Patent Application
20230096477
The present invention relates to a technique to measure a workpiece placed in a machine tool.
A machine tool performs machining on a workpiece such as drilling. As a method of checking machining results, a worker removes the workpiece from the machine tool and measures the workpiece using an instrument. When checking the machining results of the workpiece during the process of machining, it takes time and effort to remove the workpiece once, measure it, and then place it again.
There is also known a method using a touch probe attached to the main shaft as a workpiece measurement method. In this method, the main shaft is moved, and the touch probe is lowered from this position to detect contact with the workpiece. In a case of measuring a plurality of areas, an approach motion is needed in which the main shaft is moved toward each of the areas and the touch probe is lowered. In this method, much time is required for the movement of the main shaft and the approach motion of the touch probe. There is also a problem that it is difficult for the touch probe to measure the fine shapes of a workpiece, and thus the touch probe is not suitable for measurement of the orientation of a workpiece.
Patent Literature 1: JP 2019-185581 A
There is another possible method to perform measurement at a higher speed than the method using a touch probe. This method uses an imaging device to image an image of a machined workpiece, analyze the image, and measure the machining results. Measurement by means of imaging does not need the approach motion of the touch probe. A plurality of areas of the workpiece can be captured by a single imaging. This can reduce the number of times of movement of the main shaft.
Patent Literature 1 describes an example in which a photographing device is used to photograph a workpiece during the process of machining. In this example, a three-dimensional model is created using the photographed image of the workpiece, and compared with a three-dimensional model in machining simulation to determine whether there is any machining defect.
Patent Literature 1 describes that feature points are extracted from the image data, and a three-dimensional model is generated based on the feature points. However, how to specifically perform the processing is unclear. To be more specific, the location where a camera is installed, and how an image of the workpiece is captured is unclear. Accordingly, it is assumed that an image of a rough shape of the workpiece is captured, and measurement of the shape of fine portions is not assumed.
An information processing device according to an aspect of the present invention is an information processing device to process a two-dimensional image generated by being imaged by an imaging unit in a machine tool including an attachment unit to which the imaging unit or a tool is attachable, the attachment unit moving to image an image of a workpiece placed in the machine tool, the imaging unit imaging the image of the workpiece at an imaging position, the information processing device including a computation unit for: (i) calculating a position of the workpiece based on an image that is imaged after the imaging unit has moved to a predetermined position based on machine coordinates of the machine tool; and (ii) detecting an edge of the workpiece from the two-dimensional image that is imaged by the imaging unit at an imaging position after the imaging unit has been moved from the predetermined position to the imaging position, and calculating a length of the workpiece from a plurality of edges detected.
According to the present invention, it is possible to provide a technique to measure a workpiece placed in a machine tool at a high speed.
A machining system to measure a workpiece placed on a machine tool according to an embodiment will be described below with reference to the drawings. Descriptions are provided below while constituent elements having identical configurations are denoted by like reference signs.
In the present embodiment, there is described a case where a machine tool is used. In the process of machining based on an NC (Numerical Control) program, when a tool is replaced with another tool to be used next based on the NC program, this tool can be automatically attached to an attachment unit (for example, the main shaft or turret) of the machine tool by an ATC (Automatic Tool Changer).
The machine tool illustrated on the right side of
The main shaft 100 is movable in the forward-rearward direction and the rightward-leftward direction. The machining position for the workpiece W is changed by horizontally moving the position of the main shaft 100. The main shaft 100 can also move in the vertical direction. The machine tool moves the main shaft 100 upward or downward to vary the distance between the workpiece W and the main shaft 100.
In the present embodiment, an imaging unit 600 can be attached to the main shaft 100, and is used for imaging the workpiece W. The machine tool has a configuration in which the imaging unit 600 can be automatically attached to and detached from the attachment unit by the ATC. The main shaft 100 is an example of “an attachment unit to which the imaging unit is attachable”. When the machine tool is a turning center, the turret corresponds to “an attachment unit of the machine tool”. The imaging unit 600 may be attached to the turret to image the workpiece W. In a multifunction processing machine, the imaging unit 600 may be attached. In any of the cases, the imaging unit 600 is attachable to “an attachment unit of the machine tool”.
The imaging unit 600 receives an instruction from the information processing device, images an image of a workpiece positioned in the machine tool, and transmits a two-dimensional image to the information processing device. The imaging unit 600 includes an imaging element, a lens, an illuminating device, and the like of a CMOS (Complementary Metal Oxide Semiconductor) image sensor or other type of sensor. The imaging unit 600 is capable of performing wired or wireless communication. The imaging unit 600 may be supplied with power through wires or wirelessly, or may include a storage battery.
Inside the machine tool, there is a machining space for attaching a tool to the attachment unit and machining a workpiece. The position in the closed machining space is defined by machine coordinates (Xm, Ym, Zm). The machine coordinate system defines the position base on the position and orientation in the machine tool. The machine tool operates based on the machine coordinates. It is assumed that the position of the origin of the machine coordinate system is defined in advance. The workpiece W is placed at a predetermined position in a predetermined orientation. The machine tool operates on the assumption that the workpiece W is placed in such a manner as described. In a case where the machine tool performs machining such as cutting, a cutting tool is attached to the main shaft 100, and an operation to move the main shaft 100 is combined with an operation to rotate the main shaft 100 to control the cutting tool so as to touch the workpiece W, while the cutting tool is rotated.
Conventionally, when checking the machining results of the workpiece W such as the diameter of a hole drilled on the workpiece W, a worker removes the workpiece W from the machine tool to measure the machining results using an instrument. Otherwise, without removing the workpiece W from the machine tool, the measurement is performed using a touch probe attached to the attachment unit.
In the present embodiment, the imaging unit 600 images an image of the workpiece W placed in the machine tool to analyze the image, and measure the machining results. In the present embodiment, an image probe attachable to/detachable from the attachment unit is used to serve as the imaging unit 600, such that the cutting results of the workpiece W can be continuously measured while the workpiece W remains fixed in the same state as being cut in the machine tool. In the attachment unit, a tool and the imaging unit 600 are replaceable with each other. The tool attached to the attachment unit machines the workpiece W, and the imaging unit 600 attached to the attachment unit images the workpiece W. After the workpiece W is machined with the tool in the machine tool, the tool attached to the attachment unit is returned to a storage unit. Instead, the imaging unit 600 in the storage unit is attached to the attachment unit. The imaging unit 600 attached to the attachment unit moves to a position where the imaging unit 600 images the workpiece W, and images an image of the workpiece W. In such a machine tool as described above, it is unnecessary to remove the workpiece W from the machine tool and measure the workpiece W outside the machine tool. The workpiece W can be machined and imaged by moving the attachment unit. During the execution of a measurement program (imaging mode), the attachment unit having the imaging unit 600 attached thereto moves to the machine coordinates corresponding to the position of the machine-processing origin of the workpiece W. There, the imaging unit 600 images an image and acquires the image. Edge detection of the workpiece W is performed from the image. The machine-processing origin in the present embodiment is positioned at one of the corners of the rectangle when the workpiece W is viewed in the direction from the topside. Therefore, the edge at the corner is checked against the position of the machine coordinates corresponding to the machine-processing origin. When the position of the corner of the workpiece W coincides with the position of the machine coordinates corresponding to the machine-processing origin, then the attachment unit is moved to the measurement position of the workpiece W in accordance with a process of the measurement program. For example, if it is desired to measure the diameter of a hole formed by drilling, the attachment unit is moved from the imaging position for checking against the machine-processing origin to an imaging position for imaging the hole on the workpiece W. Then, an image is imaged so as to photograph the hole on the workpiece W. Furthermore, after the hole is drilled on the workpiece W, air is blown or coolant is discharged to the workpiece W to remove shavings from the measurement area (surface) of the workpiece W. After the shavings are removed in this manner, the measurement described in the present embodiment and modifications may be performed. Accordingly, in the machining process, the measurement can be performed with the door kept closed. In this case, the machine tool includes an air blower that blows air to the workpiece W, or a coolant discharge unit that can discharge coolant toward the workpiece W.
The door can be opened to change the tools. During the process of machining, an image of the workpiece W can be imaged by the imaging unit 600 to extract the edge and surface of the workpiece W from the image as feature points. For example, machining may be temporarily stopped with the door closed, and the workpiece W may be imaged by the imaging unit 600 without opening the door. Thereafter, without opening the door, a correction value of the tool can be modified on the operation panel and the machining can be resumed. It is possible not only to modify the correction value of the tool on the operation panel, but also to change the rotational speed of the tool and other parameters.
The upper section of
A machine tool 800 includes a machining unit 820, a tool changer 850, a PLC (Programmable Logic Controller) 860, and a numerical control device 880. The machining unit 820 performs machining on a workpiece. The machining unit 820 includes an attachment unit 822 to which a tool or the imaging unit 600 is attached, and a drive unit 824 such as a servo motor that rotates the shaft. In a case where the machine tool 800 is a machining center, the main shaft 100 corresponds to the attachment unit 822, and the servo motor that rotates the main shaft 100 corresponds to the drive unit 824. The drive unit 824 of the machining center also operates to move the attachment unit 822 (the main shaft) having the imaging unit 600 attached thereto. In a case where the machine tool 800 is a turning center, the turret corresponds to the attachment unit 822. The machine tool 800 includes the drive unit 824 that moves the turret in addition to the servo motor that rotates the rotational shaft having a workpiece attached thereto. The drive unit 824 of the turning center also operates to move the attachment unit 822 (the turret) having the imaging unit 600 attached thereto. The numerical control device 880 executes the NC program to operate the machining unit 820 of the machine tool 800 in accordance with the codes specified in the NC program. The numerical control device 880 also provides an instruction to change the tool or the imaging unit 600 in the process of executing the NC program. The machining unit 820 includes a tool magazine capable of accommodating therein many tools and the imaging unit 600 to be attached to the attachment unit.
The external computer 400 (a mode of the information processing device) is connected to the machine tool 800 through a communication line. Furthermore, the external computer 400 is capable of communicating with the imaging unit 600. In a case of performing wired communication, the external computer 400 and the imaging unit 600 are connected by a communication line running through the inside of the attachment unit 822. Wireless communication via a wireless LAN, etc. may also be performed between the external computer 400 and the imaging unit 600. An image imaged by the imaging unit 600 is transmitted from the attachment unit 822 through wired communication to the external computer 400. An image imaged by the imaging unit 600 may be transmitted to the external computer 400 through wireless communication. The external computer 400 is a mode of the information processing device.
A functional block of the external computer 400 is now described. Each constituent element of the external computer 400 is implemented by hardware and software. The hardware includes computation devices such as a CPU (Central Processing Unit) and various coprocessors, storage devices such as a memory and a storage, and a wired or wireless communication line connecting these devices. The software is stored in the storage device and supplies a processing command to the computation devices. A computer program may be made up of a device driver, an operating system, various application programs on the upper layer of these device driver and operating system, and a library that provides the common functions to these application programs. Blocks to be described below do not refer to configurations in units of hardware but to blocks in units of functions.
The external computer 400 includes a user-interface processing unit 410, a data storage unit 440, a communication unit 450, and a computation unit 486. The user-interface processing unit 410 is configured to perform user interface processing via a display, a keyboard, a mouse, a touch sensor, a touch panel in which the display and the touch sensor are integrated, and the like included in the external computer 400. The data storage unit 440 stores therein various types of data. The data storage unit 440 is implemented by, for example, a RAM, a ROM, a flash memory, an SSD (Solid State Device), a hard disk, or other storage device, or is implemented by an appropriate combination thereof. The communication unit 450 is configured to perform communication processing to communicate with the machine tool 800 and the imaging unit 600. The computation unit 486 performs various types of processing based on the data acquired by the communication unit 450 and the data stored in the data storage unit 440. The computation unit 486 also functions as an interface of the user-interface processing unit 410, the data storage unit 440, and the communication unit 450.
The user-interface processing unit 410 includes an input unit 420 that inputs data through a user's operation, and an output unit 430 that outputs data to be provided to the user.
The input unit 420 includes an operation reception unit 422 that receives a data input upon an event detected by an input device such as a keyboard, a mouse, a touch sensor, or a touch panel. The output unit 430 includes a display processing unit 432 that displays various types of screens on a display device such as the display or the touch panel.
The computation unit 486 detects an edge of the workpiece W from a two-dimensional image, and performs computation processing based on the edge detected in the image.
The computation unit 486 includes an edge detection unit 482, a coordinate conversion unit 488, a dimension calculation unit 490, a pattern matching unit 492, and a correction-value calculation unit 494. The edge detection unit 482 detects an edge of the workpiece W from a two-dimensional image. The coordinate conversion unit 488 converts the coordinates between the machine coordinates and the coordinates in the image (hereinafter, referred to as “imaging coordinates”) to each other. The coordinate conversion unit 488 may convert the coordinates between the machine coordinates and the coordinates that define the machine-processing position on the workpiece W (hereinafter, referred to as “machine-processing coordinates”) to each other, and may convert the coordinates between the imaging coordinates and the machine-processing coordinates to each other. The dimension calculation unit 490 calculates the dimension of the overall shape or partial shape (for example, a circular hole) of the workpiece W (the dimension is, for example, the diameter of the circular hole). The computation unit 486 is a mode of the information processing device. The external computer 400 in its entirety (including the computation unit 486) may be regarded as the information processing device, or only the computation unit 486 itself may be regarded as the information processing device. The computation unit 486 may also be operated by an independent hardware resource. For example, the computation unit 486 may use a microcomputer including a processor and a memory as a hardware resource. The functions of the computation unit 486 may be implemented by storing a program defining the processing performed by the functions of the computation unit 486 in the memory of the microcomputer, and executing the program by the processor of the microcomputer.
The pattern matching unit 492 performs pattern matching on the shape of an edge detected in an image. The pattern matching unit 492 is described later in a second modification. The correction-value calculation unit 494 calculates a correction value for correcting the placement error of the workpiece W. The correction-value calculation unit 494 is described later in a fourth modification and a fifth modification.
The data storage unit 440 includes a pattern storage unit 442 that stores therein data about a plurality of patterns that are used in the pattern matching unit 492.
The communication unit 450 includes a transmission unit 460 that transmits data to the machine tool 800 and the imaging unit 600, and a reception unit 470 that receives data from the machine tool 800 and the imaging unit 600.
The transmission unit 460 includes a transmission unit 262 that transmits a correction value for correcting the placement error of the workpiece W to the machine tool 800.
The machine tool 800 includes an operation panel 890. The operation panel 890 includes, for example, a display unit 892 such as a liquid crystal display, and hardware resources such as a processor and a memory. On the display unit 892 on the operation panel 890, processing results of the information processing device such as the external computer 400 may be displayed. For example, the external computer 400 transmits the processing results to the operation panel 890, and the operation panel 890 receives the processing results and displays them on the display unit 892.
The lower section of
The machine tool 800 may include the computation unit 486, the data storage unit 440, and the communication unit 450 that are illustrated in the upper section of
When the tool or the imaging unit 600 attached to the attachment unit 822 is changed, the numerical control device 880 transmits a tool change command to the PLC 860. When receiving the tool change command, the PLC 860 controls the tool changer 850 in accordance with the received command to replace the tool attached to the main shaft 100 with a tool accommodated in the tool magazine of the tool changer 850. In this example, in the same manner as the tool, the imaging unit 600 is attached to or detached from the attachment unit 822 by the tool changer 850.
In accordance with the machining drawing, machining is performed on the workpiece W. The top surface of the workpiece W in this example is a square with each side being 100 mm. The machining drawing illustrates the machine-processing coordinates on the top surface of the workpiece W when viewed from the imaging unit 600. The machine-processing coordinates refer to the coordinates that define the machine-processing position for the workpiece W. The origin of the machine-processing coordinate system is referred to as “machine-processing origin”. The machine-processing origin is expressed as (Xp=0 mm, Yp=0 mm). The machine-processing origin in this example is the lower left corner point of the workpiece W. The rightward direction in the machining drawing shows the positive direction of Xp. The upward direction in the machining drawing shows the positive direction of Yp. A position on the top surface of the workpiece W is defined by the machine-processing coordinates (Xp, Yp).
The example of this machining drawing shows that machining is performed to drill two circular holes on the top surface of the workpiece W. A hole with a diameter of 4 mm (hereinafter, referred to as “first hole”) is drilled with its center at (Yp=30 mm, Xp=30 mm) on the lower left side. In addition, a hole with a diameter of 15 mm (hereinafter, referred to as “second hole”) is drilled with its center at (Xp=70 mm, Yp=70 mm) on the upper right side.
In the machining drawing, the position of the machine-processing origin (Xp=0 mm, Yp=0 mm) is specified by predetermined machine coordinates as a condition for placing the workpiece W. That is, the workpiece W is placed in such a manner that the machine-processing origin is positioned at the predetermined machine coordinates. In the descriptions below, the machine coordinates of the machine-processing origin are assumed to be (Xm=100, Ym=100).
As another condition for placing the workpiece W, the orientation of the workpiece W is specified. The direction of Xp of the machine-processing coordinates is aligned with the direction of Xm of the machine coordinates. The direction of Yp of the machine-processing coordinates is aligned with the direction of Ym of the machine coordinates.
On the premise of the machining drawing and the conditions for placing described above, an imaging condition for measuring the machining results is determined. Specifically, the position of the center of an image (referred to as “imaging center”) is determined as an imaging condition. For example, in the trial production phase, a worker may determine the imaging condition at any timing during the machining process. For another example, in the mass production phase, an imaging condition may be determined in advance during the planning of the production process.
After drilling of the first hole and the second hole, the machining system measures the first hole and the second hole formed by the drilling. As a condition for imaging the first hole, the center of the first hole (Xp=30 mm, Yp=30 mm) is set as an imaging center A. The imaging region is the inside of the rectangle defined by the thin lines around the imaging center A as illustrated. Since the first hole in its entirety falls inside the imaging region, measurement of the first hole is completed by a single imaging. In a case where a measurement target in its entirety falls inside the imaging region as described above, the measurement is referred to as “on-screen measurement”.
When the machine coordinates of the imaging center A are expressed as (Xm (imaging center A), Ym (imaging center A)), Xm (imaging center A) is calculated as 100 mm+30 mm=130 mm, and Ym (imaging center A) is calculated as 100 mm+30 mm=130 mm.
Since the second hole does not fit within the imaging region, a plurality of areas of the second hole are imaged. In this example, two areas are imaged, and based on the two images, the measurement of the second hole is performed. In a case where a measurement target in its entirety does not fit within the imaging region as described above, the measurement is referred to as “offscreen measurement”.
As a condition for the first imaging of the second hole, a predetermined position (Xp=66 mm, Yp=66 mm) on the lower left side relative to the center of the second hole is set as an imaging center B. The first imaging region is the inside of the rectangle defined by the thin lines around the imaging center B illustrated. As a condition for the second imaging of the second hole, a predetermined position (Xp=74 mm, Yp=74 mm) on the upper right side relative to the center of the second hole is set as an imaging center C. The second imaging region is the inside of the rectangle defined by the thin lines around the imaging center C illustrated.
When the machine coordinates of the imaging center B are expressed as (Xm (imaging center B), Ym (imaging center B)), Xm (imaging center B) is calculated as 100 mm+66 mm=166 mm, and Ym (imaging center B) is calculated as 100 mm+66 mm=166 mm. When the machine coordinates of the imaging center C are expressed as (Xm (imaging center C), Ym (imaging center C)), Xm (imaging center C) is calculated as 100 mm+74 mm=174 mm, and Ym (imaging center C) is calculated as 100 mm+74 mm=174 mm.
Next, measurement of the first hole is described.
The top diagram of
The image in this example has a size of 1600×1200 pixels. The resolution in this example is 4.5 μm/pixel. Therefore, the image covers a region of 5.4 mm×7.2 mm in actual size. The origin of the image system (referred to as “image origin”) is the lower left corner point. The rightward direction shows the positive direction of Xi. The upward direction shows the positive direction of Yi. The image origin is expressed as (Xi=0 pixels, Yi=0 pixels).
The illustrated circle shows the edge (boundary line) detected by the edge detection unit 482 from the imaged image with the optical axis of the imaging unit 600 focusing on the imaging center A. The edge represents the contour of the first hole. The external computer 400 (an example of the information processing device) measures the positions at two points on the edge line to obtain the diameter of the first hole machined. Hereinafter, the point on the edge line is referred to as “edge point”. An edge point a1 is the lower left point of the circle at which the tangent line forms a predetermined angle (135 degrees in this example) counterclockwise relative to the Xi-axis. An edge point a2 is the upper right point of the circle at which the tangent line forms a predetermined angle (135 degrees in this example) counterclockwise relative to the Xi-axis. The edge detection unit 482 identifies the tangent lines at the predetermined angle described above through image analysis, and detects the tangent points where the tangent line overlaps the edge as the edge point a1 and the edge point a2.
The image coordinates of the edge point a1 are expressed as (Xi (edge point a1), Yi (edge point a1)). The coordinate conversion unit 488 converts the image coordinates of the edge point a1 detected by the edge detection unit 482 to machine coordinates in accordance with the procedure below. The machine coordinates of the edge point a1 are expressed as (Xm (edge point a1), Ym (edge point a1)).
Xm (edge point a1) is calculated by the following Equation 1.
Xm(edge point a1)=Xm(imaging centerA)+Rx(Xi(edge point a1)−Xi(imaging centerA)) (Equation 1)
Xm (imaging center A) is equal to 130 mm. R represents the resolution (4.5 μm/pixel). Xi (edge point a1) is detected as an image coordinate of the tangent point through image analysis. Xi (imaging center A) is equal to 800 pixels.
Ym (edge point a1) is calculated by the following Equation 2.
Ym(edge point a1)=Ym(imaging centerA)+Rx(Yi(edge point a1)−Yi(imaging centerA)) (Equation 2)
Ym (imaging center A) is equal to 130 mm. R represents the resolution (4.5 μm/pixel). Yi (edge point a1) is detected as an image coordinate of the tangent point through image analysis. Yi (imaging center A) is equal to 600 pixels.
The coordinate conversion unit 488 converts the image coordinates of the edge point a2 (Xm (edge point a2), Ym (edge point a2)) detected by the edge detection unit 482 to machine coordinates in accordance with the procedure below. The machine coordinates of the edge point a2 are expressed as (Xm (edge point a2), Ym (edge point a2)).
Xm (edge point a2) is calculated by the following Equation 3.
Xm(edge point a2)=Xm(imaging centerA)+Rx(Xi(edge point a2)−Xi(imaging centerA)) (Equation 3)
Xm (imaging center A) is equal to 130 mm. R represents the resolution (4.5 μm/pixel). Xi (edge point a2) is detected as an image coordinate of the tangent point through image analysis. Xi (imaging center A) is equal to 800 pixels.
Ym (edge point a2) is calculated by the following Equation 4.
Ym(edge point a2)=Ym(imaging centerA)+Rx(Yi(edge point a2)−Yi(imaging centerA)) (Equation 4)
Ym (imaging center A) is equal to 130 mm. R represents the resolution (4.5 μm/pixel). Yi (edge point a2) is detected as an image coordinate of the tangent point through image analysis. Yi (imaging center A) is equal to 600 pixels.
The dimension calculation unit 490 calculates the distance between the machine coordinates of the edge point a1 (Xm (edge point a1), Ym (edge point a1)) and the machine coordinates of the edge point a2 (Xm (edge point a2), Ym (edge point a2)). This distance is the diameter of the first hole. The dimension calculation unit 490 calculates the machine coordinates of the midpoint between the machine coordinates of the edge point a1 and the machine coordinates of the edge point a2. The machine coordinates of the midpoint shows the position of the center of the first hole. The dimension calculation unit 490 may obtain the radius of the first hole by dividing the diameter of the first hole by two.
The dimension calculation unit 490 may calculate the diameter of the first hole based on the distance between the image coordinates without converting the image coordinates to machine coordinates. In that case, the dimension calculation unit 490 calculates the pixel length between the image coordinates of the edge point a1 (Xi (edge point a1), Yi (edge point a1)) and the image coordinates of the edge point a2 (Xi (edge point a2), Yi (edge point a2)), and multiplies the pixel length by the resolution (4.5 μm/pixel) to calculate the diameter of the first hole. The diameter of the first hole can also be calculated based on an edge point a3 and an edge point a4 in the same manner as described above. The edge point a3 is the lower right point of the circle at which the tangent line forms a predetermined angle (45 degrees in this example) counterclockwise relative to the Xi-axis. The edge point a4 is the upper left point of the circle at which the tangent line forms a predetermined angle (45 degrees in this example) counterclockwise relative to the Xi-axis. If the diameter calculated based on the edge points a1 and a2 does not correspond with the diameter calculated based on the edge points a3 and a4, the average value of the calculated diameters may be used, or the imaging process may be restarted. If noise included in the image affects the measurement, the non-correspondence may be resolved by imaging an image again.
Subsequently, measurement of the second hole is described.
The top diagram of
The illustrated arc shows the edge (boundary line) detected by the edge detection unit 482 from the imaged image with the optical axis of the imaging unit 600 focusing on the imaging center B. The edge represents the lower left arc of the contour of the second hole. The external computer 400 (an example of the information processing device) measures the position of an edge point b on the edge line. The edge point b is on the lower left arc at which the tangent line forms a predetermined angle (135 degrees in this example) counterclockwise relative to the Xi-axis. The edge detection unit 482 identifies the tangent line at the predetermined angle described above through image analysis, and detects the tangent point where the tangent line overlaps the edge as the edge point b.
The image coordinates of the edge point b are expressed as (Xi (edge point b), Yi (edge point b)). The coordinate conversion unit 488 converts the image coordinates of the edge point b detected by the edge detection unit 482 to machine coordinates in accordance with the procedure below. The machine coordinates of the edge point b are expressed as (Xm (edge point b), Ym (edge point b)).
Xm (edge point b) is calculated by the following Equation 5.
Xm(edge point b)=Xm(imaging center B)+Rx(Xi(edge point b)−Xi(imaging center B)) (Equation 5)
Xm (imaging center B) is equal to 166 mm. R represents the resolution (4.5 μm/pixel). Xi (edge point b) is detected as an image coordinate of the tangent point through image analysis. Xi (imaging center B) is equal to 800 pixels.
Ym (edge point b) is calculated by the following Equation 6.
Ym(edge point b)=Ym(imaging center B)+Rx(Yi(edge point b)−Yi(imaging center B)) (Equation 6)
Ym (imaging center B) is equal to 166 mm. R represents the resolution (4.5 μm/pixel). Yi (edge point b) is detected as an image coordinate of the tangent point through image analysis. Yi (imaging center B) is equal to 600 pixels.
The bottom diagram of
The illustrated arc shows the edge (boundary line) detected by the edge detection unit 482 from the imaged image with the optical axis of the imaging unit 600 focusing on the imaging center C. The edge represents the upper right arc of the contour of the second hole. The external computer 400 (an example of the information processing device) measures the position of an edge point c on the edge line. The edge point c is on the upper right arc at which the tangent line forms a predetermined angle (135 degrees in this example) counterclockwise relative to the Xi-axis. The edge detection unit 482 detects the edge point c through image analysis.
The image coordinates of the edge point c are expressed as (Xi (edge point c), Yi (edge point c)). The coordinate conversion unit 488 converts the image coordinates of the edge point c detected by the edge detection unit 482 to machine coordinates in accordance with the procedure below. The machine coordinates of the edge point c are expressed as (Xm (edge point c), Ym (edge point c)).
Xm (edge point c) is calculated by the following Equation 7.
Xm(edge point c)=Xm(imaging center C)+Rx(Xi(edge point c)−Xi(imaging center C)) (Equation 7)
Xm (imaging center C) is equal to 174 mm. R represents the resolution (4.5 μm/pixel). Xi (edge point c) is detected as an image coordinate of the tangent point through image analysis. Xi (imaging center C) is equal to 800 pixels.
Ym (edge point c) is calculated by the following Equation 8.
Ym(edge point c)=Ym(imaging center C)+Rx(Yi(edge point c)−Yi(imaging center C)) (Equation 8)
Ym (imaging center C) is equal to 174 mm. R represents the resolution (4.5 μm/pixel). Yi (edge point c) is detected as an image coordinate of the tangent point through image analysis. Yi (imaging center C) is equal to 600 pixels.
The dimension calculation unit 490 calculates the distance between the machine coordinates of the edge point b (Xm (edge point b), Ym (edge point b)) and the machine coordinates of the edge point c (Xm (edge point c), Ym (edge point c)). This distance is the diameter of the second hole. The dimension calculation unit 490 calculates the machine coordinates of the midpoint between the machine coordinates of the edge point b and the machine coordinates of the edge point c. The machine coordinates of the midpoint shows the position of the center of the second hole. The dimension calculation unit 490 may obtain the radius of the first hole by dividing the diameter of the second hole by two.
In the first example, the timing of the measurement process by imaging is defined in the NC program. This sequence is used, for example, in the mass production phase.
At the stage of completion of predetermined machining (drilling of the first hole or the second hole in this example) in accordance with the NC program, the illustrated sequence is entered.
The numerical control device 880 instructs the PLC 860 to attach the imaging unit 600 (S10). In accordance with the instruction, the PLC 860 controls the tool changer 850 to replace the tool attached to the attachment unit 822 (the main shaft 100) with the imaging unit 600 accommodated in the tool magazine of the tool changer 850, and attach the imaging unit 600 to the attachment unit 822 (the main shaft 100).
The numerical control device 880 instructs the drive unit 824 to move the attachment unit 822 (the main shaft 100) such that the optical axis of the imaging unit 600 focuses on the machine coordinates of a predetermined imaging center (for example, the imaging center A, the imaging center B, or the imaging center C described above) (S12). The drive unit 824 moves the attachment unit 822 (the main shaft 100) in accordance with the instruction. It is assumed that the center axis of the attachment unit 822 (the main shaft 100) coincides with that of the imaging unit 600.
The numerical control device 880 transmits an imaging instruction to the external computer 400 (an example of the information processing device) (S14). When the reception unit 470 in the external computer 400 receives the imaging instruction, the transmission unit 460 in the external computer 400 transmits the imaging instruction to the imaging unit 600 (S16).
When receiving the imaging instruction, the imaging unit 600 images an image of the workpiece (S18). The imaging unit 600 transmits the image obtained to the external computer 400 (an example of the information processing device) (S20).
The numerical control device 880 transmits the machine coordinates of the imaging center focused in S12 to the external computer 400 (an example of the information processing device) (S22). The reception unit 470 in the external computer 400 receives the machine coordinates of the imaging center and the image. The edge detection unit 482 in the external computer 400 (an example of the information processing device) detects the edge in the image through image analysis (S24). The edge detection unit 482 detects the image coordinates of a predetermined edge point (for example, the edge point a1, the edge point a2, the edge point b, or the edge point c described above) through image analysis (S26). The coordinate conversion unit 488 converts the image coordinates of the detected edge point to machine coordinates (S28).
In a case of the above first hole, an image is imaged once, and thus the sequence of processes within the broken line is performed once. Then, two edge points are detected from a single image. In contrast, an image of the second hole is imaged twice, and a single edge point is detected from each image. Accordingly, the sequence of processes within the broken line is performed twice.
The dimension calculation unit 490 in the external computer 400 (an example of the information processing device) calculates the diameter of the first hole and the diameter of the second hole individually based on the machine coordinates of the two edge points (S30). Furthermore, the dimension calculation unit 490 calculates the machine coordinates of the center of the first hole and the machine coordinates of the center of the second hole. The display processing unit 432 in the external computer 400 or the operation panel 890 displays the image and the dimensions measured (the diameter and the center) on the screen of the display device. The external computer 400 (an example of the information processing device) may calculate and display the radius of the first hole and the radius of the second hole.
In the second example, a worker operates the external computer 400 or the operation panel 890 to perform the measurement process by imaging at any timing. This sequence is used, for example, in the trial production phase.
The display processing unit 432 in the external computer 400 or the operation panel 890 displays the machining drawing on the screen of the display device. The machining drawing is stored in the data storage unit 440. Referring to the machining drawing, the worker specifies the machine-processing coordinates of the imaging center. The worker may input numerical values of the machine-processing coordinates, or may input the machine-processing coordinates by clicking or touching the machining drawing on the display. Upon this user's operation, the operation reception unit 422 in the external computer 400 or the operation panel 890 receives the machine-processing coordinates of the imaging center (S40).
The coordinate conversion unit 488 in the external computer 400 (an example of the information processing device) converts the machine-processing coordinates of the imaging center to machine coordinates of the imaging center. Specifically, the machine coordinates of the imaging center are obtained by adding the machine-processing coordinates of the imaging center to the machine coordinates of the machine-processing origin (S42).
The external computer 400 (an example of the information processing device) instructs the numerical control device 880 to attach the imaging unit 600 (S44). This process and operation are the same as those in S10.
The external computer 400 (an example of the information processing device) instructs the numerical control device 880 to move the attachment unit 822 (the main shaft 100) to the machine coordinates of the imaging center (S46). At this time, the machining system operates in the same manner as in S12. The process and operation in S48 to S60 are the same as those in S16 to S20 and in S24 to S36.
Other than the first and second examples, a worker may manually actuate the attachment unit 822 (the main shaft 100) to move the attachment unit 822 (the main shaft 100) above a workpiece. In this method, the external computer 400 (an example of the information processing device) acquires the machine coordinates of the imaging center from the machine tool 800. The machine coordinates of the imaging center are at the position of the center axis of the attachment unit 822 (the main shaft 100).
In the embodiment, an example of the on-screen measurement has been described in which the positions of two points on the edge are measured, and based on the positions of the two measurement points, the diameter of the circular hole is calculated. In the on-screen measurement, the positions of three or more points on the edge may also be measured, and based on the positions of the three or more measurement points, the diameter of the circular hole may be calculated. In the on-screen measurement, three-point measurement has higher measurement accuracy than two-point measurement. In the on-screen measurement, the number of measurement points may be four or more. As the number of measurement points increases, measurement accuracy further improves.
The bottom diagram in
In this example, the edge detection unit 482 detects the image coordinates of each of the edge points in the on-screen measurement. The edge points are a tangent point on the tangential line forming an angle of 0 degree relative to the Xi-axis of the image system, a tangent point on the tangential line forming an angle of 120 degrees relative to the Xi-axis of the image system, and a tangent point on the tangential line forming an angle of 240 degrees relative to the Xi-axis of the image system. The dimension calculation unit 490 calculates the diameter of the circular hole based on the image coordinates of these three edge points through geometric analysis. Otherwise, the coordinate conversion unit 488 converts the image coordinates of each edge point to machine coordinates, and the dimension calculation unit 490 calculates the diameter of the circular hole based on the machine coordinates of the three edge points through geometric analysis.
The top diagram in
In this example, the edge detection unit 482 detects the image coordinates of each of the edge points that are 12 tangent points in the on-screen measurement. The dimension calculation unit 490 calculates the diameter of the circular hole based on the image coordinates of these 12 edge points through geometric analysis. Otherwise, the coordinate conversion unit 488 converts the image coordinates of each of the 12 edge points to machine coordinates, and the dimension calculation unit 490 calculates the diameter of the circular hole based on the machine coordinates of the 12 edge points through geometric analysis. At this time, the diameter of the circular hole is calculated by using the distance from the center to each edge point excluding the outlier point equal to or greater than the standard deviation. This can improve the calculation accuracy.
The bottom diagram in
In the embodiment, an example of the offscreen measurement has been described in which the positions of two points on the edge are measured in two images, and based on the positions of the two measurement points, the diameter of the circular hole is calculated. In the offscreen measurement, the positions of three or more points on the edges may also be measured in three or more images, and based on the positions of the three or more measurement points, the diameter of the circular hole may be calculated. In the offscreen measurement, three-point measurement has higher measurement accuracy than two-point measurement. In the offscreen measurement, the number of measurement points may be four or more. As the number of measurement points increases, measurement accuracy further improves.
In a case of offscreen measurement in which three-point measurement is performed, images of three areas are imaged. The three areas are the region including a tangent point on the tangential line forming an angle of 0 degree relative to the Xi-axis of the image system, the region including a tangent point on the tangential line forming an angle of 120 degrees relative to the Xi-axis of the image system, and the region including a tangent point on the tangential line forming an angle of 240 degrees relative to the Xi-axis of the image system. The edge detection unit 482 detects the image coordinates of the respective edge points in the three images. The coordinate conversion unit 488 converts the image coordinates of each of the edge points to machine coordinates. The dimension calculation unit 490 calculates the diameter of the circular hole based on the machine coordinates of the three edge points through geometric analysis.
For example, in a case of offscreen measurement in which 12-point measurement is performed, images of 12 areas are imaged, each of which includes each of the 12 tangent points. The edge detection unit 482 detects the image coordinates of the edge points in the 12 images. The coordinate conversion unit 488 converts the image coordinates of each of the edge points to machine coordinates. The dimension calculation unit 490 calculates the diameter of the circular hole based on the machine coordinates of the 12 edge points through geometric analysis. In the same manner as in the on-screen measurement, the diameter of the circular hole is calculated by using the distance from the center to each edge point excluding the outlier point equal to or greater than the standard deviation. This can improve the calculation accuracy. In a case of offscreen measurement in which 36-point measurement is performed, the processing is performed in the same manner as the processing when 12-point measurement is performed, except that the number of measurement points is different from that in the 12-point measurement. When the distance between edge points is short, two edge points may be photographed together in a single imaging, so that the number of times of imaging is reduced.
In the embodiment, an example has been described in which the diameter or radius of the circular hole is calculated by using the distance between tangent points. However, the diameter or radius of the circular hole may be determined by pattern matching.
In the second modification, many circle pattern images with slightly different diameters are stored in the pattern storage unit 442 according to the measurement accuracy. The pattern matching unit 492 reads the circle pattern images one by one, and performs pattern matching between the circle pattern image and the edge of the image. A pattern image, from which the highest degree of matching is obtained, is determined, and the diameter (or radius) of the circle in the determined pattern image is defined as a diameter (or a radius) of the circular hole. The pattern matching unit 492 defines the center position of the circle in the pattern image showing the highest degree of matching as the center position of the circular hole.
Not only the circle pattern images, but also many pattern images may be prepared in the pattern storage unit 442 according to the shape of a portion to be measured, such that the pattern matching unit 492 may determine a pattern image that shows the highest degree of matching with the edge by means of pattern matching. In this case, the dimension of the shape in the determined pattern image is regarded as a dimension of the shape of the portion to be measured.
In consideration of the case where burrs generated along a circular hole can possibly be detected as an edge, the edge detection unit 482 may detect the contour by using a Sobel filter or other type of filter, and compare the detected contour with an ideal circular shape so as to locate an irregular area such as burrs. Then, the edge detection unit 482 may identify the edge point, excluding the irregular area.
In the embodiment, an example has been described in which the machining results are measured. In a fourth modification, an example is described in which a positioning error in placing the workpiece W prior to machining is corrected. The position where the workpiece W is placed on the table is referred to as “placement position of the workpiece W”. The placement position of the workpiece W can be determined by a predetermined reference point of the workpiece W. This example shows that the predetermined reference point is defined as a machine-processing origin, and based on this machine-processing origin, the placement position of the workpiece W is determined. That is, the deviation between the machine-processing origin of the workpiece W placed in an ideal state and the machine-processing origin of the workpiece W that is actually placed corresponds to “a deviation of placement position of the workpiece W”. An example is described below in which the external computer 400 (an example of the information processing device) uses an image to measure the deviation of the machine-processing origin.
In relation to
As an imaging condition for imaging the machine-processing origin, the machine-processing origin (Xp=0 mm, Yp=0 mm) is defined as an imaging center D. The imaging region is the inside of the rectangle defined by the thin lines around the imaging center D illustrated.
The top diagram of
The top diagram of
The broken-line square shows the position and orientation of the workpiece W that meet the specified imaging conditions. That is, it shows the position and orientation of the workpiece W placed in an ideal state. The NC program runs on the assumption that the workpiece W is correctly placed at this position and in this orientation.
In contrast, the solid-line square shows the position and orientation of the workpiece W that is actually placed. Since errors can occur in the placement process, there is a possibility that the position and orientation of the solid-line square may vary each time the workpiece W is changed. The top diagram of
In this example, the orientation of the workpiece W meets the imaging condition. That is, the workpiece W is correctly oriented. However, the position of the workpiece W is displaced toward the upper right direction. Specifically, the machine-processing origin of the workpiece W placed is supposed to be at the original machine coordinates (Xm=0 mm, Ym=0 mm). However, the actual machine-processing origin is at the machine coordinates (Xm=0.9 mm, Ym=0.9 mm). Hereinafter, the machine-processing origin specified as a condition for placing the workpiece in the machining drawing is referred to as “ideal machine-processing origin”, and the machine-processing origin of the workpiece W placed is referred to as “actual machine-processing origin”.
If machining is performed as is, the positions of the first hole and the second hole are displaced toward the lower left. The machine tool 800 has a function of correcting the positional deviation of the machine-processing origin in such a case. Specifically, when an offset value (Xoff=0.9 mm, Yoff=0.9 mm) for correcting the positional deviation is set for the machine tool 800, drilling can be performed as shown in the machining drawing. Hereinafter, the external computer 400 (an example of the information processing device) analyzes an image and calculates the offset value (Xoff=0.9 mm, Yoff=0.9 mm).
The top diagram of
Two broken lines show the position and orientation of the workpiece W that meet the imaging conditions in the same manner as illustrated in the top diagram of
Xm (actual machine-processing origin) is calculated by the following Equation 9.
Xm(actual machine-processing origin)=Xm(imaging centerD)+Rx(Xi(actual machine-processing origin)−Xi(imaging centerD)) (Equation 9)
Xm (imaging center D) is equal to 100 mm. R represents the resolution (4.5 μm/pixel). Xi (actual machine-processing origin) is determined as the image coordinates of the intersection of the edge lines through image analysis. Xi (imaging center D) is equal to 800 pixels.
Ym (actual machine-processing origin) is calculated by the following Equation 10.
Ym(actual machine-processing origin)=Ym(imaging center D)+Rx(Yi(actual machine-processing origin)−Yi(imaging center D)) (Equation 10)
Ym (imaging center D) is equal to 100 mm. R represents the resolution (4.5 μm/pixel). Yi (actual machine-processing origin) is determined as the image coordinates of the intersection of the edge lines through image analysis. Yi (imaging center D) is equal to 600 pixels.
In this example, the image coordinates of the actual machine-processing origin (Xi (actual machine-processing origin)=1000, Yi (actual machine-processing origin)=800) are converted to machine coordinates of the actual machine-processing origin (Xm (actual machine-processing origin)=100.9 mm, Ym (actual machine-processing origin)=100.9 mm).
The correction-value calculation unit 494 calculates an offset value to be transmitted to the machine tool 800 from the machine coordinates of the actual machine-processing origin obtained by the coordinate conversion unit 488. The offset value is a type of parameter (correction value) that is set to correct the motion in the machine tool 800. Specifically, an offset value (Xoff, Yoff) is obtained by subtracting the machine coordinates of the ideal machine-processing origin from the machine coordinates of the actual machine-processing origin. In this example, the offset value is calculated as follows.
Xoff=Xm(actual machine-processing origin)−Xm(ideal machine-processing origin)=100.9 mm−100 mm=0.9 mm
Yoff=Ym(actual machine-processing origin)−Ym(ideal machine-processing origin)=100.9 mm−100 mm=0.9 mm
The offset value intends to associate the machine-processing coordinates (Xp, Yp) (see
The first half of
This sequence is based on the sequence on the left side of
The edge detection unit 482 detects the image coordinates of the actual machine-processing origin through image analysis (S86). The coordinate conversion unit 488 converts the image coordinates of the actual machine-processing origin detected to machine coordinates (S88). The correction-value calculation unit 494 calculates an offset value to be set for the machine tool 800 from the machine coordinates of the actual machine-processing origin obtained by the coordinate conversion unit 488. Then, the correction-value transmission unit 468 transmits the offset value (a type of correction value) to the machine tool 800 (S90). Thereafter, the machine tool 800 performs appropriate machining positioning according to the offset value.
Although not illustrated, the sequence in the fourth modification may be based on the sequence on the right side of
In the fifth modification, in addition to the deviation of the machine-processing origin, a deviation of the orientation of the workpiece W is also measured. The orientation in which the workpiece W is placed on the table is referred to as “placement orientation of the workpiece W”. The placement orientation of the workpiece W can be determined by an angle of the line segment connecting the ends at two predetermined reference points of the workpiece W. That is, the angle formed by the line segment connecting the ends at two predetermined reference points of the workpiece placed in an ideal state, and the line segment connecting the ends at two predetermined reference points of the workpiece W that is actually placed corresponds to “a deviation of placement orientation of the workpiece W”. In this example, the machine-processing origin and the point diagonal to the machine-processing origin are defined as two predetermined reference points. Hereinafter, the point diagonal to the machine-processing origin may sometimes be simply referred to as “diagonal point”. The machine-processing origin is an example of “first reference point”. The point diagonal to the machine-processing origin is an example of “second reference point”. The machine coordinates of the machine-processing origin are an example of “first coordinates”. The machine coordinates of the point diagonal to the machine-processing origin are an example of “second coordinates”.
In relation to
The bottom diagram of
Similarly to the top diagram of
The machine tool 800 has a function of correcting the positional deviation of the machine-processing origin and the deviation of orientation of the workpiece W. That is, when an offset value (Xoff=0.9 mm, Yoff=0.9 mm) and a correction rotation angle (+1) for correcting the positional deviation are set for the machine tool 800, drilling is performed as shown in the machining drawing. Hereinafter, the external computer 400 (an example of the information processing device) calculates the offset value (Xoff=0.9 mm, Yoff=0.9 mm) and the correction rotation angle (+1) based on the imaged image of the machine-processing origin and the imaged image of the diagonal point.
The middle diagram of
Two broken lines show the ideal position and orientation of the workpiece W that meet the imaging conditions in the same manner as illustrated in the top diagram of
The bottom diagram of
As illustrated in the bottom diagram of
Xm (actual diagonal point) is calculated by the following Equation 11.
Xm(actual diagonal point)=Xm(imaging centerE)+Rx(Xi(actual diagonal point)−Xi(imaging centerE)) (Equation 11)
Xm (imaging center E) is equal to 200 mm. R represents the resolution (4.5 μm/pixel). Xi (actual diagonal point) is determined as the image coordinates of the intersection of the edge lines through image analysis. Xi (imaging center E) is equal to 800 pixels.
Ym (actual diagonal point) is calculated by the following Equation 12.
Ym(actual diagonal point)=Ym(imaging center E)+Rx(Yi(actual diagonal point)−Yi(imaging center E)) (Equation 12)
Ym (imaging center E) is equal to 200 mm. R represents the resolution (4.5 μm/pixel). Yi (actual diagonal point) is determined as the image coordinates of the intersection of the edge lines through image analysis. Yi (imaging center E) is equal to 600 pixels.
In this example, the image coordinates of the actual diagonal point (Xi (actual diagonal point)=608.67, Yi (diagonal point)=1184.44 are converted to machine coordinates of the actual diagonal point (Xm (actual diagonal point)=99.14 mm, Ym (actual diagonal point)=102.63 mm).
Subsequently, the correction-value calculation unit 494 calculates an offset value and a correction rotation angle. The offset value is calculated by the same method as that in the fourth modification. A method for calculating the correction rotation angle is now described. As illustrated in the bottom diagram of
tan(45+θ)=(Ym(actual diagonal point)−Ym(actual machine-processing origin))/(Xm(actual diagonal point)−Xm(actual machine-processing origin)) (Equation 13)
Then, Equation 14 is derived from Equation 13. The correction-value calculation unit 494 calculates the correction rotation angle θ by using Equation 14.
θ=arctan((Ym(actual diagonal point)−Ym(actual machine-processing origin))/(Xm(actual diagonal point)−Xm(actual machine-processing origin)))−45 (Equation 14)
The calculation result of Equation 14 in this example is shown below.
θ=arctan((102.630−0.9)/(99.139−0.9))−45=arctan(1.0355)−45=46−45=1
Next, the sequence in the fifth modification is described.
The second half of
This sequence is continued from the sequence illustrated in the first half of
The edge detection unit 482 detects the image coordinates of the actual diagonal point through image analysis (S114). The coordinate conversion unit 488 converts the image coordinates of the actual diagonal point detected to machine coordinates (S116). The correction-value calculation unit 494 calculates the correction rotation angle from the machine coordinates of the actual machine-processing origin and the machine coordinates of the actual diagonal point. Then, the correction-value transmission unit 468 transmits the correction rotation angle (a type of correction value) to the machine tool 800 (S118). Thereafter, the machine tool 800 performs appropriate machining positioning according to the offset value and the correction rotation angle.
The dimension calculation unit 490 may calculate the distance between the actual machine-processing origin and the actual diagonal point. When the calculated distance does not correspond with the distance in the machining drawing, the output unit 430 in the external computer 400 or the operation panel 890 may issue an alert indicating that the workpiece W is inappropriate.
In the fifth modification, two reference points may be used other than the machine-processing origin and the point diagonal to the machine-processing origin. For example, the lower right corner point may be used as the first reference point, while the upper left corner point may be used as the second reference point.
The external computer 400 (an example of the information processing device) may measure the deviation of orientation of the workpiece W by only using an imaged image with the machine-processing origin at the center. In the sixth modification, the slope of the edge line detected from the image is obtained.
The edge detection unit 482 obtains a correction rotation angle through image analysis. The correction rotation angle is formed by a boundary line (a thick solid line extending substantially in the horizontal direction in the middle diagram of
When the external computer 400 (an example of the information processing device) detects a deviation of the machine-processing origin of the workpiece W or a deviation of the orientation of the workpiece W through the operation shown in the fourth to sixth modifications, then the output unit 430 in the external computer 400 or the operation panel 890 may issue an alert indicating that there is a deviation of the machine-processing origin of the workpiece W or a deviation of the orientation of the workpiece W. This alert informs the worker of the deviation of the position or orientation of the workpiece W, so that the worker can correctly place the workpiece W back. Therefore, a machining defect due to inappropriate placement of the workpiece W can be prevented.
The embodiment and the above modifications have illustrated an example in which a circular hole is measured. The circular hole (a cylindrical recessed portion) is an example of the measurement target portion, that is, a portion of the workpiece W to be measured. The measurement target portion may have any other shape with a circle when viewed from the Z-axis direction, such as a conical recessed portion, a cylindrical portion (a cylindrical protruding portion), or a conical protruding portion. The measurement target portion may have a shape with lines extending in parallel when viewed from the Z-axis direction, such as a groove or a wall. The width of the shape with lines extending in parallel may be measured. The measurement target portion may have any other shape such as an elongated circle, an ellipse, or a polygon (for example, a rectangle or hexagon) when viewed from the Z-axis direction. Any other dimension such as the center, diameter, side length, diagonal-line length, and angle of the corner (the angle formed by two sides of the edge) may be measured. The measurement target portion may be a protruding portion, a recessed portion, or a sloped portion. The shape of the measurement target portion may be measured either before or after machining.
There is a case where the dimension calculation unit 490 determines that the dimension calculated based on the edge is different from a predetermined dimension of the workpiece W, referring to the data on the machining drawing stored in the data storage unit 440. In that case, the output unit 430 in the external computer 400 or the operation panel 890 may output an alert indicating that the workpiece W placed does not have a predetermined size or a predetermined shape. For example, an alert is output when the length of the side of the workpiece W calculated based on the edge does not correspond with the length of the side of the workpiece W included in the data on the machining drawing. For another example, an alert is output when the angle of the corner of the workpiece W calculated based on the edge does not correspond with the angle of the corner of the workpiece W included in the data on the machining drawing. That is, when the dimension calculation unit 490 determines that the workpiece W placed does not have a predetermined size or a predetermined shape based on the calculated dimension of the workpiece W, the output unit 430 in the external computer 400 or the operation panel 890 outputs an alert.
In the first example of the sequence (on the left side of
In a tenth modification, the external computer 400 (an example of the information processing device) transmits the machine coordinates of the imaging center to the numerical control device 880. In the present modification, the external computer 400 operates while reading a simple program created by a user (hereinafter, referred to as “measurement program”) in the computation unit 486. This measurement program is stored in the data storage unit 440. The user describes the machine coordinates of the imaging center into the measurement program. The external computer 400 reads the machine coordinates of the imaging center from the measurement program, and instructs the numerical control device 880 to move the attachment unit 822 to the position of the machine coordinates read.
The numerical control device 880 instructs the PLC 860 to attach the imaging unit 600 in the same manner as in S10 in
The numerical control device 880 instructs the drive unit 824 to move the attachment unit 822 in accordance with the received attachment unit movement instruction (S126). The drive unit 824 moves the attachment unit 822 such that the optical axis of the imaging unit 600 focuses on the machine coordinates of the imaging center.
The external computer 400 receives the machine coordinates of the attachment unit 822 from the numerical control device 880 as needed (S128). When the received machine coordinates of the attachment unit 822 corresponds with the machine coordinates of the imaging center, the external computer 400 determines that movement of the attachment unit 822 has completed, and transmits an imaging instruction to the imaging unit 600 (S130). The subsequent processes in S132 to S142 are the same as those in S18, S20, and S24 to S30 in
In the on-screen measurement, as an image is imaged only once, the sequence of processes within the broken line in
Focusing on the points in common between the above embodiment (
The imaging unit 600 images an image of the area to be measured once (S18 and S50 in
The dimension calculation unit 490 calculates an approximate circle from the detected edge points by means of the least squares method (S30 and S60 in
Next, the process flow of offscreen measurement is also summarized. The imaging unit 600 images images of the area to be measured in multiple times. Here,
The top diagram of
The edge detection unit 482 detects edge points at a pitch of 20 degrees on the arc in each of the image 1 and the image 2 (S24 and S54 in
The coordinate conversion unit 488 converts the edge positions from the form of image coordinates to the form of machine coordinates (S28 and S58 in
The edge detection unit 482 calculates an approximate circle from the edge points (machine coordinates) by means of the least squares method (S30 and S60 in
Next, an editing tool used for generating a measurement program is described.
The editing tool runs on the external computer 400.
The positioning image 714 is, for example, a live view image (on the upper left of the diagram). The live view image is an image currently being imaged by the imaging unit 600 attached to the attachment unit 822. The user can slide the live view image by operating a jog attached to the external computer 400 and thereby moving the imaging unit 600.
The positioning image 714 may be the whole or a portion of a stitching image. The stitching image is an image including the workpiece W in its entirety, and is made by connecting N×M images together, that are obtained from the imaging unit 600.
The positioning image 714 may be the whole or a portion of a CAD (Computer Aided Design) image. The CAD image is a graphic image generated from CAD data of a workpiece.
In a list 704, commands included in the measurement program are displayed. For example, with the fourth item “Circle” selected, the editing tool accepts parameters related to circle measurement. The machine coordinates corresponding to the center of an annular pattern 712 set in the positioning image 714 are written into the measurement program as the imaging center. The outer diameter and inner diameter of the annular pattern 712 are also written into the measurement program as parameters for determining the target region for edge detection. The editing tool may allow for setting of the machine coordinate in the Zm direction (height direction).
The editing tool may generate a stitching image by connecting live view images imaged plural times together, and may display the stitching image as the positioning image 714 on the editing screen 700.
The top diagram of
The display processing unit 432 displays a live view image imaged by the first imaging in the region of the positioning image 714. A first live view image 720 is displayed at a predetermined display magnification. Thereafter, the user operates the jog to move the imaging unit 600 for the second imaging. Then, the operation reception unit 422 receives an imaging instruction and the imaging unit 600 performs the second imaging.
The middle diagram of
The display processing unit 432 generates a stitching image by connecting a second live view image 722 with the first live view image 720, and displays the stitching image in the region of the positioning image 714. As illustrated in the middle diagram, in a case where an additional live view image 722 fits within the region of the positioning image 714 at the current display magnification, the display processing unit 432 does not change this display magnification. Further, the user moves the imaging unit 600 to perform the third imaging.
The bottom diagram of
The display processing unit 432 generates a stitching image by connecting a third live view image 724 with the first live view image 720 and the second live view image 722, and displays the stitching image in the region of the positioning image 714. When an additional live view image is displayed at the current display magnification in the state illustrated in the middle diagram of
In the bottom diagram of
The top diagram of
The bottom diagram of
The display processing unit 432 may change the display magnification such that a hole to be measured, included in a stitching image, can fit entirely within the region of the positioning image 714, and may display the whole or a portion of the stitching image on the display unit. In this case, it is allowable that the remaining portion of the live view image, other than the hole to be measured, does not fit within the region of the positioning image 714. For example, an image of the outside of the hole to be measured may extend over the region of the positioning image 714.
The display processing unit 432 acquires the first live view image 730 associated with first machine coordinates (for example, machine coordinates of the imaging center) by the first imaging. Based on the first machine coordinates, the display processing unit 432 displays the first live view image 730 such that the shape (contour line) of the hole to be measured, included in the first live view image 730, fits entirely within the region of the positioning image 714.
Next, the second imaging is performed. The display processing unit 432 acquires the second live view image 732 associated with second machine coordinates (same as above). Based on the first machine coordinates and the second machine coordinates, the display processing unit 432 displays a stitching image such that the shape of the hole to be measured, included in the first live view image 730 and in the second live view image 732, fits entirely within the region of the positioning image 714. The display processing unit 432 performs the same processing after the third imaging.
The shape of a hole to be measured is determined by, for example, machine coordinates of an edge line obtained by edge detection in a live view image. The display processing unit 432 may acquire information on the shape of the hole to be measured in advance. When the display processing unit 432 holds machine coordinates that determine the shape of the hole to be measured, it is permissible that the edge detection processing is not performed. The same applies to measurement targets other than a circular hole.
A supplemental explanation of the edge detection performed by the edge detection unit 482 (S24 and S54 in
The image illustrated in
The edge detection unit 482 generates a cutout image by rotating the image within the measurement range by using affine transformation and bilinear interpolation. The affine transformation is a type of coordinate conversion method. In this example, the image is rotated clockwise by 10 degrees. The bilinear interpolation is a type of pixel-value conversion method. In this example, the origin of the cutout image is defined as the lower left corner point, the rightward direction is defined as the positive direction of the Xc-axis, and the upward direction is defined as the positive direction of the Yc-axis. In this process, based on the position and pixel value of each pixel included in the measurement region, the pixel value of each pixel of the cutout image is calculated. Image degradation due to the rotation can be reduced by using the affine transformation and bilinear interpolation.
Focus is on the variations in brightness in the Xc-axis direction in order to locate the edge point. The edge detection unit 482 calculates a first-order differential value of the brightness (hereinafter, sometimes simply referred to as “first-order differential value”). The graph in
The measurement accuracy can be improved by interpolating the first-order differential values. The edge detection unit 482 calculates a spline interpolated value of the first-order differential values. The spline interpolated value is shown by the curve in the graph. The edge detection unit 482 defines the pixel Xc corresponding to a maximum value of the spline interpolated values as an edge point. As described above, the peak of the interpolated value corresponds to the position of the edge point in the cutout image. The pixel Xc corresponding to a negative minimum value of the spline interpolated values may sometimes be defined as an edge point. When the spline interpolated value is used in the manner as described above, a resolution of 0.02 pixel is achieved. This makes it possible to perform the measurement with high accuracy. The spline interpolation described in this example is a type of subpixel processing.
The edge detection unit 482 converts the coordinates (Xc, Yc) of the edge point in the cutout image to the image coordinates (Xi, Yi) of the edge point in the image. At this time, the coordinate conversion is performed by rotating the image reversely to the rotation using the affine transformation described above.
A supplemental explanation of the method for calculating the distance between edge points is given below.
In the above embodiment, the distance between edge points is calculated based on the difference in machine coordinates of the edge points. That is, the distance between edge points is calculated as the square root of the sum of squares of the distance in the Xm direction between the edge points (difference in Xm coordinates) and the distance in the Ym direction between the edge points (difference in Ym coordinates).
In the present modification, the distance between edge points is calculated based on the difference in image coordinates between the edge points and the difference in machine coordinates between the images. The edge point b and the edge point c illustrated in
First, the difference in image coordinates between edge points is described. When the spacing (pixel) in the Xi direction between edge points is expressed as Dx (edge point b, edge point c), the spacing in the Xi direction between edge points is calculated by the following Equation (15).
Dx(edge point b,edge point c)=Xi(edge point c)−Xi(edge point b) Equation (15)
Similarly, when the spacing (pixel) in the Yi direction between edge points is expressed as Dy (edge point b, edge point c), the spacing in the Yi direction between edge points is calculated by the following Equation (16).
Dy(edge point b,edge point c)=Yi(edge point c)−Yi(edge point b) Equation (16)
Next, the difference in machine coordinates between images (the difference in machine coordinates of the imaging center) is described. When the distance in the Xm direction between imaging centers is expressed as Lx (imaging center B, imaging center C), the distance in the Xm direction between imaging centers is calculated by the following Equation (17).
Lx(imaging center B,imaging center C)=Xm(imaging center C)−Xm(imaging center B) Equation (17)
Similarly, when the distance in the Ym direction between imaging centers is expressed as Ly (imaging center B, imaging center C), the distance in the Ym direction between imaging centers is calculated by the following Equation (18).
Ly(imaging center B,imaging center C)=Ym(imaging center C)−Ym(imaging center B) Equation (18)
Therefore, the distance in the Xm direction between edge points is calculated by the following Equation (19). R represents the resolution as described above.
Lx(edge point b,edge point c)=Lx(imaging center B,imaging center C)+RxDx(edge point b,edge point c) Equation (19)
The distance in the Ym direction between edge points is calculated by the following Equation (20).
Ly(edge point b,edge point c)=Ly(imaging center B,imaging center C)+RxDy(edge point b,edge point c) Equation (20)
The distance between edge points is calculated using the square root of the sum of squares of the distance in the Xm direction between edge points and the distance in the Ym direction between the edge points. The calculation method in the present modification substantially corresponds to that in the embodiment, except the calculation procedure.
In the on-screen measurement, since the distance Lx in the Xm direction between imaging centers and the distance Ly in the Ym direction between the imaging centers are both 0, the calculation of the first term can be omitted from Equations (19) and (20). That is, since it suffices that only the second term is calculated, the distance between edge points is obtained without using the machine coordinates (Xm, Ym).
The following configuration can be derived based on the above embodiment and modifications.
A) The machine tool 800 includes the attachment unit 822 to which either the imaging unit 600 or a tool is attachable.
B) The external computer 400 (an example of the information processing device) processes a two-dimensional image generated by being imaged by the imaging unit 600 in the machine tool 800 in which the attachment unit 822 moves to image the workpiece W placed in the machine tool 800, and the imaging unit 600 images the workpiece W at the imaging position.
C) The process of checking the position where the workpiece W is placed is performed (see the fourth modification). Therefore, the imaging unit 600 moves to a predetermined position based on the machine coordinates in the machine tool 800, and thereafter images to obtain the image.
D) The computation unit 486 calculates the position of the workpiece W on the basis of the image.
E) The machine tool 800 performs machining on the workpiece W.
F) The measurement process for the workpiece W is performed. In order to image a measurement target, the imaging unit 600 moves from a predetermined position to the imaging position, and thereafter images an image of the measurement target at the imaging position. A two-dimensional image can thereby be obtained.
G) The computation unit 486 detects an edge of the workpiece W from the two-dimensional image, and calculates the length of the workpiece W from a plurality of edges detected.
In the above configuration A, the imaging unit 600 can be used by being attached to the attachment unit 822 to which a tool can also be attached. Therefore, the workpiece W can be measured while remaining placed after machining. This enables additional machining to be immediately performed when needed. That is, conventionally, the workpiece W is removed from the table for measurement, and when additional machining is needed, it is necessary to correctly place the workpiece W back on the table. This requires much time for setup. In this regard, the above embodiment and modifications do not cause such a problem as described. It is unnecessary to move the workpiece W, and this can ensure sufficient accuracy in measuring the correct length. Higher measurement accuracy can be achieved compared to the case where the device only needs to grip the workpiece W.
In the above configurations F and G, as illustrated in
In the above configurations D and G, the computation unit 486 performs edge detection from an image imaged at a predetermined position to calculate the position of the workpiece W. Furthermore, the computation unit 486 calculates the position of the edge point that is a measurement target in the image from the distance between the position of the workpiece W and the center of the image. Specifically, the computation unit 486 calculates the distance in the Xm direction and the distance in the Ym direction between the position of the workpiece W and the center of the image. Next, the computation unit 486 calculates the spacing in the Xi direction and the spacing in the Yi direction between the center of the image and the edge point. The computation unit 486 multiplies the spacing in the Xi direction by the resolution, and multiplies the spacing in the Yi direction by the resolution, and can thereby calculate the distance in the Xm direction and the distance in the Ym direction between the center of the image and the edge point. The above two distances in the Xm direction are added to obtain the Xp coordinate of the edge point based on the position of the workpiece W. Similarly, the above two distances in the Ym direction are added to obtain the Yp coordinate of the edge point based on the position of the workpiece W. When the positions of two edge points are obtained, then the length between the edge points can also be obtained accordingly based on the obtained positions.
In the above configuration C, the “image” is an image that is used for checking the position. The “predetermined position” refers to the position of the imaging unit 600 that images an image for checking the position where the workpiece W is placed. The “predetermined position” is different from the “imaging position” in the above configuration F. The “image” imaged at the “predetermined position” includes a reference position (a specific edge point) of the workpiece W for calculating the position of the workpiece W. In the example in
The “imaging position” described in the above configuration F indicates a position on the horizontal plane (Xm coordinate, Ym coordinate: see
The present invention is not limited to the above described embodiment or the modifications thereof, and may be embodied while modifying the components without departing from the scope of the invention. Various inventions may be formed by appropriately combining plural components disclosed in the above described embodiment and the modifications thereof. Further, several components may be omitted from the entire components described in the above described embodiment and the modifications thereof.
In the embodiment and the modifications, an approach motion is not needed unlike measurement with a touch probe. In addition, since a plurality of areas can be imaged in a single imaging, the movement distance of the main shaft can be reduced. Therefore, workpiece measurement can be performed at a high speed. In the embodiment and the modifications, examples of the length of the workpiece calculated by the computation unit 486 include the diameter of a hole, the side length or diagonal-line length of a surface of the workpiece, and the arc length. Specifically, the radius and diameter of the first hole (
In the measurement using a touch probe, in addition to a repeatability error of the probe, a repeatability error in the machine tool 800 affects the measurement accuracy. However, in the on-screen measurement using the imaging unit 600, it is unnecessary to move the attachment unit 822 (the main shaft 100). Accordingly, a repeatability error in the machine tool 800 does not occur and the measurement accuracy is improved.
Measurement of a fine shape of a workpiece, or measurement of the orientation of the workpiece can be facilitated, although it is difficult when a touch probe is used. The fine shape is, for example, such a small hole or a narrow groove that a touch probe cannot be inserted. A workpiece with low rigidity or a thin portion of the workpiece cannot be measured correctly when a touch probe is used, for the reason that the low-rigidity workpiece or the thin portion may be deflected or deformed by the touch probe. In contrast, in the embodiment and the modifications, a pressure is not applied to a workpiece, which is advantageous in performing correct measurement without causing a deflection or deformation.
Further, placement error of a workpiece is automatically determined, and the correction value is set for the machine tool 800. This can reduce the machining error in the machine tool 800. Furthermore, an alert is issued when the placement error of the workpiece is determined to have occurred. This can prevent a machining defect or can inform a worker of the incorrect placement of the workpiece to prompt the worker to correct the inappropriate placement.
This application is a continuation application of International Application No. PCT/JP2022/034919, filed on Sep. 20, 2022, which claims priority to and the benefit of Japanese Patent Application No. 2021-158845 filed on Sep. 29, 2021 and Japanese Paten Application No. 2022-107560 filed on Jul. 4, 2022. The contents of these applications are incorporated herein by reference in their entirety.
| Number | Date | Country | Kind |
|---|---|---|---|
| 2021-158845 | Sep 2021 | JP | national |
| 2022-107560 | Jul 2022 | JP | national |
| Number | Date | Country | |
|---|---|---|---|
| Parent | PCT/JP2022/034919 | Sep 2022 | US |
| Child | 17976989 | US |
