Work conveying device and mapping method

Abstract

To enable accurate determination of a load state of a workpiece while using a reflection type sensor when performing mapping for determining the load state of the workpiece in an accommodation unit having a plurality of slots arranged in a first direction.SOLUTION: Wafers as workpieces are accommodated one by one in all slots of a cassette stage which is an accommodation unit (step 101), and the reflection type sensor is moved together with a hand to the position of the cassette stage under that state (step 102). An operation of moving the sensor along the first direction (vertical direction) to detect reflected light from a wafer is repetitively performed while moving the sensor along a second direction (horizontal direction) orthogonal to the first direction, thereby acquiring detection distribution data (step 105). Mapping positions to be used during actual operation are determined based on the detection distribution data (step 106).SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to the conveyance of a workpiece using a conveyance device such as a conveyance robot, and more particularly to a conveyance device and a mapping method that improve the accuracy of mapping for determining the in-stock state of a workpiece in a storage unit such as a cassette or a cassette stage.

Background Art

[0002] When a plate-shaped workpiece such as a semiconductor wafer or a glass substrate is conveyed by a conveyance device such as a conveyance robot, the workpiece to be conveyed is stored in a storage unit called a cassette or a cassette stage. The conveyance device extends a hand provided at its tip into the storage unit at the conveyance source to take out the workpiece, conveys the workpiece with the workpiece placed on the hand, and stores the workpiece in the storage unit at the conveyance destination. An example of a conveyance robot for conveying a workpiece such as a semiconductor wafer is shown in Patent Document 1. Generally, since the storage unit can store a plurality of workpieces, it is necessary to check the in-stock state of where the workpiece is located in the storage unit before actually performing the conveyance, and this is called mapping. Assuming that a plurality of slots are provided in the storage unit so as to be stacked in the vertical direction, by executing the mapping, it is possible to know which slots the workpiece is present in, and an optimal conveyance sequence can be assembled according to the result. In the conveyance sequence, it is specified from which slot of the storage unit at the conveyance source the workpiece is taken out and into which slot, which is the currently empty slot in the storage unit at the conveyance destination, the workpiece is stored.

[0003] Mapping is performed by non-contact detection of the workpiece in the housing using a sensor to determine its position. The sensors used for mapping include a transmissive sensor that detects the workpiece by blocking the optical path with the workpiece, and a reflective sensor that irradiates the workpiece with detection light and detects the light reflected by the edge (or end face) of the workpiece, and detects the workpiece based on the amount of reflection and the reflection position. The workpiece is, for example, a semiconductor wafer, and its thickness is, for example, 1 mm or less. Also, the mutual distance between the slots where the workpieces are placed one by one in the housing is about 5 mm to 10 mm. For these reasons, in the case of using either a transmissive or reflective sensor, the detection light is preferably laser light with a narrow beam diameter and sharp directivity. When using a transmissive sensor, there is an advantage that stable detection results can be obtained regardless of the material of the workpiece, but it is necessary to bring the sensor close to the housing or the workpiece, and there is a risk of interference or contact with the housing or the workpiece. Also, assuming that the sensor is attached to the hand, the hardware structure such as the hand needs to be in a shape that can detect the workpiece through transmission. Examples of performing mapping using a transmissive sensor are described in Patent Documents 2 and 3.

[0004] When using a reflective sensor, since the incoming state of the workpiece can be known from a position away from the workpiece, interference between the sensor and the workpiece can be prevented, and there are no restrictions on the hardware structure. For example, when the conveying device is a conveying robot, if a sensor can be arranged on the arm or hand of the robot, there is no need to make the surrounding of the sensor into a special shape for mapping. Therefore, mapping using a reflective sensor has great advantages in terms of introduction and operation. However, in mapping using a reflective sensor, the detection accuracy may deteriorate depending on the material, dimensions, and surface state of the workpiece. If the absorption of light by the workpiece can be ignored and the surface of the workpiece is rough, the light will be diffusely reflected, so the workpiece can be detected reliably. On the other hand, for example, when the workpiece is a silicon wafer and its surface is mirror-finished, depending on the mapping position of where to irradiate the detection light from the sensor, the detection light may be totally reflected, and in such a case, the workpiece cannot be detected appropriately. More specifically, the mapping position is the position on the workpiece where the light from the sensor is irradiated along the direction (referred to as the second direction) orthogonal to the first direction on the front surface of the accommodating portion when mapping is performed by moving the sensor in the direction in which the slots are arranged (referred to as the first direction). The mapping position can be changed by actually moving the sensor in the second direction, or it can also be changed by changing the irradiation direction of the light from the sensor without changing the position of the sensor in the second direction. When performing mapping using a reflective sensor, in order to perform stable detection, it is necessary to manually adjust the mapping position while referring to the detection results when the workpiece position is known.

[0005] When accommodating a workpiece in the cassette stage, one workpiece needs to be accommodated in each slot. However, for some reason, two or more workpieces may be stacked in one slot, or a workpiece may be placed so as to straddle two slots. When attempting to remove a workpiece from the cassette stage in such a defective state where the workpiece is accommodated in the cassette stage in an incorrect form, there is a risk of damaging the workpiece. Therefore, it is desirable that the mapping recognizes such defective states and issues an alert, thereby preventing damage accidents and the like. Such defective states can be discriminated by the thickness of the workpiece detected when mapping is performed and the position of the workpiece within the slot. If there is an error in discriminating the in-stock state by mapping, problems such as the robot going to acquire a workpiece at a slot where there is no workpiece and stopping due to an error, or overlapping and placing a workpiece on a slot where there is already a workpiece and damaging the workpiece may occur. Since the conveyance of the workpiece is continuously performed, it is required that the mapping can be accurately performed over a long period of time.

Prior Art Documents

Patent Documents

[0006]

Patent Document 1

Patent Document 2

Patent Document 3

Summary of the Invention

Problems to be Solved by the Invention

[0007] Mapping using a reflective sensor has the advantage that mapping can be performed without approaching the workpiece or the housing portion, but depending on the material and surface state of the workpiece, it may become impossible to accurately discriminate the in-stock state of the workpiece.

[0008] An object of the present invention is to provide a conveying device for conveying a workpiece, which can accurately determine the presence or absence of the workpiece while using a reflective sensor, and a mapping method capable of accurately performing such mapping.

Means for Solving the Problems

[0009] When conveying a workpiece for a housing portion having a plurality of slots arranged in a first direction and capable of accommodating one workpiece per slot, mapping is performed before actually conveying the workpiece to determine the presence or absence of the workpiece in the housing portion. In the mapping, the sensor is moved along the first direction to determine whether a workpiece is present in each slot. When using a reflective sensor to prevent the sensor from interfering with the workpiece or the housing portion during mapping execution, depending on the mapping position, the detection of the workpiece may become unstable due to the shape and material of the workpiece, so it is necessary to obtain an optimal mapping position in advance. The present invention aims to automate the determination of the mapping position as a preprocessing before conveying the workpiece. Once the mapping position is determined, the same mapping position can be used as long as the shape and material of the workpiece do not change.

[0010] Therefore, the conveying device of the present invention is a conveying device that conveys workpieces for a housing part having a plurality of slots arranged in a first direction and capable of accommodating one workpiece per slot, and includes a hand that holds the workpiece when conveying the workpiece, a reflection type sensor that irradiates the workpiece with light and detects the reflected light, a moving mechanism that moves the hand and the sensor with respect to the housing part, and a control means that drives and controls the moving mechanism. The control means moves the sensor toward the position of the housing part in a state where one workpiece is accommodated in each of all the slots, and then repeatedly executes an operation of moving the sensor along the first direction to detect the reflected light from the workpiece, by moving the light irradiation position from the sensor along a second direction orthogonal to the first direction on the front surface of the housing part, and obtains detection distribution data. Based on the detection distribution data, a mapping position is determined. The mapping position is the position along the second direction of the irradiation position on the workpiece when the sensor is moved along the first direction for performing mapping to discriminate the presence state of the workpiece in the housing part.

[0011] With respect to the housing part in a state where one workpiece is accommodated in each of all the slots, repeatedly executing the operation of moving the sensor along the first direction to detect the reflected light from the workpiece, by moving the sensor along a second direction orthogonal to the first direction on the front surface of the housing part, can be said to be repeatedly performing mapping at different mapping positions with respect to the housing part. In this case, data indicating the detection intensity of the workpiece can be obtained from the sensor for each mapping position and each slot. These data can be represented as detection distribution data with the mapping position and the slot position as coordinates respectively. Since it is known that one workpiece is accommodated in each of all the slots, based on this, the mapping position where the workpiece is detected with an appropriate detection intensity for each slot can be extracted from the detection distribution data. In the conveying device of the present invention, by determining the mapping position from the detection distribution data in this way, it becomes possible to accurately discriminate the presence state of the workpiece in the mapping when actually conveying the workpiece during actual operation.

[0012] In the conveying device of the present invention, it is preferable that the control means determines the moving range of the sensor along the first direction when acquiring the detection distribution data and when performing mapping by moving the sensor along the first direction in front of the housing part before acquiring the detection distribution data. By determining in advance the moving range of the sensor along the first direction in this way, the amount of movement of the sensor along the first direction can be minimized, improving the working efficiency. Also, by determining the moving range along the first direction, it becomes possible to easily determine in which slot the workpiece is present from the position of the sensor within that moving range.

[0013] In the conveying device of the present invention, it is preferable that the sensor is attached to the hand. Although it is necessary to move the sensor for mapping, by attaching the sensor to the hand that moves relative to the housing part, the configuration of the moving mechanism in the conveying device can be simplified.

[0014] In the conveying device of the present invention, for example, the workpiece is a plate-like object horizontally accommodated in the slot, and the first direction is the vertical direction. Detection by a reflection-type sensor tends to become unstable when the workpiece is a plate-like object, but according to the conveying method of the present invention, it becomes possible to perform mapping that can surely determine the in-stock state of the workpiece even in such a case.

[0015] In the conveying device of the present invention, when the control means controls the moving mechanism to convey the workpiece, it can execute control to perform mapping on the storage unit using the determined mapping position. In such a conveying device, when actually conveying the workpiece after determining the mapping position as described above, mapping is performed at the determined mapping position, so that the in-stock state of the workpiece in the storage unit can be accurately determined. In that case, the control means determines the first mapping position and the second mapping position from the detection distribution data, and when performing mapping, moves the sensor at the first mapping position along the forward direction of the first direction, and then may move the sensor at the second mapping position along the reverse direction of the first direction. Since mapping is performed at different mapping positions while reciprocally moving the sensor along the first direction, the accuracy of mapping can be improved while minimizing an increase in the moving amount of the sensor.

[0016] The mapping method of the present invention is a mapping method for determining the in-stock state of workpieces in a storage unit having a plurality of slots arranged in a first direction and capable of accommodating one workpiece per slot, and includes a moving step of moving a reflection type sensor toward the position of the storage unit in a state where one workpiece is accommodated in each of all the slots, and after the moving step, repeating an operation of moving the sensor along the first direction to detect reflected light from the workpiece by moving the irradiation position of the light from the sensor along a second direction orthogonal to the first direction on the front surface of the storage unit to obtain detection distribution data, and a determination step of determining a mapping position based on the detection distribution data. The mapping position is the position along the second direction of the irradiation position when moving the sensor along the first direction for mapping.

[0017] In the mapping method of the present invention, in order to obtain the optimal mapping position, mapping is repeated while changing the mapping position for the accommodation part in a state where one workpiece is accommodated in each slot, and detection distribution data is obtained based on the data from the sensor at that time. Since it is known that one workpiece is accommodated in each slot, the mapping position where the workpiece is detected with an appropriate detection intensity can be extracted for each slot from the detection distribution data. In the mapping method of the present invention, by determining the mapping position from the detection distribution data in this way, it becomes possible to accurately determine the presence state of the workpiece in the mapping when actually transporting the workpiece during actual operation.

[0018] In the mapping method of the present invention, before the determination step after the movement step, it is preferable to move the sensor along the first direction on the front surface of the accommodation part to determine the movement range of the sensor along the first direction when performing the acquisition step and the mapping. By determining the movement range of the sensor along the first direction in advance in this way, the movement amount of the sensor along the first direction can be minimized and the work efficiency can be improved. Also, by determining the movement range along the first direction, it becomes possible to easily determine in which slot the workpiece exists from the position of the sensor in that movement range.

[0019] In the mapping method of the present invention, for example, the workpiece is a plate-like object and is horizontally accommodated in the slot, and the first direction is the vertical direction. Detection by a reflection type sensor tends to be unstable when the workpiece is a plate-like object, but according to the mapping method of the present invention, it becomes possible to perform mapping that can surely determine the presence state of the workpiece even in such a case.

[0020] In the mapping method of the present invention, when the workpiece is conveyed after the determination step, mapping with respect to the storage unit can be performed using the mapping position. In this method, when the workpiece is actually conveyed after determining the mapping position, mapping is performed at the determined mapping position, so that the in-stock state of the workpiece in the storage unit can be accurately determined. In that case, in the determination step, the first mapping position and the second mapping position are determined, and when mapping is performed, the sensor is moved at the first mapping position along the forward direction of the first direction, and then the sensor is moved at the second mapping position along the reverse direction of the first direction. Since mapping is performed at different mapping positions while reciprocally moving the sensor along the first direction, it is possible to improve the mapping accuracy while minimizing an increase in the movement amount of the sensor.

Advantages of the Invention

[0021] According to the present invention, it is possible to obtain a conveying device for conveying a workpiece, which can accurately determine the in-stock state of the workpiece while using a reflection type sensor, and a mapping method capable of performing such accurate mapping.

Brief Description of the Drawings

[0022]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Mode for Carrying Out the Invention

[0023] Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 shows a transfer robot which is a transfer device according to an embodiment of the present invention. The transfer robot 20 shown in FIG. 1 is used for transferring a workpiece which is a plate-like object. Hereinafter, it is assumed that the workpiece is a wafer 50 such as a silicon wafer, and the wafer 50 is accommodated in a cassette stage 40 which is an accommodation unit. The cassette stage 40 includes a plurality of slots arranged in a first direction, specifically, the vertical direction here. In this cassette stage 40, it is assumed that one wafer 50 is horizontally accommodated for each slot. Before transferring the wafer 50, the transfer robot 20 executes a mapping for determining the in-stock state of the wafer 50 in the cassette stage 40.

[0024] The transfer robot 20 shown in Fig. 1 is the same as, for example, the horizontal articulated robot described in Patent Document 1, and includes three arms 22 to 24 serially attached to the base 21, and a hand 25 attached to the tip of the tip-side arm 24. Fig. 1 depicts the state where the arms 22 to 24 are folded, but the arms 22 to 24 can also be in an extended state. In particular, the arms 22 and 23 form a link mechanism in which the tip of the arm 23 moves radially in the horizontal plane around the attachment position of the arm 22 to the base 21. The arm 24 is rotatable in the horizontal plane with respect to the arm 23, and the hand 25 is rotatable with respect to the arm 24. The base 21 is provided with a lifting mechanism (not shown) for raising and lowering the base-side arm 22 and a rotating mechanism (not shown) for rotating the arm 22 around the vertical axis. A reflective sensor 29 for performing mapping is attached to the hand 25. A robot controller 30 for driving and controlling the transfer robot 20 is further connected to the transfer robot 20. The robot controller 30 controls the lifting mechanism and the rotating mechanism provided on the base 21, the link mechanism composed of the arms 22 and 23, the mechanism for rotating the arm 24 with respect to the arm 23, and the mechanism for rotating the hand 25 with respect to the arm 24, and processes the detection signals from the sensor 29 described later. In particular, the robot controller 30 executes control for operations for mapping positioning, mapping operations, and operations for transferring workpieces.

[0025] Figures 2(a) and 2(b) show the relationship among the hand 25, the sensor 29, and the wafer 50 when mapping is being performed. The hand 25 is an elongated member provided with a fork portion 26 at one end thereof, and a reflective sensor 29 is attached to the other end of the hand 25. The fork portion 20 has a shape in which the tip side branches into two, and is a portion that holds a workpiece such as the wafer 50 when transporting the workpiece. When transporting, the workpiece, which is a plate-like object, is horizontally placed on the surface of the fork portion 26. The sensor 29 is an integrated unit of a light-emitting portion that generates and emits laser light 60 by a laser diode (LD) or the like, and a light-receiving portion that detects the reflected light when the laser light 60 is reflected by a workpiece such as the wafer 50.

[0026] When the workpiece is the wafer 50, the mapping of the wafer 50 is performed by irradiating the laser light 60 from the sensor 29 toward the end face (wafer edge) of the wafer 50 while moving the sensor 29 in the thickness direction of the wafer 50, and detecting the reflected light from the end face of the wafer 50. The thickness of the wafer 50 can also be detected from the amount of movement of the sensor 29 within the period when the reflected light is being detected. However, the direction and intensity of the reflected light vary greatly depending on the surface state of the end face of the wafer 50. When the end face of the wafer 50 is a rough surface, the irradiated laser light 60 is diffusely reflected at the end face of the wafer 50 as shown in Fig. 2(a), and a part of the reflected light due to the diffuse reflection reaches the light-receiving portion of the sensor 29. As a result, the sensor 29 can detect the wafer 50 regardless of the mapping position. On the other hand, when the end face of the wafer 50 is, for example, a mirror surface, the irradiated laser light 60 is specularly reflected in one direction at the end face of the wafer 50 as shown in Fig. 2(b). In this case, even if the laser light 60 is irradiated onto the wafer 50 with an inappropriate mapping position, the reflected light may not return to the sensor 29, and the sensor 29 may not be able to detect the wafer 50. The transfer robot 20 of the present embodiment attempts to automatically determine an appropriate mapping position based on the material and shape of the wafer 50 by the transfer device itself before performing mapping for the transfer of the wafer 50.

[0027] The transfer robot 20 performs processes such as taking out one wafer 50 from the cassette stage 40 at the transfer source and transferring the wafer 50 toward the device or cassette stage 40 at the transfer destination, and accommodating the wafer 50 transferred from the device or cassette stage 40 at the transfer source into the empty slot of the cassette stage 40 at the transfer destination. The cassette stage 40, which is the storage unit, is a front-opening type such as a front-opening unified pod (FOUP), which is a front-opening integrated transfer and storage box defined by SEMI (Semiconductor Equipment and Materials International) standard E47.1. Therefore, when the transfer robot 20 takes out the wafer 50 from the slot in the cassette stage 40 and when accommodating the wafer 50 in the empty slot in the cassette stage 40, the hand 25 of the transfer robot 20 enters the cassette stage 40 aiming at a predetermined slot through the opening in the front of the cassette stage 40 with the fork portion 26 side as the tip.

[0028] The determination of the mapping position in this embodiment will be described. FIG. 3 shows the procedure for determining the mapping position. Steps 102 to 106 in the procedure shown in FIG. 3 are executed by the robot controller 30 controlling the transfer robot 20. Since the mapping position needs to be determined for each shape and material of the wafer 50, first, in step 101, wafers 50 having the same shape and the same material as the wafer to be transferred are accommodated one by one in all the slots of the cassette stage 40 that is the mapping target. Next, in step 102, teaching of the cassette stage position for normal wafer transfer is performed. In this case, teaching is performed to move the hand 25 to the standby position, which is the position in the front of the target cassette stage 40 and where the hand 25 waiting to access the inside of this cassette stage 40 waits.

[0029] Next, in step 103, the hand 25 is rotated with respect to the arm 24 so that the reflective sensor 29 attached to the hand 25 faces the stage cassette 40, and the laser light 60 from the sensor 29 irradiates almost the center of the wafer 50 in a state where the cassette stage 40 is viewed from the front in the opening side. With the lower end position of the lowermost slot of the cassette stage 40 as the mapping lowest point LL, the teaching of the hand 25 is performed. This teaching is for mapping, and the teaching position is registered in the robot controller 30 as for mapping. Subsequently, the hand 25 is moved upward in the vertical direction, and similarly, teaching is performed with the upper end position of the uppermost slot of the cassette stage 40 as the mapping uppermost point UL. FIG. 4 is a diagram for explaining the teaching here. The section between the mapping lowest point LL and the mapping uppermost point UL and the sections of slightly set margins at both ends of this section are determined as the moving range of the sensor 29 in the vertical direction when determining the mapping position in step 104 and when actually performing mapping after determining the mapping position, that is, the mapping range which is the moving range of the hand 25.

[0030] From the teaching height TL when teaching the hand 25 to the standby position in step 102 and the mapping lowest point LL, an offset between the actual slot height represented in the coordinate system of the transfer robot 20, in other words, as seen from the transfer robot 20, and the height detected by the irradiation of the laser light 60 can be obtained. If this offset is called the mapping offset, the mapping offset is Mapping offset = LL - TL represented by. The mapping offset can be used to determine which slot from the bottom the wafer 50 is present in the cassette stage 40 when detecting the wafer 50. Here, the vertical direction is the Z direction, the direction from the hand 25 in the standby position toward the cassette stage 40 is the Y direction, and the direction orthogonal to the Z direction on the front surface of the cassette stage 40 is the X direction. The X direction and the Y direction are in the horizontal plane and orthogonal to each other.

[0031] By executing Steps 101 to 104, the preliminary procedures for obtaining the optimal mapping position are completed. Thereafter, in Step 105, detection distribution data used for determining the mapping position is acquired. FIGS. 5(a) and 5(b) are diagrams for explaining the acquisition of the detection distribution data. In the acquisition of the detection distribution data, as shown in FIG. 5(a), with respect to the cassette stage 40 in which the wafers 50 are accommodated in all slots, while emitting the laser beam 60 in the same manner as the mapping performed during the conveyance of the wafer 40, the sensor 29 is moved along the vertical direction, that is, the Z direction within the mapping range, and an attempt is made to detect each wafer 50 within the cassette stage 40. This is called the detection operation. Then, this detection operation is repeatedly executed while moving the sensor 29 in the X direction. FIG. 5(b) shows at which positions in the X direction the detection operation is to be performed by the laser beam 60 from the sensor 29. Specifically, centering on the position in the X direction when the mapping lowest point LL is obtained, a range of, for example, ±10 mm in the positive and negative directions of the X direction is set, and the detection operation is performed while moving the sensor 29 along the X direction at a pitch of 0.5 mm to 1 mm within this range. Here, from among the positions where the detection operation is performed, as will be described later, the optimal mapping position is searched for. In the example shown in FIG. 5(b), the detection operation is performed at 11 positions A to K in the X direction, and among them, the position F is the position when the mapping lowest point LL is obtained.

[0032] In the above description, it is assumed that the sensor 29 is actually moved in the X direction and the detection operation is repeated. However, since the mapping position is the irradiation position of the laser beam 60 along the X direction on the wafer 50, the detection operation may be repeated while changing the emission direction of the laser beam 60 within a range of, for example, ±5° at a pitch of 0.2° to 0.4° in the horizontal plane centered on the Y direction while fixing the position of the sensor 29 (or the hand 25) in the X direction. The emission direction of the laser beam 60 can be easily changed, for example, by changing the angle of the hand 25 with respect to the arm 24. Furthermore, in order to improve the accuracy of determining the optimal mapping position, the detection operation may be performed while shifting the position of the sensor 29 within a range of, for example, ±5 mm in the Y direction.

[0033] When the detection operation is repeated while changing the position of the sensor 29 in the X direction, for each slot of the cassette stage 40 and for each position in the X direction where the detection operation is performed, a detection result is obtained from the sensor 29 as the thickness of the wafer 50 accommodated in the corresponding slot. A series of detection results thus obtained can be expressed as two-dimensional detection distribution data with the position in the X direction as one dimension and the position of the slot in the Z direction as the other dimension. FIG. 6 shows an example of the detection distribution data obtained when the number of slots in the cassette stage 40 is 10, the slot positions are sequentially numbered from 1 to 10 from the bottom, the positions of the detection operations in the X direction are 11 locations A to K described above, and silicon wafers with a thickness of 0.9 mm are stored as the wafers 50 in each slot. In the figure, the region represented in black is the region where the sensor 29 detects the wafer 50. The larger the dimension of the region in the Z direction (the vertical direction in the figure), the thicker the wafer 50 is detected. The dotted line indicates that the wafer 50 could not be detected despite the presence of the wafer 50. Table 1 shows the actual thickness of the wafer 50 detected for each position in the X direction and for each position of the slot. The unit of the numerical values in Table 1 is mm.

[0034]

Table 1

[0035] Once the detection distribution data is obtained, in step 106, based on the detection distribution data, the mapping position in the X direction to be used in the mapping during actual conveyance is determined. Hereinafter, the determination of the mapping position from the detection distribution data will be described.

[0036] In the example shown in FIG. 6 and Table 1, since the wafer 50 with a thickness of 0.9 mm is stored in each slot of the cassette stage 40, the X-direction position where the detected thickness is 0.9 mm in all slots is most suitable as the mapping position. Considering the results in Table 1 from this perspective, at positions A, B, and K, although there are actually 10 wafers 50 in the cassette stage 40, there are wafers 50 that cannot be detected. This is presumably because the irradiation position of the laser beam 60 is far from the center position of the wafer 50 along the X direction, so correct reflection of the laser beam 60 on the end face of the wafer 60 cannot be obtained, and thus the wafer 50 cannot be detected correctly. Since the mapping results are not reliable, mapping should not be performed at these positions A, B, and K.

[0037] At positions F, G, and H, although the actual thickness of the wafer 50 is about 1 mm, there are slots that detect the wafer 50 with a large thickness ranging from 2 mm to 4 mm. Also, although wafers 50 of the same thickness are the detection targets, variations more than twice the detected thickness can also be seen. This is presumably because over-reflection occurs due to the relationship of the reflection angle of the laser beam 60 on the wafer 50. In this state, since it is not reliable when trying to detect the presence or absence of wafer overlap in the slot from the detected wafer thickness, mapping at these positions F, G, and H should not be performed either.

[0038] Since positions A, B, F, G, H, and K are unsuitable as mapping positions, an X-direction position suitable for mapping is selected from the remaining positions C, D, E, I, and J as the mapping position. Here, position D or position I is determined as the mapping position. After determining the mapping position in this way, the teaching position previously registered as the position for mapping in step 103 is automatically corrected to the mapping position determined this time. The conditions for determining the mapping position include the following (1) to (3).

[0039] (1) The average value of the detected thickness is closest to a predetermined thickness (e.g., the thickness of the wafer 50); (2) The variation among all the values of the detected thickness is the smallest; (3) The difference between the maximum value and the minimum value of the detected thickness is the smallest.

[0040] When determining the mapping position, any one of the conditions (1) to (3) may be used, or two of these conditions may be used, or all of the conditions (1) to (3) may be used. Also, in the above, the detection operation is performed while shifting the position in the X direction. However, the detection operation is similarly performed while shifting the axial angle of the hand 25 with respect to the arm 24 or the position in the Y direction of the sensor 29, and the axial angle and the Y-direction position that can obtain stable detection results are found, and at least one of the axial angle of the hand 25 and the position in the Y direction and the position in the X direction may be combined as the mapping position.

[0041] By executing the processes of steps 101 to 106, an optimal mapping position for performing mapping to determine the in-loading state of the wafer 50 on the cassette stage 40 is determined. Once the mapping position is determined in this way, thereafter, when actually operating the transfer robot 20 to transfer the wafer 50, the mapping is executed at this determined mapping position. Since the teaching position registered as the mapping position in step 103 is corrected to the mapping position determined as described above in step 106 so that the laser beam 60 irradiates the center position of the wafer 50 in the X direction, the mapping during actual transfer can be executed by moving the hand 25 to the teaching position registered for mapping.

[0042] In order to complete the mapping in the shortest possible time, at a predetermined mapping position, the sensor 29 is moved only once upward or downward within the mapping range to perform the detection operation. However, in order to improve the mapping accuracy, the sensor 29 is moved by a reciprocating motion combining upward movement and downward movement to perform the detection operation, and the detection values of the forward path and the return path can be averaged. When performing mapping by the reciprocating motion, it is preferable to obtain two different mapping positions in step 106 and then perform the detection operation at different mapping positions for the forward path and the return path. When position D and position J are determined as the mapping positions in the example shown in FIG. 6 and Table 1, as shown in FIG. 7, the detection operation is performed at mapping position D when the sensor 29 ascends, and the detection operation is performed at mapping position J when it descends. Then, the detection results are averaged for each slot to obtain the final mapping result. In FIG. 7, the thick line 70 indicates the locus of the irradiation position of the laser beam 60.

[0043] In the transport robot 20 of the present embodiment described above, detection distribution data is acquired by performing a detection operation while changing the position in the X direction with respect to the cassette stage 40 in which the wafers 50 are accommodated in all the slots, and the optimal mapping position is automatically determined from this detection distribution data. Then, during the actual transport of the wafer 50 thereafter, the determined mapping position is used. As a result, it becomes possible to accurately execute the mapping for discriminating the loading state of the wafer 50 on the cassette stage 40 while using a reflective sensor without manually adjusting the mapping position.

Description of Reference Numerals

[0044] 20... Transport robot; 21... Base; 22 - 24... Arms; 25... Hand; 26... Fork portion; 29... Sensor; 30... Robot controller; 40... Cassette stage; 50... Wafer; 60... Laser beam; 70... Mapping path.

Claims

1. A conveying device for conveying a workpiece, which has a plurality of slots arranged in a first direction and a housing portion capable of accommodating one workpiece for each slot, comprising: a hand for holding the workpiece when conveying the workpiece; a reflection type sensor for irradiating the workpiece with light and detecting reflected light; a moving mechanism for moving the hand and the sensor relative to the housing portion; control means for driving and controlling the moving mechanism, wherein the control means moves the sensor toward the position of the housing portion in a state where one workpiece is accommodated in each of all the slots, and then moves the sensor along the first direction to detect the reflected light from the workpiece, and repeats the operation of moving the irradiation position of the light from the sensor on the workpiece along a second direction orthogonal to the first direction on the front surface of the housing portion to obtain detection distribution data, determines a mapping position based on the detection distribution data, wherein the mapping position is the position along the second direction of the irradiation position when the sensor is moved along the first direction to perform mapping for discriminating the presence state of the workpiece in the housing portion.

2. The conveying device according to claim 1, wherein the control means moves the sensor along the first direction on the front surface of the housing portion before obtaining the detection distribution data to determine the moving range of the sensor along the first direction when obtaining the detection distribution data and when performing the mapping.

3. The conveying device according to claim 1 or 2, wherein the sensor is attached to the hand.

4. The conveying device according to any one of claims 1 to 3, wherein the workpiece is a plate-shaped object horizontally accommodated in the slot, and the first direction is a vertical direction.

5. The conveying device according to any one of claims 1 to 4, wherein the control means executes control for performing mapping on the housing portion using the determined mapping position when controlling the moving mechanism to convey the workpiece.

6. The control means determines a first mapping position and a second mapping position from the detected distribution data, and when performing the mapping, moves the sensor at the first mapping position along the forward direction of the first direction, and then moves the sensor at the second mapping position along the reverse direction of the first direction. The conveying device according to claim 5.

7. A mapping method for determining the in-stock state of the work in a storage unit that includes a plurality of slots arranged in a first direction and can accommodate one work for each slot, a moving step of moving a reflective sensor toward the position of the storage unit in a state where the work is accommodated in each of all the slots; an acquisition step of, after the moving step, repeatedly performing an operation of moving the sensor along the first direction to detect the reflected light from the work by moving the irradiation position of the light from the sensor on the work along a second direction orthogonal to the first direction on the front surface of the storage unit to acquire detection distribution data; a determination step of determining a mapping position based on the detected distribution data; characterized by The mapping position is the position along the second direction of the irradiation position when the sensor is moved along the first direction for performing the mapping. A mapping method.

8. Before the determination step after the moving step, the sensor is moved along the first direction on the front surface of the storage unit to determine the moving range of the sensor along the first direction when performing the acquisition step and the mapping. The mapping method according to claim 7.

9. The work is a plate-shaped object and is horizontally accommodated in the slot, and the first direction is the vertical direction. The mapping method according to claim 7 or 8.

10. When performing the conveyance of the work after the determination step, mapping with respect to the storage unit is performed using the mapping position. The mapping method according to any one of claims 7 to 9.

11. In the determination step, a first mapping position and a second mapping position are determined, and when performing the mapping, the sensor is moved at the first mapping position along the forward direction of the first direction, and then the sensor is moved at the second mapping position along the reverse direction of the first direction. The mapping method according to claim 10.

Images