JP6026746B2 - Workpiece assembly procedure calculation method, assembly procedure calculation program, part manufacturing method, and automatic assembly robot - Google Patents

Description

  The present invention relates to a multi-point fitting assembly in which a workpiece is assembled by inserting a plurality of fitting feet into corresponding fitting holes.

  In recent years, automatic assembly robots have attracted attention as production equipment that can handle production of a wide variety and a small quantity due to diversification of market needs. Generally, this automatic assembly robot is equipped with a robot arm having a robot hand with a force sensor at the tip, and it can handle various assembly by changing the control program related to the operation of the robot arm according to the product to be assembled It is a feature that can be done.

  The automation of product assembly at production sites has been studied and realized using such automatic assembly robots, but replacement by automatic assembly robots is not easy, and there are still assemblies that rely on humans for much of them. . One of them is multi-point fitting assembly. The multipoint fitting assembly is an assembly of a work for fitting a plurality of fitting feet into corresponding fitting holes. It is characterized by creating a high restraint force between the mating object and the mating object by inserting and mating multiple mating feet into the corresponding mating hole. Often used for assembly.

  One-point fitting assembly that fits one fitting foot into one fitting hole is an assembly that is generally performed by an automatic assembly robot, while many of the multi-point fitting assembly is still manual. The reason why it is relied on is due to the technical difficulty of the multi-point fitting assembly as follows.

  In the one-point fitting assembly, the position of the robot arm is aligned with one fitting leg provided on one of the fitting and the fitting and one fitting hole provided on the other. When the fitting foot is inserted into the fitting hole, it is assembled using compliance control using a force sensor. On the other hand, in multi-point fitting assembly, in addition to the dimensional tolerances of each fitting foot and each fitting hole, there is also a distance tolerance between the fitting feet and the fitting holes. The positioning accuracy required for the robot arm at the time of alignment is dramatically increased. Depending on the work to be assembled, due to dimensional tolerances, the fitting hole and the fitting foot may physically interfere with each other, and it may be impossible to simultaneously insert all the fitting feet.

  A person can position such a multi-point fitting assembly with high precision while avoiding the situation where simultaneous insertion is impossible due to dimensional tolerances by supporting the insertion feet to be inserted with fingers and elastically deforming them. First, assemble the fitting object and the fitting object. However, this assembling method that imitates humans has been difficult to realize with conventional automatic assembly robots due to factors such as the number of fingers, the number of finger joints, size, and control complexity.

  Therefore, conventionally, parts with relatively high strength, such as automobile body parts, are selected in order from soft fitting feet with relatively low rigidity, and inserted into the fitting holes sequentially using the elastic deformation of the fitting feet. Thus, a method of performing multi-point fitting assembly has been devised (see Patent Document 1).

  Here, the reason for inserting in order from the low-rigid fitting foot is that the fitting between the already-inserted fitting foot and the fitting hole and the fitting object are restrained by the low-rigid fitting. This is because the elastic deformation of the foot can be used to hold it in a loose state. Thereby, when inserting the next fitting leg, the robot arm can operate without difficulty, and the insertion operation can be advanced more easily.

Japanese Patent No. 2768210

  By the way, there are many products that are subject to multi-point fitting assembly such as exterior assembly of small precision devices such as digital cameras as well as large products such as cars. Therefore, even for such a small product having a relatively low strength, the development of automation of multi-point fitting assembly is expected.

  However, in the case of multi-point fitting assembly of the above-mentioned small products, the fitting feet are very thin and thin, and the number of precise shapes increases, so the fitting feet are likely to be plastically deformed and broken during assembly. Become. The cause of the plastic deformation and breakage of the fitting foot is that an excessive load is generated on the inserted fitting foot when multiple fitting feet are sequentially inserted.

  In the method described in Patent Document 1, assembly is performed by determining the insertion order by paying attention to the relative rigidity of each fitting foot. The reason for this is that if the insertion is performed from a fitting foot having low rigidity, the relative position between the fitting and the fitting object can be easily adjusted by deformation of the inserted fitting foot.

  However, the rigidity value and the resistance to breakage do not always match. Even if the rigidity is high, there is a fitting foot with a shape and material that allows a large amount of deformation of the fitting foot until it breaks. There is also a leg. For this reason, in the above-described method for determining the insertion order based on the rigidity of the fitting feet, the insertion order may not be such that the fitting feet are prevented from being destroyed by a load applied to the fitting feet of the workpiece.

  Further, in Patent Document 1 above, the determination of the insertion order is determined only by the rigidity (scalar amount) of the fitting foot alone, and the direction of softness is not taken into consideration. In some cases, an excessive load is applied to the inserted fitting foot. This problem will be described by taking the multi-point fitting assembly of the workpiece in FIG. 17 as an example.

  The workpiece W shown in FIG. 17 is divided into a fitting object W1 and a fitting object W2, and the fitting article W1 includes three fitting legs 201, 202, 203, and the fitting object W2 has a structure including fitting holes 301, 302, and 303. Then, by inserting the fitting foot 201 into the fitting hole 301, the fitting foot 202 into the fitting hole 302, and the fitting foot 203 into the fitting hole 303, the fitting object and the fitting object are fitted. It is configured to be.

  Table 1 shows the rigidity value of each fitting foot. According to the values in Table 1, when the insertion order is determined by focusing on the lowest rigidity value of each fitting foot, the fitting foot 202 (X-axis direction rigidity value 20 g / mm) is inserted first, and then the fitting is performed. A leg 201 (Y-axis direction rigidity value 40 g / mm) is inserted. Then, the mating feet 203 (XY-axis direction rigidity value 200 g / mm) are inserted, and when all the mating feet are inserted into the mating holes, the mating feet are inserted to the full depth of each mating foot. Insertion completes the mating.

  When the fitting foot 202 and the fitting foot 201 are inserted into the corresponding fitting holes 302 and 301, the two fitting feet of the fitting foot 202 and the fitting foot 201 are placed between the fitting object and the fitting object. Restraints in the X-axis direction and Y-axis direction based on the rigidity work. As the combined rigidity value of the fitting foot 202 and the fitting foot 201, if the sum of the rigidity of both fitting feet is simply taken as an evaluation value (the rigidity value in the X-axis direction is 540 g / mm, the rigidity value in the Y-axis direction) 620 g / mm). In the insertion order focusing on the lowest rigidity value of each fitting foot described above, the last fitting foot 203 must be inserted by moving the fitting W1 in such a restrained state.

  Here, for example, consider the case where the insertion order is the order of the fitting foot 202, the fitting foot 203, and the fitting foot 201. Then, when the last fitting foot 201 is inserted into the fitting hole 301, the rigidity value (the rigidity value in the X-axis direction is 600 g / mm, the rigidity value in the Y-axis direction is 120 g). / Mm) restrains the fitting object and the fitting object. At this time, the rigidity value in the X-axis direction is not significantly different from the insertion order focusing on the lowest rigidity value of each fitting foot, but the rigidity value in the Y-axis direction is much smaller in this insertion order. .

  As a result, the conventional method of determining the insertion order based on the rigidity (scalar amount) of the mating foot alone does not necessarily result in an optimal insertion order for reducing the load on the inserted mating foot. I understand.

Accordingly, an object of the present invention is to provide a workpiece assembly procedure calculation method, an assembly procedure calculation program , a part manufacturing method, and an automatic assembly robot in consideration of loads applied to fitting feet in multi-point fitting assembly.

In the workpiece assembly procedure calculation method according to the present invention, the calculation device includes the first and second workpieces among the first and second workpieces that are assembled to each other by fitting the fitting feet into the corresponding fitting holes . An obtaining step for obtaining data obtained for at least a horizontal direction of an allowable deformation amount, which is a deformation amount that can be deformed within an elastic region, for each of the plurality of fitting feet provided on at least one of the workpieces; but on the basis of data of the acquired tolerance deformation amount, for each of the insertion order of the plurality of the Hamagoashi, allowed relative movement of said first and second workpiece when inserting the Hamagoashi sequentially a computation step asking you to amount as an assembly flexibility, with a, characterized in that.

The automatic assembly robot according to the present invention grips one of the first and second workpieces that are assembled to each other by fitting a plurality of fitting feet into the corresponding fitting holes, and the other one placed. Based on the robot arm assembled to the workpiece and the deformation amount data that can be deformed within the elastic range in the horizontal direction of the plurality of mating feet, the mating feet are inserted in order for each of the plurality of mating foot insertion orders. and determined Mel computing device allowed relative movement of the first and second workpiece when inserted as an assembly flexibility, with a, characterized in that.

  By calculating the range in which the first workpiece can move relative to the second workpiece (assembly degree of freedom) in terms of size and direction (orientation), and evaluating the insertion order of the fitting feet based on this degree of assembly freedom, It is possible to select an insertion order that can maintain a large degree of assembly freedom until the second half. As a result, in the multi-point fitting assembly of the workpiece, in the process of inserting the fitting foot into the fitting hole, an excessive load that leads to plastic deformation or breakage is applied to the inserted fitting foot. Can be prevented.

The schematic diagram which shows the workpiece assembly system which concerns on embodiment of this invention. The schematic diagram which shows the robot controller of the automatic assembly robot which concerns on embodiment of this invention. The flowchart which shows the fitting procedure of the workpiece | work which concerns on embodiment of this invention. The flowchart which shows the determination method of the insertion order of the fitting leg which concerns on embodiment of this invention. It is a schematic diagram which shows the workpiece | work which concerns on embodiment of this invention, (a) is a schematic diagram which shows a 1st workpiece | work (fitting thing), (b) is a schematic diagram which shows a 2nd workpiece | work (fitting thing). (A) is a side view of the workpiece | work shown in FIG. 5, (b) Sectional drawing of the workpiece | work shown in FIG. The figure which shows the sum total of the assembly freedom in the insertion order of each fitting leg in embodiment of this invention, and an assembly freedom. It is a figure which shows the difference in the assembly freedom by the insertion order from Example 1, (a) is a graph which shows the difference in the assembly freedom of an X-axis direction, (b) is the difference in the assembly freedom in a Y-axis direction. Graph showing. It is a figure which shows the difference in the assembly freedom by the insertion order between Example 2, (a) is a graph which shows the difference in the assembly freedom of the X-axis direction, (b) is the difference in the assembly freedom in the Y-axis direction. Graph showing. The flowchart which shows the determination method of the insertion order of the fitting leg different from the method of FIG. It is a schematic diagram explaining the determination method of an assembly flow, (a) is a schematic diagram explaining insertion of the 1st side of a 1st workpiece | work, (b) is a schematic diagram explaining insertion of the 2nd side. It is a schematic diagram explaining the determination method of an assembly flow, (a) is a schematic diagram explaining insertion of the 1st side of a 3rd workpiece | work, (b) is a schematic diagram explaining insertion of the 4th side. The schematic diagram explaining the track | line which does not apply a load to the already inserted fitting leg. The schematic diagram for demonstrating the track | line where the fitting leg inserted next does not contact the edge of a fitting hole. The schematic diagram for demonstrating the determination method of an insertion track | orbit. The schematic diagram for demonstrating the track | orbit production | generation in case a guide exists in a fitting hole. The schematic diagram for demonstrating the shape of the workpiece | work which concerns on multipoint fitting assembly.

  Hereinafter, a calculation method of a work assembly procedure according to the present invention and an automatic assembly robot (fitting device) using the calculation method will be described. In the following description, ROM (Read Only Memory) refers to a nonvolatile semiconductor memory and a read-only medium other than the semiconductor memory. A RAM (Random Access Memory) refers to a memory that can be randomly accessed for both reading and writing, and the ROM and RAM constitute a main storage device of a computer.

  Furthermore, the yield stress is a stress value that becomes a boundary at which plastic deformation occurs when an additional stress is applied to the elastic body, and the deformation does not return to the original state even when the stress is zero. Further, the allowable deformation amount refers to the deformation amount of each fitting foot until the yield stress δmax is reached, that is, the deformation amount that the fitting foot f can deform within the elastic region.

[Configuration of automatic assembly system]
As shown in FIG. 1, the automatic assembly system S for workpieces W includes first and second workpieces W1, which are fitted to each other by fitting a plurality of fitting feet f... Into corresponding fitting holes h. This is an automatic assembly system for assembling W2. This automatic assembly system S includes a fitting supply device 21 for supplying a first workpiece (fitting object) W1, a fitting supply device 23 for a second workpiece (fitting object) W2, and the first of these. And an automatic assembly robot 1 for assembling the second workpieces W1 and W2.

  The automatic assembly robot 1 is configured by mounting a robot arm 2 on an assembly base 20, and a mating object mounting table 22 on which a second workpiece W2 is mounted is provided on the assembly base 20. Yes. Further, the fitting supply device 21 is disposed between the attachment base of the robot arm 2 and the fitting object mounting table 22 on the assembly base 20. Furthermore, a to-be-fitted material supply device 23 is provided on the opposite side of the to-be-fitted material mounting table 22 with respect to the fitting material supply device 21.

  The automatic assembly system S uses the robot arm 2 to place the second workpiece W2 on the workpiece mounting table 22, and then grips the first workpiece W1 from the fitting supply device 21. The workpiece W1 is automatically assembled to the second workpiece W2 on the fitting object mounting table.

[Configuration of automatic assembly robot]
Next, the configuration of the automatic assembly robot 1 described above will be described in detail with reference to FIGS. The automatic assembly robot 1 includes the robot arm 2, an imaging device 6 that images the first and second workpieces W1 and W2, and a robot controller (computer) 10 that controls them.

  The robot arm 2 includes an arm unit 2a formed of a vertical articulated robot, and a robot hand 3 attached to a movable end 2b of the arm unit 2a via a force sensor 5. Then, by detecting the reaction force applied to the robot hand 3 by the force sensor 5, the detected reaction force is fed back to the position control value or the torque control command value of the arm portion 2a. The delicate assembly like this can be done.

  As shown in FIG. 2, the robot controller 10 is configured by connecting the robot arm 2 and the imaging device 6 to a computer main body having a calculation device 102 and a storage device 103. Also connected to the computer main body are an input device 106 and a display device 107 for an operator to perform input work.

  The storage device 103 stores data such as shape data, physical property data, and allowable deformation amount of the fitting foot f for the first and second workpieces W1, W2. In addition, a control driver such as the robot arm 2 and the image pickup device 6, an assembly procedure calculation program CPR for causing the computer (the calculation device 102) to calculate a workpiece assembly procedure from the allowable deformation amount of the fitting foot, etc. The various programs are stored.

  More specifically, the computer main body includes the CPU 102a as a main body, and includes the image processing device 102b and the sound processing device 102c to constitute the arithmetic device 102. In addition to the image processing device 102b and the sound processing device 102c, a ROM 103a and a RAM 103b are connected to the CPU 102a via a bus 111. The ROM 103a stores programs necessary for basic control of the computer and various programs and data such as the assembly procedure calculation program CPR described above. A working area for the CPU 102a is secured in the RAM 103b. The image processing apparatus 102b controls a liquid crystal display as the display apparatus 107 in accordance with a drawing instruction from the CPU 102a, and displays a predetermined image on the screen. The sound processing device 102c generates a sound signal corresponding to the sound generation instruction from the CPU 102a and outputs the sound signal to the speaker 109.

  A keyboard 106a and a mouse 106b as an input device 106 are connected to the CPU 102a via an input interface 106c connected to the bus 111, and designation information necessary for assembling the workpiece W or other instructions can be input. It is said.

  The bus 111 is connected to the robot arm 2 and the imaging device 6, and is also connected to a recording disk reader 115. The bus 111 reads a recording medium D on which an assembly procedure calculation program CPR and the like are recorded. It can be stored. The storage device 103 described above includes a computer-readable recording medium D and other external storage devices in addition to the ROM 103a and the RAM 103b which are main storage devices.

  Further, the communication device 112 is connected to the bus 111, and the assembly procedure calculation program CPR distributed from the Internet or the like via the communication device 112 can be downloaded without using the recording medium D as described above. It is configured.

[Outline of assembly procedure]
Next, the multi-point fitting assembly procedure of the workpiece based on the assembly procedure calculation program CPR will be described. The automatic assembly robot 1 moves the robot hand 3 holding the first workpiece W1 to the assembly start position of the assembly space by the arm portion 2a of the robot arm 2, and then assembles according to a series of procedures.

FIG. 3 is a flowchart showing an assembly procedure by the automatic assembly robot 1. The assembly is roughly divided into an offline procedure calculation (orbit generation) S _off and an online control S _on of the robot arm 2. In the off-line trajectory generation, the insertion order of a plurality of fitting feet f is first determined (step S1 in FIG. 3). Next, an assembly flow for inserting the fitting feet f into the corresponding fitting holes h in the order determined in step S1 is created (S2). The assembly flow disassembles the movements performed by the automatic assembly robot 1 such as moving the first work W1 to a predetermined position, moving the first work W1 until contact with the second work W2, and controlling the trajectory for inserting the fitting foot f. Are arranged in order. Then, an insertion trajectory for each insertion order is determined at the end of the offline S _off (S3).

Next, control of the automatic assembly robot performed online (S _on ) will be described. In the online control, the robot arm 2 is operated according to the trajectory determined offline (S4). Then, during assembly in which contact between the first and second workpieces W1 and W2 occurs, a reaction force applied to the robot hand 3 is detected so that an excessive load is not applied to the first workpiece W1 and the second workpiece W2. Further, fine adjustment of the position of the first workpiece W1 is performed by the robot arm 2 (S5). Online control is constituted by these two procedures.

  Although not shown in FIG. 3, when an abnormal reaction force is detected during assembly, a retry operation may be performed by estimating the assembly state. If an abnormal reaction force is detected by the force sensor during online control, retry or assembly stop, fitting or mating object will be performed based on the threshold or reaction force profile set by the user handling the automatic assembly robot. Judge the replacement of other parts.

[Calculation method of assembly procedure]
Next, in the assembly procedure described above, the work order shown in FIGS. 5 and 6 is described with respect to the insertion order determination method (S1), the assembly flow determination method (S2), and the insertion track determination method (S3). W will be described in detail as an example.

  The work shown in FIGS. 5 and 6 includes a first work W1 that is a fitting object and a second work W2 that is a fitting object. Fitting feet f are formed, and a total of six fitting feet, fa1, fa2, fb1, fb2, fc, and fd, are provided. The second workpiece W2 includes six corresponding fitting holes ha1, ha2, hb1, hb2, hc, hd for inserting the fitting feet f of the first workpiece W1.

  Further, the dimensions of the first and second workpieces W1 and W2 are such that the length of the side where fa1, fa2, and fb1, fb2 exist is 90 mm, and the length of the side where fc and fd exist is 60 mm. All the fitting feet f are metal plates made of SUS (stainless steel) having a peg-like thin shape having a width d1 of 5 mm and a thickness t1 of 0.2 mm. Further, the length p1 of the fitting feet fa1, fa2, fb1, and fb2 is 12 mm, and the hole depth p2 of the corresponding fitting holes ha1, ha2, hb1, and hb2 is 15 mm. The length p3 of the fitting feet fc and fd is 12 mm, and the depth p4 of the corresponding fitting holes hc and hd is 15 mm.

[Determination method of mating foot insertion order]
First, a method for determining the insertion order (step S1 in FIG. 3, the step for determining the insertion order) will be described with reference to FIGS. The fitting foot f is SUS having a thickness of 0.2 mm. In the direction in which deformation is most likely to occur, the yield stress δmax is less than 100 g, and it is so soft that plastic deformation easily occurs even with the force of a human fingertip. Therefore, the insertion order of the fitting feet f is determined based on the amount of deformation until the yield stress δmax of each fitting foot is reached.

  As described above, when a stress greater than the yield stress δmax, that is, a deformation greater than the allowable deformation amount is applied to the fitting foot f, the workpiece to be assembled undergoes plastic deformation or breakage, resulting in an assembly failure. In order to prevent assembly failure, when the fitting feet f are sequentially inserted into the fitting holes h, the first workpiece is within the allowable deformation amount range of the already inserted fitting feet. It is necessary to adjust the relative position between W1 and the second workpiece W2 and insert the next fitting foot into the corresponding fitting hole.

  The range in which the automatic assembly robot 1 can move during assembly is equivalent to the minimum value of the allowable deformation amount in each direction of the inserted fitting foot f, and this is the degree of freedom during assembly of the automatic assembly robot 1. I can say that. Therefore, the movable range that can be taken during the assembly of the automatic assembly robot calculated from the allowable deformation amount of each fitting foot is defined as the degree of assembly freedom.

  In the present embodiment, the insertion order for sequentially inserting the fitting feet of the multi-point fitting assembly is determined while leaving as much as possible the degree of assembly freedom as described above. Assembling is possible while preventing assembly failure due to load and destruction.

  FIG. 4 shows a flowchart of the insertion order determination method for that purpose. In order to determine the insertion order of the fitting feet f, first, the allowable deformation amounts of the two orthogonal axes (X direction and Y direction) of the respective fitting feet, that is, the allowable deformation amount in the horizontal direction are first experimentally and actually measured. It is obtained and stored in the internal memory (storage device 103) of the robot controller 10. In this embodiment, the allowable deformation amount stores data for two orthogonal axes, but for example, the allowable deformation amount in the Z-axis direction may be added to store data for two or more axes. The first step in determining the insertion order of the fitting feet f is performed by reading (acquiring) the allowable deformation amount of each fitting foot stored in the storage device 103 (step S11 in FIG. 4). , Acquisition process).

  Table 2 shows data obtained by the computing device 102 for each of the fitting feet of the first workpiece W1 and the allowable deformation amount at least in the horizontal direction.

  Next, the computing device 102 of the robot controller 10 computes the degree of assembly freedom when these fitting feet fa1 to fd are sequentially fitted (S12, assembly degree of freedom computing step). That is, based on the acquired allowable deformation amount data, the first and second workpieces W1, W1 in each insertion step when the fitting feet are inserted in order for each of the plurality of fitting feet insertion order. The allowable relative movement amount of W2 is obtained as the degree of assembly freedom. Specifically, the degree of assembly freedom is a value obtained by extracting the numerical value of the fitting foot with the smallest allowable deformation amount in the X-axis and Y-axis directions from the combination of already inserted fitting feet. The possible combinations of the insertion order in the present embodiment are the number of fitting feet f, so that there are a total of 720 factorials.

  Here, when there is a minimum degree of assembly freedom necessary for the automatic assembly robot 1, the degree of assembly freedom obtained in the degree-of-freedom calculation step is compared with a threshold value stored in advance in the internal memory of the storage device 103. (YES in S13). Then, the insertion order that gives the degree of assembly freedom below the threshold is excluded from the options (S14 exclusion step). In addition, it is preferable that the minimum assembly freedom required for the fitting device is determined by the positional accuracy of the automatic assembly robot 1 (robot arm 2).

  For example, if the straightness of the robot arm 2 is 0.1 mm, when the next fitting foot is inserted under the condition that the degree of freedom of assembly is 0.1 mm or less, an excessive load is applied to the inserted fitting foot. There is a high probability that it will be applied. Therefore, the minimum value of assembly freedom is set to 0.15 mm. When the minimum required assembly freedom does not exist (NO in S13), the combination of all insertion orders becomes an option and the process proceeds to the next step.

  When the calculation of the assembly degree of freedom is completed, the arithmetic unit 102 calculates the sum of the assembly degrees of freedom extracted for each insertion step in each insertion order excluding the insertion order removed in the exclusion step (S15). ), And determines the insertion order that maximizes the total degree of assembly freedom (S16).

For example, the degree of assembly freedom when the fitting feet are inserted in the order of fb1, fb2, fd, fc, fa1, and fa2 is shown below (this insertion order is taken as Example 1).
Allowable deformation amount of each fitting foot (dx [mm], dy [mm]): fb1 (0.1, 2) → fb2 (0.1,2) → fd (4,0.1) → fc (1 , 0.1) → fa1 (0.2,3) → fa2 (0.2,3).
Assembly freedom level Sn [mx [mm], my [mm]] for each insertion step: second insertion steps S2 [0.1, 2], S3 [0.1, 2], S4 [0.1, 0.1], S5 [0.1, 0.1], S6 [0.1, 0.1].

  Note that when inserting the first fitting foot fb1, there is no restriction between the first workpiece W1 and the second workpiece W2, so the description of the first insertion step S1 is omitted. It can be seen from the transition of the degree of freedom of assembly in Example 1 that the degree of freedom of assembly decreases as the fitting feet are sequentially inserted, and the value becomes lowest when all the fitting portions are inserted.

  The sum of the assembly degrees of freedom at this time is ΣS = S1 + S2 + S3 + S4 + S5 + S6 = [0.5, 4.3]. In this way, the result of calculating the total sum ΣS of the assembly degrees of freedom for every combination of the insertion orders for inserting the fitting feet is as shown in FIG. Then, the arithmetic unit 102 selects an insertion order in which the total sum ΣS of assembly degrees of freedom is maximized.

As a result, in the workpiece W to be assembled in the present embodiment, the insertion order with the highest total degree of freedom of assembly ΣS is fa1 → fa2 → fb1 → fb2 → fc → fd. Therefore, this insertion order is selected and assembly is performed. The degree of assembly freedom and the total sum of the selected insertion order are as follows.
Allowable deformation amount of each fitting foot (dx [mm], dy [mm]): fa1 (0.2, 3) → fa2 (0.2,3) → fb1 (0.1,2) → fb2 (0 .1,2) → fc (1.0,0.1) → fd (4,0.1).
Assembly degree of freedom of each insertion step Sn [mx [mm], my [mm]]: S2 [0.2, 3], S3 [0.2, 3], S4 [0.1, 2], S5 [0 .1, 2], S6 [0.1, 0.1].
Total of assembly freedom: ΣS = S1 + S2 + S3 + S4 + S5 + S6 = [0.7, 10.1].

  Here, when the transition of the degree of freedom during assembly of the insertion order of Example 1 shown above and the selected insertion order is shown in FIG. 8 and compared, the selected insertion order is free to assemble until the second half of the assembly. You can see that the degree is high.

Also, as a reference, the assembly freedom when assembled in another insertion order fd → fa1 → fa2 → fb1 → fb2 → fc and the total of the assembly degrees of freedom are shown (this insertion order is Example 2). This insertion order is for each of the six fitting feet, focusing on the axial direction with the highest allowable deformation amount, and fitting the fitting feet in the descending order of the allowable deformation amount.
Allowable deformation amount of each fitting foot (dx [mm], dy [mm]): fd (4,0.1) → fa1 (0.2,3) → fa2 (0.2,3) → fb1 (0 .1,2) → fb2 (0.1,2) → fc (1,0.1).
Assembly degree of freedom of each insertion step Sn [mx [mm], my [mm]]: S2 [4, 0.1], S3 [0.2, 0.1], S4 [0.2, 0.1] , S5 [0.1, 0.1], S6 [0.1, 0.1].
Total of assembly freedom: ΣS = S1 + S2 + S3 + S4 + S5 + S6 = [4.6, 0.5].

  FIG. 9 shows the transition of the degree of freedom during the assembly of Example 2 and the selected insertion order. In Example 2, the degree of assembly freedom is greatly reduced in Step S2. If fitting feet in which the axial direction with a large allowable deformation amount and the small axial direction are orthogonal (opposite) are inserted in the initial insertion step, the workpiece cannot be moved in either of the two axes. From this, it can be reconfirmed that, in determining the insertion order, it is preferable to evaluate and determine the allowable deformation amount by using a vector in consideration of the direction (orientation) as in the present embodiment instead of a scalar.

  In the selected insertion order, since the fitting feet fa1 and fa2 and fb1 and fb2 have the same allowable deformation amount, the sum of the assembly degrees of freedom does not change regardless of which is inserted. In such a case, the insertion order of the fitting feet fa1 and fa2, and fb1 and fb2 may be from either. If priority is given to tact, it is also possible to insert fitting feet fa1 and fa2 and fb1 and fb2 at the same time. However, if importance is attached to the reliability and certainty of assembly, it is better to insert them in order.

  As described above, the automatic assembly robot 1 determines the total assembly freedom obtained in each insertion step for each of the plurality of insertion orders of the fitting feet, thereby assembling the workpieces in the entire insertion order. The degree can be evaluated. Then, one of the order of insertion in which the sum of the assembly degrees of freedom is a predetermined value or more (in the present embodiment, the sum of the degrees of freedom is the maximum) is selected from the total of the obtained plurality of assembly degrees of freedom. (S15, 16, evaluation process, insertion order selection process). Accordingly, an insertion order in which the relative movement amount of the first work W1 with respect to the second work W2 can be selected throughout the entire insertion order, and a small product with relatively low rigidity of the fitting foot f can be selected. The point fitting assembly can be automated by the automatic assembly robot 1.

  Next, a method for determining the insertion order of the second fitting feet, which is different from the method for determining the insertion order of the fitting feet described above (method for determining the insertion order of the first fitting feet), will be described. The method for determining the insertion order of the second fitting feet is a method of selecting and assembling the order in which the degree of assembly freedom when inserting the last fitting feet is the maximum, and until step S14, Since it is the same as the determination method of the insertion order of 1 fitting leg, the description is abbreviate | omitted.

  As shown in FIG. 10, the determination method of the insertion order of the second fitting feet is the same as the determination method of the insertion order of the first fitting feet. First, an acquisition step (S11), an assembly freedom degree calculation step (S12) and an exclusion process (Yes of S13, S14) are performed. After these steps, the order of insertion of the plurality of fitting feet is evaluated based on the degree of freedom of assembly when the last fitting foot is inserted, and the degree of assembly freedom when the last fitting foot is inserted is One of the insertion orders of a predetermined value or more is selected (S18, evaluation process, insertion order selection process). Specifically, in the present embodiment, the degree of freedom of assembly when the arithmetic device 102 inserts the sixth fitting foot described in the sixth insertion step S6 of FIG. 7 is maximized. Select the insertion order.

  At this time, if there are multiple combinations of the order in which the degree of assembly freedom when inserting the last fitting foot is maximum, the insertion step is moved back one by one until one insertion order is determined, and the degree of assembly freedom is compared. It is preferable. In the present embodiment, when the method for determining the insertion order of the second fitting legs is used, the selected insertion order is fa1 → 1a2 → fb1 → fb2 → fc → fd.

  As described above, there are two main methods for determining the insertion order from the combination of insertion orders. One is a method of selecting and assembling an order in which the total sum of degrees of freedom of assembly in each insertion step until all the fitting feet shown in the flowchart of FIG. 4 are inserted is selected. The other is a method of assembling by selecting an order in which the degree of freedom in assembly when the last fitting foot shown in the flowchart of FIG. 10 is inserted is maximized.

  Although the same insertion order may be derived by the two determination methods as in the present embodiment, if different insertion orders are determined, the following selection may be made. If the position control accuracy of the robot arm 2 is small with respect to the degree of assembly freedom, the method for determining the insertion order of the first fitting feet is selected, and the position control accuracy of the robot arm 2 with respect to the degree of assembly freedom is afforded. If there is, it is preferable to select a method for determining the insertion order of the second fitting feet. In other words, the method for determining the insertion order of the first mating feet is a difficult assembly that requires assembly accuracy, and the method for determining the insertion order of the second mating feet is an assembly method that takes into account tact. It ’s better to choose when.

  When comparing the size using the degree of freedom of assembly, if there is no insertion order that maximizes the degree of freedom of assembly in the X-axis direction and the degree of assembly freedom in the Y-axis direction, the degree of freedom of assembly in the X-axis direction. It is preferable to take the product of the degree of freedom of assembly in the Y-axis direction and select the insertion order that maximizes the product.

  Some of the workpieces to be assembled are designed with clearance between the fitting feet and the fitting holes. Such dimensional play due to clearance can also be considered as the degree of freedom of assembly of the workpiece. Therefore, in this embodiment, the degree of freedom of assembly is obtained by calculating only from the allowable deformation amount of each fitting foot.However, the degree of freedom of assembly is calculated from the value obtained by adding the size that becomes play for the clearance to the allowable deformation amount. You may make it calculate. In that case, it is necessary for the user to input the dimensions of the clearance between the fitting foot and the fitting hole in advance to the storage device 103 of the robot controller 10 so that they can be used for calculation of the degree of assembly freedom. Further, when the allowable deformation amount is actually measured, it is preferable that the allowable deformation amount is measured by actual measurement including the dimensional play due to the clearance.

  Further, in the present embodiment, an evaluation step for evaluating the insertion order of a plurality of fitting feet based on the obtained degree of assembly freedom, and an insertion for selecting one insertion order from the insertion order of the plurality of fitting feet. Although the order selection step is performed simultaneously, it may be performed separately. Further, the insertion order selection step may be performed manually.

  In addition, as for the insertion order of the fitting feet whose assembly freedom is equal to or less than a predetermined threshold, the excluding step of excluding from the insertion order selection candidates may be performed any time before the insertion order selecting step, for example, An exclusion process may be incorporated into the assembly degree of freedom calculation process.

[Method of determining assembly flow]
Next, an assembly flow determination method (step S2 in FIG. 3, assembly flow determination step) will be described. The assembly flow of the workpiece W includes adjusting the relative position of the first workpiece with respect to the second workpiece, moving the first workpiece so that the first workpiece contacts the second workpiece, and controlling the trajectory for inserting the fitting foot. The movements performed by the automatic assembly robot 1 are disassembled and arranged in order.

In the following description, the side with the fitting feet fa1 and fa2 of the first workpiece W1 is the first side, the side with the fitting feet fb1 and fb2 is the second side, and the side with the fitting feet fc is the third side. The side having the side and the fitting foot fd is referred to as a fourth side. Moreover, if each side of the first workpiece W1 is defined in this way,
In the present embodiment, the fitting foot f of the first workpiece W1 is set to the first side, the second side, the third side, and the fourth side based on the insertion order determined in the insertion order determining step (S1 in FIG. 3). In the order of the sides, they are inserted into the corresponding fitting holes h of the second workpiece W2.

  First, a procedure for inserting the fitting feet fa1 and fa2 on the first side into the fitting holes ha1 and ha2 will be described with reference to FIG. First, the X-axis positions of the fitting feet fa1 and fa2 and the fitting holes ha1 and ha2 are aligned. At the same time, the height of the front ends of the fitting feet fa1 and fa2 on the first side is moved to a height position above the fitting holes ha1 and ha2. The Y axis moves in parallel to the positions of the fitting holes ha1 and ha2. At this time, the fitting feet fa1 and fa2 are inclined several degrees with respect to the fitting holes ha1 and ha2. Regarding the angle, it is assumed that the tips of the fitting feet fb1 and fb2 on the second side do not contact the fitting holes hb1 and hb2.

  Next, the fitting feet fa1 and fa2 are translated in the Y-axis direction until the tips of the first-side fitting feet fa1 and fa2 come into contact with the side surfaces of the fitting holes ha1 and ha2, and at the same time, the Z-axis direction is also lowered. Move in the direction. Although the contact is detected by the output value of the force sensor 5, since a time delay occurs in the control of the automatic assembly robot 1, the smallest possible value is taken into consideration in consideration of the speed of the robot, the sensor information acquisition cycle, and the sensor accuracy. Set as a haptic threshold for detecting contact. Next, it inserts in a Z-axis direction to the 2nd side insertion start position, and controls a Y-axis. At this time, compliance control is performed in the X-axis direction to prevent the fitting feet fa1 and fa2 from being broken. Since the two fitting feet fa1 and fa2 on the first side have the same allowable deformation amount, they are simultaneously inserted to shorten the tact time required for assembly.

  Next, a procedure for inserting the fitting feet fab1 and fb2 on the second side into the fitting holes hb1 and hb2 will be described with reference to FIG. The second side is inserted while the fitting feet fa1, fa2 on the first side are inserted into the fitting holes ha1, ha2. First, the fitting feet fb1 and fb2 are rotated around the X axis around the insertion position of the fitting feet fa1 and fa2 on the first side. At this time, “the rotation trajectory based on the insertion position of the fitting feet fa1 and fa2” and “the trajectory in which the fitting feet fb1 and fb2 to be inserted next do not contact the edges of the corresponding fitting holes hb1 and hb2”. A combined track is created and the fitting foot is controlled in the Y-axis direction. The method for generating the trajectory will be described later. At this time, compliance control is performed in the X-axis direction to prevent destruction of the fitting foot f. Since the two fitting feet fb1 and fb2 on the second side have the same allowable deformation amount, they are simultaneously inserted in order to shorten the tact time required for assembly.

  Next, a procedure for inserting the fitting foot fc on the third side into the fitting hole hc will be described with reference to FIG. In the insertion of the fitting foot fc on the third side, there is no inserted fitting foot fd on the fourth side which is the opposite side, so the inserted fitting feet fa1, fa2, inserted on the first side and the second side. The insertion takes into account only fab1 and fb2. Here, the fitting foot is rotated with the tip of the fitting foot fd on the fourth side as the center. Further, it is necessary to take a track so that the tip of the fitting foot fc on the third side does not collide with the edge of the corresponding fitting hole hc. Therefore, “the rotation trajectory based on the insertion position of the fitting feet fa1, fa2, fb1, fb2” and “the trajectory in which the next fitting foot fc to be inserted does not contact the edge of the corresponding fitting hole hc” are combined. Make a trajectory. Then, the fitting foot fc is controlled in the X-axis direction based on the combined trajectory. Similar to the second side, the method for generating the trajectory will be described later. Furthermore, compliance control is performed in the Y-axis direction to prevent destruction of the inserted fitting feet fa1, fa2, fb1, and fb2.

  Next, a procedure for inserting the fitting foot fd on the fourth side into the fitting hole hd will be described with reference to FIG. First, the fitting foot fd is rotated around the Y axis around the insertion portion of the fitting foot fc on the third side. At this time, a trajectory is formed by synthesizing the “rotating trajectory generated from the fitting foot fc as a starting point” and the “trajectory in which the next fitting foot fd to be inserted does not contact the edge of the corresponding fitting hole hd”. The fitting foot fd is controlled in the direction. The method for generating the trajectory will be described later as in the second side and the third side. Furthermore, compliance control is performed in the Y-axis direction to prevent destruction of the inserted fitting feet fa1, fa2, fb1, fb2, and fc.

  As described above, in the assembly flow determination step, the insertion operation of the fitting feet is determined for each side of the workpieces W1 and W2 having the fitting feet f.

[How to determine the insertion trajectory]
Finally, the method for determining the insertion trajectory (step S3 in FIG. 3, trajectory calculation step) will be described in detail. FIG. 13 is a diagram for explaining a “trajectory in which no load is applied to the inserted fitting foot”. In order to prevent plastic deformation or destruction of the fitting foot during assembly, when the fitting feet f are sequentially inserted into the fitting holes h, the load applied to the already inserted fitting feet fd is reduced. It is necessary to insert the next fitting foot fn on the track.

  For example, in FIG. 13A, there are already inserted fitting feet fa1 and fa2, and the next fitting feet to be inserted are fitting feet fb1 and fb2. When the fitting feet fb1 and fb2 are inserted into the fitting holes hb1 and hb2, not only the relative positional deviation between the fitting feet and the fitting holes is corrected, but also the inserted fitting feet are plastically deformed or broken. It is necessary to control the position of the top plate TP of the first workpiece W1 in such a path that does not apply a load.

  Therefore, in the present embodiment, when the next fitting foot fn is inserted into the fitting hole hn, the robot arm 2 generates a trajectory for moving the top plate TP of the first workpiece W1 by combining the following two trajectories. Generate. Of these two combined tracks, the first track O1 is “the track calculated based on the load on the inserted fitting foot”, specifically, “in the vicinity of the inserted fitting foot”. It can be said that the trajectory is generated starting from an arbitrary reference point. In other words, it can also be said to be a “trajectory generated when the first or second workpiece is rotated around the reference point and the inserted fitting foot is moved to the target position”. Furthermore, the second track O2 is a “track that can be inserted without bringing the inserted fitting foot into contact with the edge of the corresponding fitting hole at the tip portion”.

  FIG. 13B shows an example of the first trajectory O1. When the fitting foot f is inserted, a restriction is generated between the inserted fitting foot f and the fitting hole h. One point is determined as a reference point based on the position of the already inserted fitting foot fd, and the reference point is determined as follows.

  When there is one inserted fitting foot fd, it is set as a point near the contact point where the fitting foot fd contacts the fitting hole hd. Further, when there are two inserted fitting feet, it is set as a point near the intermediate position between the two fitting feet. Furthermore, when there are three or more inserted fitting feet, the points are in a region surrounded by the inserted fitting feet or in the vicinity of the region. Therefore, the object shapes of the fitting feet f and the fitting holes h, and the distances between the fitting feet and the fitting holes need to be input as data to the storage device 103 of the robot controller 10 in advance.

  For example, it is assumed that the fitting foot is one of the fitting feet fa1 after being inserted in FIG. 13A, and the fitting foot to be inserted next is the fitting foot fb1. The reference point P at that time is the intersection of the virtual position of the fitting foot fa when the fitting foot fa is not stressed (dotted line n in FIG. 13A) and the inner wall Ia1 of the fitting hole ha1. Determine. The rotation trajectory centered on the reference point P shown in FIG. 13B is a trajectory O1, in which a load is not easily applied to the inserted fitting foot fa1, that is, the first trajectory.

  However, as can be seen from FIG. 13B, the track O1 is in contact with the upper edge UE of the fitting hole hb1, and the fitting foot fb1 cannot be smoothly inserted by interfering with the second workpiece W2 as it is. . As described above, in the rotation trajectory based on the insertion point of the inserted fitting foot, a trajectory that cannot be inserted due to interference with the edge of the fitting hole may be generated. Therefore, the second track “the track in which the tip of the fitting foot to be inserted next does not contact the edge of the corresponding fitting hole” is synthesized with the first track to generate the final track O3.

  The generation of the second trajectory O2 will be described with reference to FIG. The second track O2 is a “track where the tip of the fitting foot fn to be inserted next does not contact the edges A and B of the corresponding fitting hole hn”. First, Start and Goal are determined with respect to the track of the tip of the fitting foot fb1 (fb2). Next, in order to plan an insertion trajectory that avoids the upper edge UE of the fitting hole hb1 (hb2), which is an obstacle viewed from the fitting foot fb1 (fb2), the potential according to the shape of the insertion hole fb1 (fb2) Set the field. The potential field is a virtual gravity field in which terms of attractive force and repulsive force are defined by mathematical formulas, and potentials at arbitrary coordinates on a plane or space are converted into MAP and converted into MAP. As for the attractive force and repulsive force for creating the potential field, since the shape of the object shape of the fitting foot and the fitting hole is known, an attractive force is generated from the Goal point with respect to the determined Start point, while the fitting hole Set to generate repulsion with the upper edge of the obstacle as an obstacle. The points where the repulsive force is generated are shown as points A and B in FIG.

  In the example of FIG. 14A, repulsive force is generated at two points of the fitting hole, point A and point B. However, the repulsive force is set so as to be generated from the entire plane serving as the upper edge UE of the fitting hole. May be. An example of the potential map created in this way is shown in FIG. The point potential described as “Start” is the highest, and the potential at the point described as “Goal” is the lowest. The trajectory at the tip of the fitting foot fb1 to be inserted next moves from the Start point to the Goal point in order from the high potential to the low potential.

  Even if the fitting foot comes into contact with the inner wall of the fitting hole, the fitting foot fb1 (fb2) is inserted along with the inner wall with moderate elastic deformation following the inner wall. Do not define repulsion. Further, the Goal point is a point that becomes the center of the width and thickness direction of the fitting hole.

  Such a potential field can be defined by the following equation.

ここで、式１．１はＧｏａｌ地点ｑ_ｇｏａｌから空間上の任意の点ｑへ働く引力、式１．２は、障害物ｑ_{ｏｂｓｔａｃｌｅ}から空間上の任意の点ｑへ働く斥力、式１．３は式１．１と式１．２の和をとった任意の位置ｑのポテンシャルを示す式である。また、ｑ、ｑ_ｇｏａｌ、ｑ_{ｏｂｓｔａｃｌｅ}は平面上の座標（ｘ，ｙ）を表し、ρは２点間の距離ｄを計算する関数である。

Here, the expression 1.1 is an attractive force that works from the _goal point q _goal to an arbitrary point q in the space, and the expression 1.2 is a repulsive force that works from the _obstacle q _obstable to an arbitrary point q in the space, an expression 1.3 Is an expression showing the potential at an arbitrary position q, which is the sum of Expression 1.1 and Expression 1.2. Further, q, q _goal , and q _obstable represent coordinates (x, y) on a plane, and ρ is a function for calculating a distance d between two points.

For example, ρ (q, q _goal ) represents the distance between two points q and q _goal . α, m, β, and n represent the inclination and height of the potential field, and are coefficients that must be determined by the user. In the present embodiment, the first trajectory O1 is used when determining these four parameters α, m, β, and n. The trajectory determined as a result is a trajectory O3 in which the first trajectory O1 and the second trajectory O2 are combined.

  The above four parameters α, m, β, n determine the difference L between the first trajectory O1 and the second trajectory O2 created with arbitrary α, m, β, n as an evaluation value. As shown in FIG. 15, the difference L between the first trajectory O1 and the second trajectory O2 is calculated at each point from Start to Goal, and summed to calculate ΣL. Four parameters α, m, β, and n are calculated so that ΣL becomes small.

  In addition, various search calculations and convergence calculations can be considered as a method for obtaining these four parameters α, m, β, and n. Here, a genetic algorithm is used as one of the calculation methods. The genetic algorithm is a probabilistic solution search method, and is a representative method that can obtain a suboptimal solution relatively easily in real time. Since the calculation method of the genetic algorithm is a known content and is not related to the present invention, the description is omitted.

  From the potential field created by α, m, β, n determined by the above method, the trajectory drawn by the tip of the fitting foot f1b is finally determined. By inserting the next fitting foot fn into the fitting hole hn by this track, it becomes possible to smoothly insert the next fitting foot without applying an excessive load to the inserted fitting foot fd. As a result, destruction of the workpiece W during assembly is prevented.

  Some of the second workpieces W2 are provided with guides 30 for inserting fitting feet as shown in FIG. 16 into the fitting holes h. The insertion trajectory of the fitting foot f with respect to the second workpiece W2 having such a design is generated as shown in FIG. The difference from the case where there is no guide is that the position of the obstacle when the potential field is created is set at a point C in the figure on the upper edge of the guide.

  The generated track of the fitting foot tip forms a track that is inserted into the fitting hole h following the guide 30 while contacting the guide 30 as indicated by a dotted line in FIG. be able to. By creating a track in this way, the position accuracy requirement at the time of insertion can be reduced and insertion can be performed more easily. The above is the method for determining the insertion trajectory (S3).

  If the above-described insertion order determination (S1), assembly flow determination (S2), and insertion trajectory determination (S3) are completed off-line, the robot controller 10 controls the robot arm 2 and assembles in order. Perform the action. During actual assembly, a fine correction of the trajectory is performed online by compliance control based on the detection value of the force sensor 5 with respect to the insertion trajectory determined by the determination of the insertion trajectory.

[Summary]
Thus, according to the present embodiment, the degree of freedom of assembly in which the first work W1 can move relative to the second work W2 is obtained in the direction in which the first work W1 moves horizontally, and the obtained degree of freedom of assembly is obtained. Based on this, the insertion order of the fitting feet f is evaluated. Thereby, even if the strength of the fitting foot f is low, the multipoint fitting assembly can be performed without destroying the fitting foot f already inserted during the assembly.

  Further, when calculating the insertion trajectory of the fitting foot inserted after the second insertion step, “the first trajectory calculated based on the load on the inserted fitting foot” and “inserted” The 2nd track | orbit which can be inserted without making a fitting leg contact the edge of the fitting hole to which the front-end | tip corresponds is synthesize | combined. Then, by controlling the position of the first workpiece W1 with respect to the second workpiece W2 according to the synthesized track, the fitting foot fn to be inserted can be smoothly inserted into the fitting hole hn. Further, it is possible to reduce the load generated on the inserted fitting foot fd.

  In the process of inserting the fitting foot f into the fitting hole in combination with the evaluation of the insertion order of the fitting feet f and the calculation of the trajectory, the inserted fitting feet are excessively deformed or broken. It is possible to prevent the load from being applied and to enable reliable assembly.

  In the present embodiment, the fitting foot f is formed in the first workpiece W1, and the fitting hole h is formed in the second workpiece W2. The fitting foot f and the fitting hole h are the first and first fittings. It may be formed on either of the two workpieces W1 and W2, and naturally it may be mixed or present. In this case, the robot arm 2 grips one of the first and second workpieces W1 and W2 and assembles it to the other workpiece placed.

  Further, the material of the fitting foot is not limited to SUS, and any material such as plastic may be used. The shape is not limited to a plate-like peg or circular boss, but groove fitting, dowel fitting, etc. Various fitting forms can be applied.

  Further, the insertion order selected in the insertion order selection step is not limited to the sum of the assembly degrees of freedom / the assembly freedom degree when the last fitting foot is inserted is not the maximum value, and the automatic assembly robot 1 performs the multipoint fitting. Any insertion order may be selected as long as the insertion order is within a range of a predetermined value that can be combined.

  As for the allowable deformation amount of the fitting foot f, physical dimension data of the fitting foot f and physical property data such as Young's modulus are stored in the storage device 103, and the arithmetic device 102 calculates the degree of freedom of the fitting foot. You may ask for it. When this method is used, the degree of freedom can be obtained without transferring labor by transferring CAD data for manufacturing a workpiece. In addition, the robot controller 10 may have an offline calculation function in a separately configured server machine.

  Further, when the fitting foot f is inserted into the fitting hole h, if the insertion length of the fitting foot f is not inserted into the fitting hole h, only a part of the insertion length is inserted shallowly. Good. By maintaining the above state and inserting the next fitting foot, the restraint between the fitting and the fitting object generated between the already inserted fitting foot and the fitting hole is held in a looser state. be able to. That is, the insertion operation by the robot can be further facilitated.

1: automatic assembly robot, 2: robot arm, 102: arithmetic unit, l: fitting foot, W1: first workpiece, W2: second workpiece, O1: first track, O2: second track, P: reference point , CPR: assembly procedure calculation program, S3: trajectory calculation step, S11: acquisition step, S12: assembly degree of freedom calculation step, S14: exclusion step, S16, 18: evaluation step, insertion order selection step

Claims (9)

The plurality of fittings provided on at least one of the first and second workpieces among the first and second workpieces that are assembled to each other by fitting the fitting feet into the corresponding fitting holes. For each leg, an acquisition step of acquiring data obtained for at least the horizontal direction, an allowable deformation amount that is a deformation amount that can be deformed within the elastic range;
The arithmetic unit, based on data of the acquired tolerance deformation amount, for each of the insertion order of the plurality of the Hamagoashi, allowed the first and second workpiece when inserting the Hamagoashi sequentially a computation step of Ru determined as an assembly freedom relative movement amount, with a,
A workpiece assembly procedure calculation method characterized by the above. The arithmetic device further comprises an insertion order selection step of selecting one of the insertion orders from the degree of assembly freedom determined in the arithmetic step for each of the insertion orders of the plurality of fitting feet.
The work assembly procedure calculation method according to claim 1. The arithmetic unit, for each of the insertion order of the plurality of Hamagoashi, the assembly flexibility, insertion order selected to select the insertion order is a predetermined value or more when the last Hamagoashi is inserted Further comprising a process,
The work assembly procedure calculation method according to claim 1. The arithmetic unit, before the insertion sequence selection step, the assembly flexibility further comprising an insertion sequence to exclude exclusion step is below a predetermined value,
The work assembly procedure calculation method according to claim 2 or 3. Inserting said calculation device, without contact with the first track that is calculated based on the load on the insert already fitted foot, the Hamagoashi to be inserted with the edge of the fitting hole in which the tip portion corresponding A trajectory calculation step of combining and calculating a possible second trajectory;
The workpiece assembly procedure calculation method according to any one of claims 1 to 4. The first trajectory is a trajectory generated when the first or second workpiece is rotated around a reference point to move the inserted fitting foot to a target position.
The reference point when the already inserted Hamagoashi is one, the already inserted Hamagoashi the fitting hole and contacts contact point, when the already inserted Hamagoashi is two the point of the intermediate position between the two engagement legs of the insertion already, when the already inserted Hamagoashi is 3 or more, in terms of the area surrounded by the already inserted mating legs is there,
The work assembly procedure calculation method according to claim 5.   An assembly procedure computation program for causing a computer to execute the workpiece assembly procedure computation method according to any one of claims 1 to 6.   7. A component manufacturing method for manufacturing a component by assembling the first and second workpieces by an automatic assembly robot according to an assembly procedure calculated by the workpiece assembly procedure calculating method according to claim 1. A robot arm that grips one of the first and second workpieces that are assembled to each other by fitting the corresponding fitting holes into the corresponding fitting holes, and is assembled to the other workpiece placed;
Based on the data of the deformable deformation amount of an elastic region in the horizontal direction of the plurality of Hamagoashi, for each of the insertion order of the plurality of the Hamagoashi, the first when inserting the Hamagoashi sequentially and determined Mel computing device 1 and allowed the relative amount of movement of the second work as an assembly flexibility, with a,
An automatic assembly robot characterized by that.

Images