Object position detection method and industrial machine

Abstract

[Problem] To detect the position of a target object, such as a workpiece, with high accuracy. [Solution] This industrial machine comprises: an imaging control unit that causes an imaging device to capture an image of a target object, the imaging device being provided at a position at which the imaging axis of the imaging device is not perpendicular to a placement surface on which the target object is placed; and a detection unit that detects a position of the target object in the planar direction on the basis of the image acquired by imaging the target object and the dimensions of the target object in the height direction acquired beforehand.

Description

Object position detection method and industrial machine

[0001] The present invention relates to an object position detection method and an industrial machine.

[0002] At bases where industrial machinery has been introduced, robots are used to grasp target objects such as workpieces, and the delivery of the workpieces to the industrial machinery is automated. For a robot, if the robot cannot accurately recognize the position of the workpiece, it is difficult to deliver the workpiece appropriately. Patent Document 1 discloses a technology in which a camera is positioned away from directly above the workpiece to prevent interference between the robot and the camera, the camera captures an image of the workpiece, and distortion in the captured image is corrected to detect the position of the workpiece in the installation location.

[0003] Japanese Patent Application Laid-Open No. 2021-65950

[0004] Simply photographing a workpiece with a camera placed diagonally above the workpiece and correcting the distortion of the photographed image will result in a detection result that does not take into account the height of the workpiece, making it difficult to detect the position of a target object such as a workpiece with high accuracy.

[0005] An aspect of the present invention provides an object position detection method for detecting the position of a target object. The object position detection method includes an imaging step of capturing an image of the target object using an imaging device installed at a position where the imaging axis is not perpendicular to an installation surface on which the target object is installed. The object position detection method also includes a detection step of detecting the position of the target object in a planar direction based on the image captured in the imaging step and a height dimension of the target object captured in advance.

[0006] According to an aspect of the present invention, there is provided an industrial machine that detects the position of a target object. The industrial machine includes an imaging control unit that causes an imaging device installed at a position where the imaging axis is not perpendicular to a surface on which the target object is placed to capture an image of the target object. The industrial machine also includes a detection unit that detects the position of the target object in a planar direction based on an image acquired by capturing the image of the target object and a height dimension of the target object acquired in advance.

[0007] According to the object position detection method and industrial machinery of the present invention, for an object photographed from diagonally above, the position of the object in a planar direction is detected based on the photographed image and the height dimension of the object, so that the position of the object in a planar direction can be detected with high accuracy without requiring complex processing.

[0008] In the object position detection method of the above aspect, in the detection step, the position of the detection target point of the target object in the captured image is detected in the planar direction by determining the intersection between a line connecting the imaging device and the detection target point and a plane offset from the installation surface in the height direction of the target object by the height dimension of the target object. According to this aspect, the position of the detection target point of the target object in the captured image is detected by determining the intersection with a plane offset from the installation surface by the height dimension of the target object, thereby enabling more accurate detection of the position of the target object. In the object position detection method of the above aspect, the target object is circular when viewed perpendicular to the installation surface, and in the detection step, the position of the center point of the top surface of the target object in the planar direction is detected based on the height dimension of the target object acquired in advance. According to this aspect, the position of the center point of the top surface of a circular target object when viewed perpendicular to the installation surface can be detected with high accuracy without requiring complex processing. In the object position detection method of the above aspect, the target object is a workpiece of industrial machinery, and the industrial machinery picks the target object based on the position of the target object detected in the detection step. According to this aspect, the object position detection method can be applied to applications in which industrial machinery picks up target objects as workpieces. The object position detection method of the above aspect also includes an acquisition step of acquiring the height dimension of the target object using a sensor, and in the detection step, the position of the target object in a planar direction is detected based on the height dimension of the target object acquired in the acquisition step. According to this aspect, since the height dimension of the target object is acquired using a sensor, the method can be applied even when target objects with different height dimensions are placed on the installation surface. The object position detection method of the above aspect also includes a reception step of accepting an input of the height dimension of the target object through an input operation by an operator, and in the detection step, the position of the target object in a planar direction is detected based on the height dimension of the target object accepted in the acceptance step. According to this aspect, since the height dimension of the target object is accepted through an input operation by an operator, the method can be applied even when target objects with different height dimensions are placed on the installation surface.

[0009] A diagram showing an example of the system configuration of an object position detection system according to an embodiment. A diagram showing an example of the functional configuration of an industrial machine according to an embodiment. A diagram for explaining photographing of a target object by an imaging device according to an embodiment. A diagram for explaining photographing of a target object by an imaging device according to an embodiment. A diagram for explaining photographing of a target object by an imaging device according to an embodiment. A diagram for explaining an example of projective transformation according to an embodiment. A diagram for explaining an example of detection of the positions of detection target points according to an embodiment. A flowchart showing an example of the flow of object position detection processing according to an embodiment. A diagram showing another example of the functional configuration of an industrial machine according to an embodiment. A flowchart showing another example of the flow of object position detection processing according to an embodiment.

[0010] The present invention will be described below through embodiments of the invention. The following embodiments do not limit the scope of the invention. In the drawings, the scale may be appropriately changed, such as by enlarging, reducing, or emphasizing parts in order to explain the embodiments.

[0011] FIG. 1 is a diagram illustrating an example of a system configuration of an object position detection system according to an embodiment. As illustrated in FIG. 1 , the object position detection system 1 includes an industrial machine 10, an imaging device 20, and a robot 30, and detects the planar position of a target object 5 placed on an installation surface 3. The planar position of the target object 5 corresponds to its position on the installation surface 3. The object position detection system 1 includes, for example, a machine tool such as a lathe or a conveyance device installed in a factory. In the object position detection system 1, the industrial machine 10 may include the robot 30. This embodiment is applicable to machine tools and picking devices in conveyance systems used in general logistics, and the target object 5, serving as a workpiece of the industrial machine 10, is a target for loading and transport. In the object position detection system 1, the target object 5 is placed at any position on the installation surface 3 without being aligned in a predetermined position. The imaging device 20 is provided at a position where its imaging axis is not perpendicular to the installation surface 3 on which the target object 5 is placed, and captures an image of the target object 5 placed on the installation surface 3. The robot 30 grasps and moves the target object 5. For example, the robot 30 is equipped with an arm for grasping the target object 5 and carries the grasped target object 5 into the industrial machine 10 .

[0012] In the above-described configuration, the industrial machine 10 captures an image of the target object 5 using the imaging device 20, which is provided at a position where the imaging axis is not perpendicular to the installation surface 3 on which the target object 5 is placed. The imaging device 20 is positioned so as not to interfere with the robot 30, etc. Therefore, the imaging device 20 captures an image of the target object 5 from a position where the imaging axis is not perpendicular to the installation surface 3. The industrial machine 10 then detects the position of the target object 5 in a planar direction based on the image captured by the imaging device 20 and the height dimension of the target object 5 acquired in advance. Because the image captured by the imaging device 20 is not captured from directly above the target object 5, distortion occurs in the image of the target object 5. As described above, the position of the target object 5 in a planar direction at which the robot 30 can grasp it corresponds to its position on the installation surface 3. However, the position in the image captured by the imaging device 20 cannot be used as is. As a result, the industrial machine 10 eliminates distortions in the acquired image and detects the horizontal position of the target object 5 with higher accuracy using the height dimension of the target object 5. In the object position detection system 1, since the coordinate systems of the imaging device 20 and the robot 30 are different from each other, after detecting the horizontal position of the target object 5, a transformation is performed from the coordinate system of the imaging device 20 to the coordinate system of the robot 30. This allows the robot 30 to accurately grasp the position of the target object 5 and appropriately grasp the target object 5. Regarding the coordinate system, if the robot 30 operates on three axes, it is desirable that two of the axes are parallel to the installation surface 3. For example, if the XY plane and the installation surface 3 are parallel to each other among the three axes on which the robot 30 operates, it becomes easier to determine the XY coordinate position on which the robot 30 operates, thereby improving the accuracy of control of the robot 30 and speeding up processing in the object position detection system 1 as a whole.

[0013] 2 is a diagram illustrating an example of a functional configuration of an industrial machine according to an embodiment. As illustrated in FIG. 2, the industrial machine 10 includes an image capture control unit 110 and a detection unit 120. The image capture control unit 110 causes the image capture device 20, which is provided at a position where the image capture axis is not perpendicular to the installation surface 3 on which the target object 5 is placed, to capture an image of the target object 5. Specifically, in a situation where the target object 5 is placed on the installation surface 3, the image capture control unit 110 causes the image capture device 20, which is provided at a position where the image capture axis is not perpendicular to the installation surface 3, i.e., at a position diagonally above the target object 5, to capture an image of the target object 5.

[0014] Here, referring to FIGS. 3 to 5 , an example of capturing an image of a target object 5 using an imaging device 20 according to an embodiment will be described. FIG. 3 illustrates an example of capturing an image of a target object 5 using an imaging device 20 installed in a position where the imaging axis is perpendicular to the installation surface 3. The case illustrated in FIG. 3 differs from the installation target of the imaging device 20 according to the present embodiment, but is used as an example to explain the differences from FIGS. 4 and 5 . FIGS. 4 and 5 illustrate an example of capturing an image of a target object 5 using an imaging device 20 installed in a position where the imaging axis is not perpendicular to the installation surface 3. In FIGS. 4 and 5 , the target object 5 illustrated in FIG. 4 has a smaller height dimension than the target object 5 illustrated in FIG. 5 . Note that FIGS. 3 to 5 illustrate an example of a target object 5 that is a circular object (e.g., the cylindrical object illustrated in FIG. 1 ) when viewed from a direction perpendicular to the installation surface 3. The outline of the circle on the top surface is represented by a solid line and the center of the circle is represented by a square, and the outline of the circle on the bottom surface is represented by a dashed line and the center of the circle is represented by a square rotated 45 degrees. Note that the image (captured image) according to this embodiment is a two-dimensional image that does not include a distance image, meaning that a camera that does not include the capability to acquire a distance image can be used as the imaging device 20.

[0015] 3 , when the target object 5 is photographed by the imaging device 20 provided at a position where the imaging axis is perpendicular to the installation surface 3, the circle on the top surface and the circle on the bottom surface of the target object 5 are photographed as larger circles depending on the distance from the imaging device 20. Furthermore, the circles on the top surface and the bottom surface of the target object 5 are photographed as nearly perfect circles, and the centers of the circles on the top surface and the bottom surface of the target object 5 coincide. Note that when the target object 5 is photographed by the imaging device 20 provided at a position where the imaging axis is perpendicular to the installation surface 3, the circle on the bottom surface of the target object 5, represented by a dashed line, is not actually photographed.

[0016] As shown in FIGS. 4 and 5 , when the target object 5 is photographed using the imaging device 20 whose imaging axis is not perpendicular to the installation surface 3, the upper and lower circles of the target object 5 are photographed as larger circles, depending on the distance from the imaging device 20. Furthermore, the upper and lower circles of the target object 5 are photographed as not perfect circles, and the centers of the upper and lower circles of the target object 5 do not coincide. Specifically, the center of the upper circle of the target object 5 is located higher than the center of the lower circle of the target object 5 in the photographed image. When the target object 5 is photographed using the imaging device 20 whose imaging axis is not perpendicular to the installation surface 3, a portion of the lower circle of the target object 5, represented by a dashed line, is captured in the photograph. Furthermore, the larger the height dimension of the target object 5, the more the lower circle of the target object 5, represented by a dashed line, is captured. 4 and 5, the target object 5 shown in FIG. 5, which has a larger height dimension, has more circles reflected on the bottom surface than the target object 5 shown in FIG.

[0017] Returning to the description of FIG. 2 , the detection unit 120 detects the position of the target object 5 in the planar direction based on an image acquired by photographing the target object 5 and the height dimension of the target object 5 acquired in advance. Specifically, when the target object 5 is photographed by the imaging device 20, the detection unit 120 acquires the image from the imaging device 20. The detection unit 120 then performs projective transformation (e.g., trapezoidal correction) on the acquired image to make the upper and lower circles closer to perfect circles, and then detects the position of the detection point of the target object 5 in the image. The position of the detection point in the planar direction is detected by determining the intersection of a line connecting the imaging device 20 and the detection point with a plane obtained by offsetting the installation surface 3 by the height dimension of the target object 5 in the height direction of the target object 5. Note that if the target object 5 is cylindrical, the detection point is the position of the center point of the upper surface of the target object 5 in the planar direction.

[0018] When the reflection of the bottom circle is significant, the position of the target object 5 can be detected using the bottom circle. Detecting the position of the target object 5 using the bottom circle is not affected by the height of the target object 5. The reflection of the bottom circle changes depending on the position of the target object 5 relative to the shooting range of the imaging device 20, the shape of the target object 5, and other factors. The position of the target object 5 can be detected using the top circle even when conditions such as the shape of the target object 5 relative to the shooting range of the imaging device 20 are different. However, when detecting the position of the target object 5 using the top circle, the detection is affected by conditions such as the shape of the target object 5 relative to the shooting range of the imaging device 20, and correction for the effect of the height of the target object 5 becomes necessary. In other words, if the effect of the height of the target object 5 is corrected, the position of the target object 5 can be suitably detected even when conditions such as the shape of the target object 5 relative to the shooting range of the imaging device 20 are different.

[0019] Here, an example of detecting the position of a detection target point according to an embodiment will be described with reference to FIGS. 6 and 7 . FIG. 6 is a diagram illustrating an example of projective transformation according to an embodiment. FIG. 7 is a diagram illustrating an example of detecting the position of a detection target point according to an embodiment. Detected point A shown in FIGS. 6 and 7 corresponds to the center of the upper circle shown in FIGS. 4 and 5 , and point L corresponds to the center of the lower circle shown in FIGS. 4 and 5 . However, since detected point A is a point on a plane on the image and does not take into account the height dimension of the target object 5, the point corresponding to the actual center of the upper circle of the target object 5 (referred to as "detected point A'") is not detected point A. As shown in FIG. 6 , the detection unit 120 acquires an image (captured image) from the imaging device 20 and corrects distortion in the image by projective transformation of the image. As a result, as shown in FIG. 6 , the centers of the upper and lower circles are different in the captured image and the image after projective transformation. The detection unit 120 uses detected point A (xa, ya) after projective transformation to perform the height direction correction process described below. Note that the imaging device 20 may not capture images of the target objects 5 after they have been placed in predetermined positions (on the assumption that each target object 5 will be placed in the same position). In such cases, for example, the shape of the target objects 5 can be known in advance, and therefore it is sufficient to recognize that the target objects 5 are placed by detecting them using pattern matching or the like based on the shape of the target objects 5 (for example, detecting a circle).

[0020] As shown in FIG. 7 , the detection unit 120 calculates a line CA from the detection point A (xa, ya, 0) after the projection transformation and the position C (Xc, Yc, Zc) of the imaging device 20. The position C is the lens center of the camera model of the imaging device 20. The position C may be measured using a test pattern and pre-stored in the industrial machine 10, or may be input appropriately via operation. Because the detection point A is a point (two-dimensional point) on the projectively transformed captured image, its coordinate in the height direction is set to "0." When calculating the line CA, actual size values ​​or pixels (e.g., pixel number, pixel position) may be used. The detection unit 120 then calculates the z=H coordinate of the line CA (the intersection of the line CA and the plane H). The plane H is a plane corresponding to the height of the target object 5 placed on the installation surface 3. As a result, the detection unit 120 calculates the detection target point A' (xh, yh, H). The point L corresponding to the center of the circle on the lower surface is (xh, yh). The order of the three steps of projective transformation, position detection, and height correction described above does not matter, and for example, projective transformation may be performed after position detection. In addition, the position of a characteristic shape may be detected from the captured image, and the detected position may be subjected to projective transformation before height correction.

[0021] The industrial machine 10 converts the detection target point A' from the coordinate system of the imaging device 20 to the coordinate system of the robot 30 and notifies the robot 30 of the detection target point A'. At this time, if the target object 5 is cylindrical, the industrial machine 10 notifies the detection target point A' along with the diameter of the circle (or the radius of the circle). This allows the robot 30 to grasp the circular outline of the upper surface of the cylindrical target object 5. If the target object 5 is a workpiece of the industrial machine 10, the industrial machine 10 including the robot 30 picks the target object 5 based on the position of the target object 5 detected by the detection unit 120. Furthermore, the industrial machine 10 outputs an error for the detection target point A' if the position is outside the normal range, for example, if the detection target point A' is in a position that the robot 30 cannot grasp, is an outlier that is not within a predetermined range, or is otherwise an error.

[0022] FIG. 8 is a flowchart showing an example of the flow of object position detection processing according to the embodiment. As shown in FIG. 8 , the image capture control unit 110 causes the image capture device 20 to capture an image of the target object 5 (step S101). Specifically, in a situation where the target object 5 is placed on the installation surface 3, the image capture control unit 110 causes the image capture device 20 to capture an image of the target object 5 in a position where the image capture axis is not perpendicular to the installation surface 3, i.e., diagonally above the target object 5. The detection unit 120 then detects the position of the target object 5 in the planar direction based on the image captured by the image capture and the height dimension of the target object 5 (step S102). Specifically, the detection unit 120 acquires the captured image from the image capture device 20 and performs projective transformation on the acquired image to determine a line CA connecting the position C of the image capture device 20 and the detected point A in the projectively transformed image. The detection unit 120 then determines the intersection of the line CA with a plane H corresponding to the height of the target object 5 placed on the installation surface 3, thereby determining a detection target point A' (see FIGS. 6 and 7 , etc.).

[0023] After these steps, the industrial machinery 10 transforms the detection target point A' from the coordinate system of the imaging device 20 to the coordinate system of the robot 30 and notifies the robot 30 of the detection target point A'. At this time, if the target object 5 is cylindrical, the industrial machinery 10 notifies the detection target point A' along with the diameter of the circle (or the radius of the circle). This allows the robot 30 to grasp the circular outline of the upper surface of the cylindrical target object 5. If the target object 5 is a workpiece of the industrial machinery 10, the industrial machinery 10 including the robot 30 picks the target object 5 based on the position of the target object 5 detected by the detection unit 120. Furthermore, the industrial machinery 10 outputs an error for the detection target point A' if the position is outside the normal range, for example, if the detection target point A' is in a position that the robot 30 cannot grasp, is an outlier that is not within a predetermined range, or is otherwise an error.

[0024] In the description of the embodiment, it has been explained that the height dimension of the target object 5 has been acquired in advance. The height dimension of the target object 5 may be determined by using a sensor or the like and using a value obtained by the sensor. Fig. 9 is a diagram showing another example of the functional configuration of an industrial machine according to the embodiment. Note that in Fig. 9, the same components as those in the above-described embodiment are denoted by the same reference numerals, and detailed description thereof may be omitted.

[0025] 9 , the industrial machine 10A includes an imaging control unit 110, a detection unit 120, and an acquisition unit 130. The acquisition unit 130 acquires the height dimension of the target object 5 using a sensor. Specifically, when the target object 5 is carried into a predetermined position, the acquisition unit 130 acquires the height dimension of the target object 5 from the sensor, based on measurements by the sensor that can measure the height dimension of the target object 5. As a result, the detection unit 120 detects the position of the target object 5 in the planar direction, based on the image acquired by the imaging control unit 110 and the height dimension of the target object 5 acquired by the acquisition unit 130.

[0026] 10 is a flowchart showing another example of the flow of the object position detection process according to the embodiment. In FIG. 10, steps S202 and S203 are the same processes as steps S101 and S102 described above, respectively. Detailed descriptions of processes that are the same as those in the above-described embodiment may be omitted.

[0027] 10 , the acquisition unit 130 acquires the height dimension of the target object 5 using a sensor (step S201). Specifically, when the target object 5 to be photographed by the imaging device 20 is brought into a predetermined position, the acquisition unit 130 acquires the height dimension of the target object 5 from the sensor based on measurements by the sensor. Then, the imaging control unit 110 causes the imaging device 20 to photograph the target object 5 (step S202). Next, the detection unit 120 detects the position of the target object 5 in the planar direction based on the image acquired by the photograph and the height dimension of the target object 5 (step S203). In step S203, the height dimension of the target object 5 acquired by the acquisition unit 130 is used.

[0028] Furthermore, the height dimension of the target object 5 may be a value input by the operator. Fig. 11 is a diagram showing another example of the functional configuration of the industrial machine according to the embodiment. Note that in Fig. 11, the same components as those in the above-described embodiments are denoted by the same reference numerals, and detailed description thereof may be omitted.

[0029] As shown in FIG. 11 , the industrial machine 10B includes an image capture control unit 110, a detection unit 120, and a reception unit 140. The reception unit 140 receives input of the height dimension of the target object 5 through an input operation by an operator. Specifically, the reception unit 140 receives input of the height dimension of the target object 5 that is to be imaged by the image capture device 20 through an input operation by an operator using an input device (e.g., a touch panel, a keyboard, a mouse, etc.) connected to the industrial machine 10B, etc. Note that the reception process by the reception unit 140 may be performed either before or after the process by the image capture control unit 110. As a result, the detection unit 120 detects the position of the target object 5 in the planar direction based on the image acquired by the image capture control unit 110 and the height dimension of the target object 5 received by the reception unit 140.

[0030] 12 is a flowchart showing another example of the flow of the object position detection process according to the embodiment. In FIG. 12, steps S301 and S303 are the same processes as steps S101 and S102 described above, respectively. Detailed descriptions of processes that are the same as those in the above-described embodiment may be omitted.

[0031] 12 , the imaging control unit 110 causes the imaging device 20 to capture an image of the target object 5 (step S301). Then, the receiving unit 140 receives input of the height dimension of the target object 5 through an input operation by the operator (step S302). Specifically, the receiving unit 140 receives input of the height dimension of the target object 5, which is being captured by the imaging device 20, through an input operation by the operator using an input device connected to the industrial machine 10B or the like. Next, the detection unit 120 detects the position of the target object 5 in the planar direction based on the image acquired by the capture and the height dimension of the target object 5 (step S303). In step S303, the height dimension of the target object 5 received by the receiving unit 140 is used.

[0032] The object position detection system 1 is used in applications including lathes. For example, in the object position detection system 1, a target object 5 serving as a workpiece before machining is carried in by a belt conveyor, and the belt conveyor stops when the workpiece reaches a sensor position. Then, in the object position detection system 1, upon receipt of the workpiece reaching the sensor position, an industrial machine 10 serving as a lathe instructs an imaging device 20 to take an image, and the industrial machine 10 acquires the image captured by the imaging device 20. Next, in the object position detection system 1, when the industrial machine 10 detects the position of the workpiece, the detected position of the workpiece is notified to a robot 30 serving as a loader, and an instruction is given to carry in the workpiece. Then, in the object position detection system 1, the loader transports the workpiece from the belt conveyor to the spindle of the industrial machine 10, and the industrial machine 10 machines the workpiece on the spindle. In the object position detection system 1, after the industrial machine 10 has finished machining the workpiece, the loader removes the machined workpiece.

[0033] The object position detection system 1 also has applications in production lines. For example, in the object position detection system 1, a target object 5, which is a workpiece before machining, is carried in by a belt conveyor, and a sensor disposed at the top end of the belt conveyor measures the height of the workpiece. Here, the sensor measures the height of the workpiece from the change in the height of the belt conveyor and the height of the workpiece. Then, in the object position detection system 1, when the workpiece reaches the sensor position, the belt conveyor stops. Upon receiving the sensor position, the industrial machine 10 instructs the imaging device 20 to take an image, and the industrial machine 10 acquires the image captured by the imaging device 20. Next, in the object position detection system 1, when the position of the workpiece is detected by the industrial machine 10, the detected position of the workpiece is notified to the robot 30.

[0034] Although the embodiment has been described with reference to an example in which the target object 5 is a cylindrical object, this is not limiting. For example, the target object 5 may be a rectangular parallelepiped object. In this case, the object position detection system 1 detects the positions (coordinates) of the four rectangular corners of the upper surface of the rectangular parallelepiped, calculates the length and position of the vertical and horizontal sides (one side and the other side), and notifies the robot 30 of these, thereby allowing the robot 30 to grasp the target object 5. Furthermore, for example, the target object 5 may have a shape in which the upper and lower surfaces are parallel to each other, and the position and length of the sides can be estimated from the shape of the upper surface. Even in such a case, various information for pattern matching may be stored in advance so that the imaging device 20 can capture the target object 5. Furthermore, the shapes of the upper and lower surfaces of the target object 5 do not need to be identical. For example, the position of the target object 5 can be detected as long as the position of the target object 5 can be estimated from the shape of the upper surface. Furthermore, for the target object 5, the length determined from multiple positions of the target object 5, the area obtained by connecting the multiple positions with straight lines, the orientation of the target object 5, and the like can also be calculated. The conditions for the target object 5 according to the embodiment are that the upper surface has a shape whose position can be detected, the shape has a flat surface, the flat surface is parallel to the installation surface 3, and the height from the installation surface 3 is known. As a result, the position of the target object 5 can be estimated from the shape of the upper surface. As described above, if the target object 5 is a cylinder, the shape whose position can be detected is a circle on the upper surface, and the position can be estimated and detected from the center of the circle on the upper surface. In addition, if the target object 5 is a truncated cone, the centers of the circles on the upper and lower surfaces are the same, so the position of the target object 5 can be estimated and detected from the center position of the truncated cone. Furthermore, if the upper surface of the target object 5 has a surface parallel to the installation surface 3, a marker or the like can be attached to that portion, the position of the marker can be detected, and the position of the target object 5 can be estimated and detected from the position of the marker. Furthermore, if the upper surface of the cylinder of the target object 5 is spherical, the shape consisting of the spherical portion and the outline of the side surface is circular, and the outline of this circle is parallel to the installation surface 3, so the position of the target object 5 can be estimated and detected from the center position of the circle. Furthermore, the installation surface 3 is not limited to the XY plane (for example, the floor surface) shown in FIG. 1, but may be the XZ plane (for example, the wall surface) shown in FIG.

[0035] The object position detection system 1 is also applicable to a case where the target object 5 is loaded into a container and carried in by a conveyor. In this case, the height of the target object 5 from the installation surface 3 is the sum of the height of the target object 5 itself and the height of the bottom of the container. That is, when the target object 5 is loaded into a container or the like, the object position detection system 1 detects the position of the target object 5 in the planar direction by using the height including the height of the bottom of the container as the height dimension of the target object 5.

[0036] As described above, the object position detection system 1 detects the position of the target object 5 in a planar direction based on the image captured by the imaging device 20 and the height dimension of the target object 5, for a target object 5 photographed from diagonally above. This allows for highly accurate detection of the position of the target object 5 in a planar direction without requiring complex processing. Furthermore, the object position detection system 1 detects the detection target point A' of the target object 5 in the image captured by the imaging device 20 by determining the intersection with a plane H offset from the installation surface 3 by the height dimension of the target object 5. This allows for more accurate detection of the position of the target object 5. Furthermore, the object position detection system 1 can detect the position of the center point of the top surface of a target object 5 that is circular when viewed perpendicularly to the installation surface 3 with high accuracy without requiring complex processing. Furthermore, the object position detection system 1 can be applied to applications in which the industrial machine 10 picks up the target object 5 as a workpiece. Furthermore, because the object position detection system 1 acquires the height dimension of the target object 5 using a sensor, it can be applied even when target objects 5 of different heights are installed on the installation surface 3. Furthermore, since the object position detection system 1 accepts the height dimension of the target object 5 through input operations by the operator, it can also be applied when target objects 5 with different height dimensions are installed on the installation surface 3.

[0037] Although the embodiments of the present invention have been described above, the technical scope of the present invention is not limited to the above embodiments. It will be apparent to those skilled in the art that various modifications and improvements can be made to the above embodiments. Furthermore, forms incorporating such modifications or improvements are also included within the technical scope of the present invention. One or more of the requirements described in the above embodiments may be omitted. Furthermore, the requirements described in the above embodiments may be combined as appropriate. Furthermore, the execution order of each process shown in the present embodiment can be implemented in any order, as long as the output of a previous process is not used in a subsequent process. Furthermore, even if the operations in the above embodiments are described using terms such as "first," "next," and "continuously" for convenience, this order is not required. Furthermore, to the extent permitted by law, the disclosures of Japanese Patent Application No. 2023-169867 and all documents cited in the above embodiments are incorporated herein by reference.

[0038] DESCRIPTION OF SYMBOLS 1... Object position detection system 3... Installation surface 5... Target object 10, 10A, 10B... Industrial machine 20... Imaging device 30... Robot 110... Photography control unit 120... Detection unit 130... Acquisition unit 140... Reception unit

Claims

1. A method for detecting object position comprising: an imaging step of capturing an image of a target object using an imaging device disposed at a position where the imaging axis is not perpendicular to a surface on which the target object is placed; and a detection step of detecting the position of the target object in a planar direction based on the image captured in the imaging step and a height dimension of the target object that was captured in advance.

2. The object position detection method according to claim 1, wherein in the detection step, the position of the detection target point of the target object in the captured image is detected by determining the intersection of a line connecting the imaging device and the detection target point with a plane obtained by offsetting the installation surface in the height direction of the target object by an amount equal to the height dimension of the target object, thereby detecting the position of the detection target point in the planar direction.

3. The object position detection method according to claim 1, wherein the target object is a circular object when viewed from a direction perpendicular to the installation surface, and in the detection process, the position of the center point of the top surface of the target object in the planar direction is detected based on a height dimension of the target object obtained in advance.

4. An object position detection method as claimed in any one of claims 1 to 3, wherein the target object is a workpiece of an industrial machine, and the industrial machine picks up the target object based on the position of the target object detected in the detection process.

5. An object position detection method as described in claim 1, further comprising an acquisition step of acquiring a height dimension of the target object using a sensor, and in the detection step, the position of the target object in a planar direction is detected based on the height dimension of the target object acquired in the acquisition step.

6. An object position detection method as described in claim 1, further comprising a receiving step of receiving an input of a height dimension of the target object by an input operation by an operator, and in the detection step, a position of the target object in a planar direction is detected based on the height dimension of the target object received in the receiving step.

7. An industrial machine comprising: an imaging control unit that causes an imaging device, the imaging device being disposed at a position where the imaging axis is not perpendicular to a surface on which the target object is placed, to capture an image of the target object; and a detection unit that detects the position of the target object in a planar direction based on an image acquired by photographing the target object and a height dimension of the target object acquired in advance.