pre-stored actual length and the predetermined number of times.

11. The computing device according to claim 9, wherein: After the computing device executes the average calculation step, the method further includes an output step of outputting the X-axis offset and the Y-axis offset to an external electronic device.

12. The computing device according to claim 9, wherein: The computing device can control the light beam device so that the light beam device emits light beams of different wavelengths, thereby forming light spots of different colors on the marking element.

13. A self-propelled vehicle, characterized in that: The self-propelled vehicle comprises: The computing device and the processing device according to claim 9; a detection device for detecting the surrounding environment of the self-propelled vehicle to generate detection information; The detection device is capable of transmitting the detection information to the processing device; A driving device is electrically connected to the processing device, and the processing device can control the action of the driving device according to the detection information to move the self-propelled vehicle to the predetermined position.

14. The self-propelled vehicle according to claim 13, characterized in that: The self-propelled vehicle further includes the marking component, the light beam device and the image pickup device.

15. The self-propelled vehicle according to claim 13, characterized in that: After executing the accuracy measurement method for the self-driving vehicle, the computing device stores the X-axis offset and the Y-axis offset as an X-axis correction value and a Y-axis correction value for the self-driving vehicle. After the computing device stores the X-axis correction value and the Y-axis correction value, when the computing device receives movement information, the computing device controls the self-driving vehicle to move to a specified position corresponding to the movement information based on the movement information, the X-axis correction value, and the Y-axis correction value.

16. The self-propelled vehicle according to claim 15, characterized in that: The computing device can switch between a mobile positioning mode and a calibration mode. When the computing device executes the mobile positioning mode, the computing device controls the self-driving vehicle to move to the predetermined position based on the movement information, the X-axis correction amount, and the Y-axis correction amount, and controls the image pickup device to pick up the April tag or ArUco around the predetermined position to obtain corresponding identification data; when the computing device determines that the identification data obtained after the self-driving vehicle moves to the predetermined position is the same as the default identification data included in the movement information, the computing device determines that the self-driving vehicle has moved to the designated position; when the computing device is in the calibration mode, the computing device will execute the accuracy measurement method of the self-driving vehicle.

17. The self-propelled vehicle according to claim 15, characterized in that: When the computing device receives the movement information, the computing device will first determine whether the X-axis correction value and the Y-axis correction value are stored in the self-driving vehicle. If the computing device determines that the X-axis correction value and the Y-axis correction value are not stored in the self-driving vehicle, the computing device can issue a calibration warning message. If the computing device receives a calibration request message after issuing the calibration warning message, the computing device will execute the accuracy measurement method for the self-driving vehicle and store the X-axis offset and the Y-axis offset obtained after executing the accuracy measurement method as the X-axis correction value and the Y-axis correction value of the self-driving vehicle.

18. The self-propelled vehicle according to claim 15, characterized in that: The computing device can control the light beam device so that the light beam device emits light beams of different wavelengths to form light spots of different colors on the marking element.

Images