JPS611100A - Method of inserting electric element into printed circuit board - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to automated assembly equipment, particularly a hierarchical modular software system for controlling and controlling sensor-based circuit boards for assembly of printed circuit boards into factory loads.
Related to robot systems.

Automatic assembly devices for printed circuit boards are already known;
They typically include some form of gripper to support the device to be inserted into the circuit board, and utilize some method of sensing the position of the device lead relative to the insertion hole. However, no known system is capable of automatically correcting insertion device position errors through sensor-based software operations. Therefore, in known systems, the lead or both the lead and the hole had to be previously shaped with high precision so that a relatively high clearance was obtained between the hole and the lead.

A principal object of the present invention is to provide an improved method of inserting electrical components into printed circuit boards.

The present invention provides a sensor system for assembling printed circuit boards under the control and monitoring of a hierarchical modular software system.
Consists of a base robot system. Specifically, the system utilizes a suitable grip to obtain elements containing pre-cut and shaped leads. After acquiring the elements, the vision system is utilized to sense the position of the individual element leads relative to the robot. Using this information, the robot positions the device above the circuit board, near the insertion location. The sensor determines the position of the lead relative to the hole by looking through the hole and generates a signal, and the robot corrects the position of the lead in accordance with this signal to reduce the deviation between the hole and the lead. This process is performed individually for each lead of the multi-lead device. Once the device leads are properly inserted into the holes, they are cut and shrunk to temporarily hold the device until it can be permanently secured by soldering or other suitable method.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings.

The need for factory automation led to the development and implementation of automatic assembly departments. Specifically, the automated assembly thus constructed and tested was a sensor-based robot operating under the control and supervision of a hierarchical modular system.
・The system involved assembling printed circuit boards. Automated assembly stations are particularly utilized in the subprocess of printed circuit board wiring assembly. Such systems are necessarily complex due to the large number of variables involved in the assembly process.

Although the system actually constructed was intended for printed circuit board assembly, sensor-based robotic assembly systems have much broader applications.

The entire system is sensor based, with the controller reacting to the various sensors such that each lead of the element is inserted individually. The entire system is computer-aided design (C
AD) Data-driven and adaptable to any printed circuit board format based on the content of data files supplied from a work control center.

The high-level language used to program the digital fist computer as a control unit allows for real-time updating of data. Software configurations include programs that link automated assemblers with automated industrial work centers, calibration routines, insertion sequence programs, and other miscellaneous routines, as well as inks that enable real-time insertion of leads or discontinuous elements in an unattended environment. Provide FACE. The automatic assembly control device is Hewlett Pa.
The robot part of the system, which is a ckard A300 memory-based computer, is Auto■at.
ix AIDv 800 robot and its associated control device are used.

Visual ability is automatic using two TV cameras
ix Auto Vision 4 vision system. Attach one of the TV cameras and the lead shrink cutter to the Aerotech XY table. The programmable end effector that attaches to the end of the robot arm was specially designed by Westinghouse. However, any other suitable end effector handling this type of electrical element may be used, such as an automatic assembly control computer (
All HP aeoo software is used to monitor and control the device insertion process and to interface with automated work center computers. Communication to the robot and vision controller is via an R5-232 link, and control to the XY table and retraction cutter mechanism is via separate lines utilizing parallel interface sections.

The work control computer software is hierarchical, i.e.
Configure the flow of control to move from more general purpose programs to specialized programs for specific devices or functions. To maintain modularity and flexibility, subliminary libraries are utilized to allow all but most specialized programs to view external devices as "black-boxes." In addition, an interface for asynchronous message transfer to the work center allows the work department to concentrate on circuit board assembly relatively independently of events occurring in other parts of the automated factory.
Also includes programs.

In a typical manual production environment, data comes from nearly every assembly drawing and bill of materials; assembly workers must match these documents and records to materials as they progress through the assembly process. Must be.

In an automated environment, not all data is available in electronic form, so some form of "advanced" system is required. This "advanced" function is provided by the work control center in the environment assumed by the automatic assembly section, which is the subject of the present invention. All the information comes from various files in different formats, and is ultimately unified by the work control center into a format that can be processed by the automated assembly department.

In practice, a technical data library containing all the necessary information was stored on an IBM 4341 computer. This information includes the dimension "dollar code" as well as an element characteristics file containing all other information pertinent to and limiting the assembled element. Other files also include electrical testing information and technical specifications for the materials. In short, this data library contains all the data needed to assemble a particular circuit board.

Perforated tapes for automatic drilling of printed circuit boards are used in conjunction with this data library. This perforated tape was created independently on a separate computer. Additionally, computer vision design drawings in digital format are also utilized.

After collecting this data, run a program on the IBM 3031 to check the X
Create data for the automated assembly including Y position, intended insertion sequence, cut and clinch information for each element.

In this processing cycle, Materia+Aquis
ition/Robotic Kitting (MAR
K) All the code needed for the system is also created, called modules. Based on this information MA
The RK system assembles element kits for circuit board assembly using robot operations.

The elements included in this element kit also include leads that are pre-cut and curved into the desired shape. After cutting and bending the leads of the elements, the elements are stored in standardized element racks to form a kit that is used by an automatic assembly section when assembling printed circuit boards. It also creates a list indicating parts shortages and sends it back to the work control center and, in some cases, to the automatic assembly department.

When the MARK system assembles the component kit, including component shortage information, and sends it back to the work control center, the work control center sends all the data necessary for assembling a specific circuit board to the automatic assembly department, which is the subject of this invention. It becomes ready to send. In the automated assembly section, an insertion sequencer program executed by the work station controller sends the physical characteristics of the part, the pick-up location, and insertion information of the part to the robot controller, based on which the robot inserts the part from the kit. Start acquisition. The robot control device monitors four sensors installed on the end effector,
Indicates the presence or absence of a part and allows the robot to automatically adapt to part size variance. The part acquisition process is performed exclusively by the robot controller, and the work part controller and the vision system controller can perform their own control and monitoring functions. Upon component acquisition, the insertion sequencer program communicates with the XY table and vision control program for the system to accurately locate the center of the hole in the printed circuit board into which the device lead is inserted. Because the circuit board holes are located for each component, the positions of the holes and device leads can be updated in real time even if the circuit board should become misaligned during the insertion process. During the hole positioning process, a camera is placed below the circuit board and backlit from above the circuit board. The target position given by the robot command from the work center is corrected using an offset. The robot controller sends this update data to control the robot so that the insertion cycle can continue smoothly. At this point, control passes to the robot and vision system, which executes a tight servo competition loop to insert each element lead.

The device leads are cut to different lengths so that they can be inserted one at a time. Based on the difference in length, the vision system uniquely identifies each lead and its location. When installing a circuit board, a robot controller monitors a load cell mounted on a pallet and indicates when a lead comes into contact with the circuit board.

The robot uses feedback from the lead softness sensor and visual data to detect inaccurate part information and gripper
- Attempts to correct misalignment of the lead with respect to the finger and lead insertion hole. At the end of each insertion cycle, the robot controller transfers information reflecting the use of the correction technique to the workstation control computer for statistics. This statistical data is transferred to the work control center as historical data.

After the robot confirms that the margin element has been correctly inserted, the insertion sequencer routine works with the work station control computer to begin cutting and clinching the lead. During cutting and clinching, the robot controller sends statistical information obtained during the insertion process, and the workstation control computer in turn sends a set of element acquisition information to the robot control computer.

Once the cutting and clinching is complete, the next element insertion cycle begins, and so on, until all elements have been processed.

Both the workstation computer software and the robot and vision control software are designed to be adaptable to the production of any number or type of circuit board. All that is needed to begin automated assembly of printed circuit boards is a component kit, a circuit board mounted on a pallet that is inserted into a robot, and a data file from a work control center. Therefore, work orders for only one circuit board are possible, and high priority work can be introduced into the production process.

H when creating work department computer software
It takes advantage of several features of the PA 800 operating system. As control is transferred between several programs, so are the demands on state, instructions, and system constants. Since the host computer is memory-based, there can be no communication between processes via files; inter-process communication occurs in three ways. Status bits, error indicators, and constants that transition between programs use a portion of shared memory called "system common." Each program contains an offset map, without knowing the layout of that portion of shared memory. Software can query variables by logical name. Similarly, a section of general purpose memory called EN is utilized to store all data necessary to configure a particular circuit board style. Thus, the data download begins upon the beginning of the circuit board construction process, and the insertion sequencer-Φ program can easily interrogate this information, requiring only memory recall time.

The operating system is also used for inter-process communication, and C1ass Ilo is used for intra-station program communication.
A function called is used. Hewl for internodal commensurability
ett Packard's interprogram communications software module is utilized. C1ass Ilo allows multiple programs to communicate asynchronously. For example, if one program wants to write a message to another program, it can ensure that the message is queued and then processed.

If the receiving program wants this information, it can receive it.

The status code reflects whether there is a message waiting, so the receiving program (reading)
No waiting for IO.

Hewlett Packard's program-to-program communication package is similar, but there is a pair of dedicated programs for both the zero work unit control device and the work center control device, which differ in that they operate across a computer network.

The master program at the work station controller sends messages to the slave program at the work center controller, and vice versa. Messages from the work control center can be processed by the work station controller at logically convenient points during the circuit board configuration process, thereby reducing the total communication time or the need for "mailbox tube polling" or file communication. will disappear.

The software also utilizes a wide range of error recovery techniques. This technique is designed to be performed in an unattended environment and the objective is to achieve the maximum number of insertions in the minimum cycle time. By using this technology, we can construct a system that can store a large number of variables.
It has been used with favorable results. By employing calibration technology, the mechanical slippage of the various mechanical elements of the system,
Circuit board distortions and misalignments can be automatically corrected. Additionally, the visual lead insertion method allows for wide tolerances not only in the lead formation process but also in the device kit construction process. Furthermore, by adopting a lead-based discrimination method for these elements, the elements acquired by the robot can be used with minimal control over the lead tip position. This discrimination method allows the robot to calculate the XY and Z position of the lead tip relative to the robot and compensate for inaccurate or misshaped leads. By using this method, the work could be carried out without any problem even if the difference in diameter between the lead and the hole was only 0.005 inch.

In actual operation, circuit boards are assembled by using the automated assembly unit and assembly process of the present invention in a modular manner in an automated factory. For example, the system detailed in the drawings is primarily a system for inserting leads of radial elements, such as resistors, and axial elements, such as transistors, into printed circuit boards. In addition to the above components, certain printed circuit boards may require other components such as dual in-line integrated circuits (DIPs). In an automated factory, the system and process of the present invention can be utilized as described above to insert axial and radial elements into integrated circuit boards. In this case, it is necessary to perform chores such as aligning and controlling the individual working sections that insert the various elements, as well as ensuring that the individual assembly sections have the necessary elements to accomplish the specified task. Of course, all such alignment operations are performed by the work center controller. Regarding the work center control device, only how it functions in relation to the automatic assembly section of the present invention will be explained.

The hardware utilized in the automatic assembly section is schematically illustrated in FIG. The robot is a known robot, and the base member 3
It includes a typical robot arm 3o supported by 2. Supported on the robot arm 30 is a gripper mechanism 34 suitable for gripping assembly elements. Controlling the robot arm is a known digitally programmed robot control computer 38.

Components to be assembled in an automatic assembly section to form a printed circuit board are placed in one or more component holders or racks 38.
will be instructed. The devices whose leads are trimmed and curved and placed in a rack in a predetermined order constitute a device kit. Although the configuration of this support rack or tray is arbitrary, detailed configurations of suitable support racks are described below.

The base member 32 of the robot includes an opening 40 for locating and securing by suitable means the circuit boards into which the components are inserted.In the process flow charts and software, the circuit board support means are referred to as pallets.

A typical circuit board is partially illustrated at reference numeral 42.

A first TV camera 44 is arranged above the base member 32,
This allows the individual leads of the element to be distinguished and their positions to be detected, as will be described later. An XY table mechanism 4B is arranged below the base member 32, a second TV camera 48 for monitoring upward is placed on the XY table 46, and a lead cutter/
Place the crimper mechanism 50. A visual control computer 52 processes video information and a TV camera 44.4.
Control signals are provided to the XY table from work center controller computer 3B, which provides control signals for work center control computer 56, and the entire system is coordinated to provide the necessary information from work center control computer 56.

The modular hierarchical structure of the system software is shown in Figure 1A. Since the structure in FIG. 1A is modular, with each block representing an individual program module,
Any software module can be easily modified as long as the software and hardware interface requirements are met. In systems utilizing multiple computers, a modular structure allows software to be moved from one computer to another subject to interface requirements. Also included are interfaces, or communication software modules, that allow computers to communicate with each other and with hardware elements on a "black box" basis.

As mentioned above, one automatic assembly section of the present invention is used in an automatic factory. As already mentioned, in the environment of an automated factory and in the actually configured system of the present invention, the IB
A technical data library is provided by the M4341 computer. This library contains dimensions, part identification numbers, "dollar
Provides specific component characteristics such as "code".

More specifically, this library also includes (a) a list of standard elements for the circuit board to be assembled, (b) a file of electrical test information for each element, (C) technical details of the materials, and (d) Data from this library, which also includes circuit board perforation tape and (e) digital computer vision blueprints, is used to confirm the X-Y position of each element to be inserted and to ensure the desired location. Select an element insertion sequence, define approaches and directions for cuts and clinches, and set indicators to indicate elements that cannot be automatically inserted.

Kit information is sent to the "Materials^quisition and Robotic Kitting (MARK) module." (This system is disclosed in U.S. Patent Application No. 1,999,829.) The MARK module acquires the devices, trims and bends the leads, and arranges the devices in a standard kit. (This element kit will be described later.) By sending element shortage data to the automatic assembly section via the work control center 57, it is possible to prevent the operation of attempting to insert a missing element, and also to prevent the circuit to be assembled. Maintain statistical files of important data about the board.

The software communication package with the work control center can be logically divided into a data transmission program from the automatic assembly section to the work control center and a data reception program from the work control center.
The modules are illustrated with reference numbers 35 and 33, respectively.

Interaction between the work control center and other data collection and processing facilities is coordinated by the main sequencer program module 31. For example, the main sequencer program 31 utilizes interfaces 35 and 33 to communicate with the work center controller 57. The circuit board pallets and component kits are identified by bar codes attached thereto. At the beginning of circuit board assembly, these codes are read by a wand held in the operator's hand. The codes read by the wand are read and checked by the main sequencer program 31 to ensure that the correct circuit board and component kit are obtained. System "wake-up" and "shutdown" initiations are also performed by this program.

Coordination of the entire circuit assembly process is performed by insertion sequence module 38. This program handles all communications with the robot system, vision system, X-Y table, and lead cutter/clincher mechanism. Additionally, the program handles hole-up operations and receives error messages from the robot to allow the operator to intervene in the circuit board assembly process.

Data for the circuit board assembly process is provided by the database e-base master module 41, and data for this base is provided by the work center controller 57. Similarly, X-Y table, lead cutter/
The operating program for controlling the clincher, robot, and vision system is the X-Y table module 45.
, cutter module 43, robot controller N53, and visual controller 55.

Communication between system components also requires an interface program to accommodate data transfer between different interfaces. This transfer is performed by shared interface program module 51.

In FIG. 1A, the letter in parentheses is not allowed at the end of the module name. For example, the letter (A) is placed after "Main Sequence" 31. These characters are used in Appendix A to identify the actual software that performs the function, e.g.
In Appendix A, the software for the main sequence is designated (A).

The following sensors and signals are used to interface with the hardware. These signals are separate from those provided by the manufacturer with the various components of the system, such as robot controller 38 and visual controller 52.
A sensor field with interfaces to robots and vision systems developed by the manufacturer.
A computer (A) that interfaces with the sensor output connects the gripper-robot control device 3B.To detect the maximum movement of the fingers, a total of 4 glue switches (one for each gripper finger) (B) detects the gripping force. A total of 2 pairs of strain gauges (C) for the robot control device 3B, one pair for each pair of gripper fingers (C) 7 emergency stop switches Robot control device 38 and switch Work part control device 56 (D) stop switch Work part control Device 58 (E) Four load cells installed on the robot control device 38 on the circuit board pallet (F) Work section control device 5B limit switch installed on the cutter (G) The presence of the work section control device 56 on the circuit board pallet Presence of two limit switch (H) element kits that sense the two limit switches (1) that sense the work part control device 56. Work part control device 5B and mouth drive motor pot control device 36 operator. The interface between the machine and the automatic assembly section is composed of a work section display device 56, a robot φ system display device 63, a visual system display device 65, and a stop button display panel 67.

The assembled elements are supported by element links 38 as previously described. Device rack 38 includes a plurality of rails, each including a support device specially configured to support a particular device. Each part of the support rail is illustrated in detail in FIGS. 2-6. A suitable frame (not shown) is utilized to support the plurality of rails so that the rails are juxtaposed substantially parallel to each other. The frame also includes a suitable indexing mechanism (such as a pin) which allows the frame to be indexed with respect to the rest of the mechanical device, including the automatic assembly, along with the rails attached thereto and the elements supported by the rails. It can be positioned at a fixed location as shown in Figure 1.

Support rack 38 includes a plurality of support rails, a typical example of which is shown at reference numeral 3 in FIGS. 2.3.4 and 5, mounted on a suitable support frame. Each rail has a rectangular cross-section and includes a beveled top portion 80;
A typical row includes a plurality of transverse grooves as shown at 62 in FIGS. 2.3.4 and 5. If necessary, a thin lower portion 71 is provided to reinforce the support rail 58.

A plurality of element holders, each formed according to the type of element to be supported, are attached to each of the rail members 60. Specifically, each element holder includes an upper portion 73 as shown in FIG. 6, and two elastic arm members 60 projecting downward. An inwardly projecting ■-shaped portion 62 is formed near the bottom of the arm member 60. A second downwardly projecting V-shaped portion 64 transverse to the inwardly projecting member 62 is disposed between the resilient arm members 62 . When in use, the elastic arm member 60 is expanded outward to form the rectangular portion 6 of the rail member 58.
0, the triangular member 64 protruding downward is fitted into one of the grooves 62, and the elastic member 60
occupies a position along the lower side of the curved edge of the rectangular portion 60 to fix the element support structure 73 in a predetermined position with respect to the rail 58 . Any number of element holders 73 can be placed on any rail. Each rail includes holes 66 for securing the rail to a support structure using known fasteners, such as bolts and nuts.

The upper part of the element holder 58 may be configured to have a slot 68 on its top surface for inserting the leads of the element, as shown in FIG. The holes may be circularly shaped as indicated by reference numeral 70. In the embodiment shown in FIG. 3, each element holder may include a central portion 72 of resilient material, such as silicone rubber. In this embodiment, the leads of the element can be fixed by inserting them into the elastic material. Another embodiment of the device holder is shown in FIG. 4, in which each device holder includes two upwardly projecting outer edge portions 72, 74, each outer edge portion including a notch. These are typical resistor 76
It is useful as a holder for axial elements such as.

FIG. 7 shows a pair of typical grip fingers 78.8
0 shows the manner in which a typical element 82 is gripped. The component leads 84,86 are cut to different lengths so that they can be individually identified and individually inserted into the circuit board. Two optical fiber light sources 88.90 are provided along the outer edges of the fingers 78.80. This is used to aid in the calibration of the Roponto, as described below. The inner edges of the fingers are provided with a pair of pushers 92, 94 for pushing the components out of the fingers and inserting their leads into the circuit board. Two pairs of orthogonal fingers are used to handle "can" shaped devices such as transistors.

It is easy for a person skilled in the art to design a suitable element gripper as the case may be.

FIG. 8 illustrates the use of the optical fiber 88,80 of FIG. 7 to calibrate the position of the gripper fingers relative to the vision system. The robot arm, gripper-1 support structure and vision system can be mechanically calibrated since the position of the vision system relative to the rest of the mechanical equipment can also be calibrated as described below. Specifically, as shown in FIG. 1, the circuit board into which the device is to be inserted is removed from the opening 40 of the robot base structure 32, and then the gripper is inserted into three arbitrary positions as shown in FIG. The robot arm is positioned to sequentially position the 34 fingers 78,80. Next, by moving the XY table 46 shown in FIG. 81, the TV left camera 8 is positioned approximately below the gripper finger 34. Here, light is emitted through the optical fiber 88.80, and the image of the light from the optical fiber is T
V camera 4, and using this, an optical fiber light source 88 for the TV left camera 8 is formed. QQ
and calculate the coefficients for the position of fingers 78,80. In at least one of the calibration positions, the finger 7
8. By rotating the 80 by 180°, the gripper
Be able to calculate the rotation axis of The base 32 of the robot structure also includes three calibration holes 27, 28, 28;
Using this calibration hole, the positional relationship between the TV left camera 8 and the robot base 32 is determined.

The operation of the software and automated assembly operation process is illustrated in Figures 9-19.

For convenience of explanation - 1-1 In the figure, the software required by the various computers in which the automated assembly is fully utilized is not stored in the operating memory, but rather is stored in the various mechanical parts of the system, such as the robot arm 34, the gripper, etc.
34, X-Y table 46, etc. occupy unknown positions. The first step in activating the system is the "system wake-up" process. The operation of this process is shown in FIG.

The purpose of system wake-up is to power the entire system, position each mechanical device to a known position, and establish communications with the work center 57. As described above, when the entire automatic assembly group is powered up or down, the memories of the robot control device 3B, visual control device 52, and work unit control device 56 are erased. Therefore, the operating program (software) must be stored in a suitable memory. The operating program is supplied to the robot and visual control devices 3B, 52 from the cassette tape. The work station controller 56 is activated by files stored on the HP-/E 1000 computer (which is part of the work station control center) and downloaded to the work station controller 56.

After startup, operating software and application software appear in each system's memory.

At this point, the work station control computer application software is activated to position the automated assembly to a known position and send messages to the robot and vision system to test the communication lines. Furthermore, the X-Y table 46
5. Home" to test its mechanical and electrical operation and place the cutter 50 in the lowered open position.

More specifically, the first step in the wake-up process is to wake up the basic robot, vision system, and associated computer. This function is shown in the flowchart in Figure 9 with reference number 100.
It was shown in The example system includes circuitry that senses the power supply to each device in the system and provides signals instructing the various computers to load the necessary operating programs. That is, the operating programs for the vision system and the robot's control computer 52.36 are stored on conventional cassette tapes located near the robot system and the vision system itself. When primary power is supplied to the robot and vision systems, the robot and vision control computers are automatically enabled and the cassette φ tape is started to load the necessary programs into the associated computers. This function corresponds to reference numeral 102 in FIG.

To ensure reliable system operation, the program load operation 102 must be monitored to ensure that it completes correctly. This action 10
4 is performed by a cooperating computer. If the load operation is not performed correctly, a message is generated to indicate this on the operator console. This operation corresponds to reference numeral 106 in the figure. These messages may include simple messages instructing the operator to resume the loading operation or indicating that the computer or cassette recorder has malfunctioned and requires system repair. If the operating program is not loaded correctly without operator intervention, the operation of the working unit will eventually stop. This stage is indicated by reference number 110 in the figure.

If the load operation is successful, the next step in the system-wake-up operation is to power the basic motors that drive the robot. As a result, the motor drives the robot arm 3o, gripper 34 and associated hardware. This operation is simply an operation by the operator to turn on the corresponding power supply switch, and corresponds to reference number 110 in the figure.

Following power application to the robot and gripper drivers, the operator activates the main hardware of the robot by manipulating the robot activation switch or button to move the robot arm 30 through its range of motion. This activation is illustrated at reference numeral 112.

Operation of the robot arm 30 is checked by monitoring various limit switches that are activated when the robot arm 30 reaches its limits of travel. If this operation is performed correctly, a signal will be generated to indicate this.

If there is any disturbance in the motion of the robot arm 30 within the range of motion, a message is generated and displayed on the operator console indicating that a problem has been detected. This function is also incorporated into the message controller, as indicated by reference numeral 106, and will ultimately generate a work station stop signal if the operator does not intervene, as indicated by reference numeral 110.

In parallel with program loading and robot arm competition testing for the robot system and vision system, the work station control computer 56 is also loaded with appropriate programs from the work control center. This feature is illustrated at reference numeral 116. The loading of robot control computer 36 and vision system control computer 52 is monitored by the loaded computer and a signal is generated to indicate that the program has been loaded correctly or that the program load cycle has not been completed. This operation corresponds to reference number 118 in the figure.

Once the robot control computer 36, vision system computer 52, and workpiece controller 58 are properly loaded, a signal is generated indicating that all operations are complete. These signals activate a wake-up function for the robot and vision system as shown at 120. This wake-up function essentially consists of positioning the vision system and the robot to a known position and corresponds to reference numeral 122 in the figure. Similarly, the act of positioning lead cutter/clincher 50 and X-Y table 46 to a known position corresponds to reference numeral 124. Following these operations, all of the basic hardware included in the automated assembly is verified to be operational and positioned for precise gripping by the associated computer that controls the hardware. Once the wake-up is complete, a message indicating this is sent to the station control computer 58 as indicated by reference numeral 126 in the figure as a signal to initiate accurate calibration of the automatic assembly. vision system, robot cutter,
Alternatively, if the wake-up of the XY table 46 is not performed correctly, a signal indicating this is generated. The operator may be required to attempt a second wake-up or print out workpiece diagnostic messages that are automatically obtained from various computers.

The former is designated by reference number 128 and the latter by reference number 130, respectively. Once the message is printed, the appropriate message is formed from the contour and the work station is ultimately stopped. This stage is designated by reference number 1013, 1 in the figure.
Corresponds to 10.

Upon successful completion of all functions described above in connection with FIG. 9, the second stage of the wake-up operation is to calibrate the viewing and lighting portions of the system. The purpose of the optical calibration process is to compensate for variations in the position of the light beam within the field of view of the below-board TV camera 48 due to disassembly or adjustment of the optical system. TV camera/light source assembly 48 such that the light beam passes through the holes in the circuit board with little illumination of the underside of the circuit board.
To locate the above position, the above position must be detected. To do this, the calibration holes in the robot structure are used to determine the position of the light source and TV camera 48 relative to the robot, and the beam is directed to the reflected light located in the plane of the circuit board (which is a permanent part of the structure). Calculate the center of the illumination field for camera 48. The process is initiated by work controller 56 and the results are stored by visual control computer 52. This calibration process is shown in FIG.

The first step in the optical calibration process is to position the X-Y table 46, the light source fixed thereto, and the TV camera 48 at known positions. This position can be any position within the range of the x-y table 46 as long as it is a precise known device (initial position). This operation corresponds to reference number 132 in the figure.

After positioning the X-Y table 46 at a known position, the work section control computer 56 performs the following operations on the X-Y table 46.
A command is given to move the base member 32 to a position that is considered to be directly below any one of the calibration holes (27, 28, 28). To answer this command, any calibration hole, e.g.
Calculate the coordinates corresponding to the calibration hole 27 and create the X-Y table 4
6 to the desired position. This operation corresponds to reference numeral 134 in the figure.

Below Margin When the X-Y table 46 moves to the desired position,
The TV camera 48 starts and uses ambient light to detect the calibration hole 27.
A normal TV image is formed. Using this information, the position of the calibration hole 27 with respect to the field of view of the TV camera 48 is calculated.

This operation corresponds to reference numeral 136 in the figure. Calibration hole 2
From the initial image No. 7, the offset amount of the image with respect to the desired position is calculated. It is also possible that the first result of this calculation does not result in an image at all, in which case an erroneous message is formed, as indicated by reference numeral 138. If no available images are detected, a search pattern is activated and the X-Y table 46 is spirally moved in small increments according to the search pattern. At each location in the search pattern, station control computer 56 restarts TV camera 48 to form a new image and determines availability by having X-Y coordinates and offset H calculated.

This operation corresponds to reference numerals 140 and 142 in the figure. The spiral search pattern continues until a usable image is obtained during the course of the search pattern. Once the search pattern is complete, the operator is notified of this, as indicated by reference numeral 141. Once the initial image is detected, the offset from the center of the window is calculated, as indicated by reference numeral 144. If the statue is close enough to the center of the window that it can be used,
The X-Y table 46 is moved in a direction that reduces the amount of offset and the vision system is activated again to form a new image. This cycle is repeated until an image is obtained which shows that the calibration hole 27 corresponding to reference numeral 146 in the figure is within the window and that the amount of offset is acceptable. Once such an image is obtained, the calibration hole 27
and the offset between the calibration hole and the TV right camera are calculated and recorded. This operation is indicated by reference number 148 in the figure.
corresponds to Following this operation, the precise positioning of the X-Y table 46 and the attached TV left camera 8 relative to the structure is known. In this case, X-Y
It is convenient to move the table 46 to a known position, as indicated by reference numeral 150 in the figure, and temporarily fix it in this position.

For the system to work properly, the basic sensitivity of the vision system must be measured under known lighting conditions.

For this purpose, a reflective surface is provided on the bottom surface of the structural base plate 32, and the X-Y table 46 is positioned so that this reflective surface is within the field of view of the TV left camera 8. The process of positioning the TV left camera 8 below this reflective surface corresponds to reference numeral 152 in the figure. After setting up the camera to visually recognize the reflective surface, the TV left camera 8 is operated to form 10 images of the reflective surface, and the average offset lfr is calculated. This operation corresponds to reference numeral 154 in the figure. It is also necessary to perform a comprehensive operational check on the camera, as illustrated at 156, to detect whether any reflections are seen from the reflective surface. If no reflections are observed, appropriate false alarm messages are generated, as illustrated at 108 and 110. If the camera works and the TV picture mentioned above is acceptable, then
Basic sensitivity calibration constants are recorded at appropriate locations in memory, as indicated by reference numeral 160.

Once the optical calibration has been successfully completed as described above, the next step in activating the automated assembly is to calibrate the basic structure. The purpose of structural calibration is to correct the X-Y table 46 due to variations associated with the disassembly or homing process of the
Or to compensate for the misalignment of the pallet holder of the circuit board. The position of the reference edge of the pallet holder for the circuit board 42 with respect to the calibration holes 27.2B and 29 is constant;
It is used to locate reference holes in circuit board 42 for circuit board calibration. Further, by using this calibration, nonlinearity or nonorthogonality of the XY table 46 can be compensated for. For this purpose, three reference holes 27, 28, .
Obtain 28 images. Since the coordinates of the X-Y table 48 and the pallet geometry are known to the workpiece controller 56, the controller 56 performs all calculations for this calibration. This calibration provides a paced X-Y reference frame for all subsequent calibration processes. This structural calibration detects the position and accuracy of the provides the information necessary to accurately position the X-Y table relative to the As the first step in this process, the first calibration hole 27 is
Move the table 46. This stage corresponds to reference numeral 162 in the figure. At this position, the vision control computer 52 activates the TV left camera 8 to form an image of the first calibration hole 27 and to calculate the amount of offset of the image from the center of the window. This stage corresponds to reference numeral 164 in the figure. Once this calculation is complete, it is determined whether the image actually exists. If the image is not present, a false message is formed as shown at reference numeral 167 and a false message is formed as shown at reference numeral 18.
8, a spiral search pattern is initiated as shown at 168. At each new position of the spiral search pattern, the TV left camera 6 is activated again to image the first calibration hole 27 and calculate the offset amount. This process corresponds to reference numeral 184 in the figure. If no available image is detected upon completion of the search pattern, the operator is notified of this, as indicated by reference numeral 106.110. Conversely, once a usable image is obtained in this process, the actual offset amount is calculated, as indicated by reference numeral 170. In light of this offset amount, if it is found that the TV camera 48 is positioned in a manner that does not allow offset (ffi), the position of the X-Y table 46 is corrected to reduce this offset amount,
Form a new image. This process is repeated as indicated by reference numeral 172. This process is repeated until the XY offset amount calculation 170 determines that the offset amount is small enough to be usable. Reference number 174
The offset information is recorded in the computer's memory as shown in .

In the system of the embodiment, the robot structure has three calibration holes 27.
．． 28.28 included. In this embodiment system, after the first offset information is recorded as indicated by reference numeral 174, the positions of several calibration holes are detected, and it is confirmed whether the allowable offset amount has been calculated (or not). If all three holes have been erected and the offset amount has been calculated as described above, the workpiece controller will move the X-Y table 46 and the robot structure (base plate 3
Calculate all calibration factors between 2) and proceed to the next process. Conversely, if calibration is not performed for all calibration holes or calibration positions, X-
Move the Y-table 46 to the next calibration hole and repeat the above process. This stage corresponds to reference numeral 180 in the figure.

Through the operations described above, the X-Y table 48 is calibrated with respect to the basic robot support structure (base plate 37), and the vision system attached to the X-Y table 46, the TV camera 48, and the The light source is calibrated. Upon completion of this portion of the calibration cycle, sufficient information is available to calibrate the robot itself. Robot calibration consists of sensing the precise position of the fingers of gripper 34 that hold the components to be lead-inserted into circuit board 42 relative to other system mechanical structures. The purpose of robot calibration is to correct robot calibration relative to the base reference frame due to variations in the work breakdown or robot homing process.
The purpose is to compensate for the movement of the X-Y reference frame. To perform robot calibration, first image the optical fibers of each gripper finger in each of the two wrist orientations that form an angle of 180° with respect to each other, and calculate the position of the wrist vertical axis of rotation from this image. Then, the intersection with the base reference frame is determined as the robot's position, and then the robot is moved to two other positions to form an image. From the image and coordinates of the X-Y table 46, the workstation control computer 56 senses the position, linearity, angular orientation, and orthogonality of the robot's X and Y axes in the base frame. At this point, the base frame (X-
The robot position in the Y table is known.

This calibration also serves to compensate for changes in gripper geometry due to disassembly or damage. Since the geometry of the finger grooves and optical fibers are known to the workpiece controller 56, the gripper-finger groove spacing can be calculated from the first two images.

This value is sent to the robot controller 36 to compensate for the change by modifying the gripper's self-calibration.

During this operation, a light source (FIG. 7, optical
the light transmitted by the fiber 88.8o) and the finger 78 relative to other mechanical structures of the automatic assembly.
．． Calculate the 8° position. The first step in this process is to remove the pallet supporting one circuit board 42 from the robot, thereby removing the fingers 78, 80 of the gripper 34.
can be clearly seen within the field of view of the TV right camera 8. This stage corresponds to reference numeral 182 in the figure.

Once the pallet is installed, the work station controller 56 sends a message to the operator, indicated at 184, instructing the operator to remove the pallet for robot calibration. If the pallet has not been removed or the operator observes the system and determines that the pallet has actually been removed despite the signal to support remaining pallets, the automatic assembly section control will issue a reference number. Operator overrides are accepted from the operator as shown at 185 and 188. Both of these processes allow robot calibration to proceed.

During the calibration process, the work station controller 56, shown at 188, operates the X-Y table 46 to a known or initial position. After positioning the X-Y table 46 at a known position, the robot controller 3
6 actuates the robot to move the robot arm 30 and gripper fingers 88, 80 to known positions.

This step corresponds to reference numeral 180 in the figure. After positioning the fingers 88, 80 in the first calibration position, the work station control computer 56 operates the X-Y table 46 to position the TV right camera 8 directly below the first finger 78, as indicated by reference numeral 192. Position it in a position where it can be considered. At this position, the TV right camera 8 is placed on the finger 7.
8 is attached to the light source 90, and the amount of offset of the image is calculated. This process corresponds to reference numeral 134 in the figure. Following the calculation of the offset amount, a determination is made whether the offset is within acceptable or available limits. This operation is illustrated at reference numeral 196.

If out of tolerance or no image is visible, initiate a progressive spiral search pattern, reference numeral 198
, 200, the X-Y table 46 is moved stepwise according to the spiral search pattern, and a new offset amount is calculated at each position. If this spiral search pattern does not yield results, reference number 1
A message appears on the operator console as shown at 98.

If the occurrence of an erroneous message indicated by reference numeral 196 supports the presence of an image, an X-Y offset amount is calculated as indicated by reference numeral 202.

If the offset amount is not within the limit, X-Y table 4
6 to reduce the offset amount and repeat the calibration.

This process corresponds to reference number 203 in the figure. X-Y
After moving the table 46, a second offset amount calculation is attempted. Once it is determined that the offset amount is within the allowable range, finger calibration is performed at multiple positions. If the current position is the first calibration position as indicated by reference numeral 204, it is checked whether both fingers 78, 80 of the gripper 34 are visible. Both fingers 78. at the current position.
If 80 is not seen, the work station controller generates a signal to move the X-Y table 46 to the second finger position, as indicated by reference numeral 208, and the offset calculation process is repeated.

Conversely, if both fingers are visible, reference number 210
As shown in , gripper 34! Check to see if it has rotated 800 times. If the gripper has not rotated 180 degrees, it is rotated as indicated by reference numeral 212 and the process of imaging light source 88.8o and calculating its position is repeated. In the next cycle of this process, 1800 revolutions of the gripper 34 are commanded, as indicated by the reference numeral 210, and in the next step, the offset amount calculations are performed in all seven calibration positions, as indicated by the reference numeral 214. is confirmed. If all locations have not been reached, the process will be returned to the step indicated by reference numeral 180 and the entire process described above will be repeated. Finally, if the calculations have been made at all calibration positions, the automatic assembly control computer 56
The amount of offset of the gripper 34 is calculated at all these calibration positions as shown at 16, and the robot calibration process is completed.

Margin below As already mentioned, the automatic assembly section includes a second TV force meter 44 for checking the component leads to be inserted into the circuit board. The position of the TV camera 44 must be determined. If the position of the TV camera 44 that identifies the lead shifts due to mechanical adjustment, the camera position is corrected.

Use a proofreading process. This position calibration is also used to check the glyph bar 34 for damage. T
Calibration of the V-camera 44 is accomplished by forming images of the fingers 78,80 of the gripper 34 that are 180 degrees apart from each other. Next, the robot/robot corrects the camera position to position the robot wrist axis within the field of view of the TV camera 44. This calibration is activated by the work part controller 56, the calibration result is calculated by the robot,
be remembered.

The first step in calibrating the system is to position the robot so that the fingers 78,80 of the gripper 34 are within the field of view of the lead identification TV camera 44. This process is designated by reference number 218. Fusinger 78.80 on TV
After positioning within the field of view of camera 44, vision control computer 52 activates TV camera 44 to generate an image of gripper fingers 78,80.

Examination of this image determines whether both fingers 78,80 of gripper 34 are visible. These two processes are illustrated with reference numerals 220 and 222. If both fingers 78, 80 are not visible, the robot sends an error message to the operator control to instruct the operator to malfunction. This process is illustrated at reference number 224. If the re-finger 78.80 is seen, a check is made as to whether the finger has rotated 180°. If not, a message is sent to rotate the finger and reference number 2
A second image is formed as shown at 2B, 228. After forming the two images, an offset amount is calculated and stored, as illustrated at 230. Once the calibration cycle is complete, the autoassembly is ready to begin processing work orders. The process begins by supplying the automated assembly station with the component kits and circuit boards necessary to fulfill the work order. Additionally, information is sent from the work center computer to the automated assembly department specifying what component kits are needed. As a first step in processing a work order, an operator indicates to the automated assembly unit what parts have been supplied. Therefore, the components are supplied to the automated assembly station in the form of a pre-organized kit containing a bar code identifying the parts included in the kit. The operator reads this bar code using a hand-held bar code reader or wand to indicate to the automatic assembly section what element kit is being supplied. This operation is illustrated by reference numeral 232 in FIG. After reading the bar e-code and identifying the supplied component kit, the work station controller 56 sends a message back to the work control center indicating whether the required component kit was supplied for the circuit board being assembled. This action is referred to as 234.
, 23Ei. ” If the bipedal bar code reading indicates that additional component kits are required for the initial assembly operation, a signal is sent to the work station controller 58 and the work station controller 56 is activated in response. Generates signals to the operator.

These effects are illustrated by reference numbers 238 and 240.

Once the required component kits to be delivered are identified, a progress indicator instructs the operator to load the circuit board pallet onto the robot, as illustrated at reference numeral 242.244.

Once the pallet is loaded, the work station controller 56 sends a message to the work control center indicating the type of component kit and circuit board pallet. If a suitable circuit board is not present, a message is sent to work part control 56.
is sent and the operator is requested to attach another parent. This process is referred to as reference number 250,
252.

Once the appropriate pallet has been provided, the work station controller 56 and component kit download it as indicated by reference numeral 254. Following this download, the station controller initiates circuit board calibration as indicated by reference numeral 58.

Upon completion of the process shown in Figure 14, the system
The state is now ready for circuit board calibration as shown in Figure 5. Circuit board X-Y
Calibration is necessary to ensure that the proper hole is found during each hole centering cycle regardless of changes in circuit board position within the pallet or pallet position within the holder.

Therefore, an image of a predetermined separation hole is obtained with the TV left camera 8 located below the circuit board. One of these holes determines the origin of the circuit board reference frame in the base frame, and the other hole is used to calculate the amount of rotation between both frames. This process is performed by station control computer 56, and the results are used to calculate any subsequent board coordinates needed to control the positioning of X-Y table 46.

A circuit board height calibration process is required to compensate for circuit board or pallet thickness changes, robot Z-axis homing, or gripper 34 or workpiece disassembly that affects two directions. This process determines the position of the circuit board plane in the robot Z-axis. To this end, the robot moves over and lowers the circuit board, sensing contact between the gripper 34 and the circuit board using force sensors located within the pallet. Since the robot control computer 36 is the only device associated with movement in the Z direction, this information is generated and stored exclusively by the computer 36.

The circuit board calibration process provides all the data necessary to know where on the circuit board various components should be inserted. Functionally, the circuit board calibration process is performed by the work section control device 5B.
-Starts by commanding the Y table 46 to move to a known or initial position. This effect is illustrated at reference number 258. After moving the X-Y table 46 to a known position, the system checks to see if a pallet containing circuit boards has been placed. If the circuit board is not present, a signal is generated instructing the operator to load the pallet. These stages are designated by reference number 2. ElO, 262
Illustrated in. .

If a pallet containing a circuit board is placed, the work unit control device moves to the first position relative to the X-Y table 46, and the TV left camera 8 on the X-Y table 46 is calibrated for the first calibration of the circuit board. Command it to be positioned almost below the hole. At this position, an image of the circuit board calibration hole is formed on the TV left camera 8, and the center of this hole is calculated. These two stages are illustrated with reference number 284, 2E18. After calculating the calibration hole center, examine the result to determine whether it is appropriate. This determination is possible because the hole size and expected image are known. The error checking process is illustrated at reference numeral 268. If there is an error, it is determined whether it is the first error, and if so, the calibration process is repeated as illustrated at 270 and 272. If there is no error in the calculation, check whether the positions of all calibration holes have been detected. Reference number 27 for this stage
Illustrated in 4. If there are no errors and the calibration is not complete for all calibration holes, the calibration process is repeated for the remaining calibration holes as indicated by reference numeral 276. Once all calibration holes have been calibrated, work station controller 56 calculates the offset of the circuit board relative to the rest of the robot, as indicated at 278.

Gripper 34 is then moved 0.1 inch above the circuit board, as shown at 229. As indicated by reference numeral 231, check to see if the circuit board has been contacted, and if so, raise the gripper 34 0.002 inch and recheck to see if it has contacted the circuit board. This process is indicated by reference number 237.24 in the figure.
Corresponds to 3. If the circuit board is not in contact, record the height of the circuit board as indicated by reference numeral 247.

If not in contact with the circuit board, 0.005 until contact occurs.
Lower it in inch increments or try this three times. This stage has been designated with reference numbers 241,245, 248.

If it does not contact the circuit board after three attempts, it will send an error message and try height calibration again. This step is designated by reference numbers 251 and 253.

Once the circuit board offset calculation is complete, the system
The assembly process shown in FIG. 6 can now be started.

The circuit board is assembled using the process illustrated in FIG. The circuit assembly process consists of assembling individual elements from a prearranged element kit.
'L, Cg'e 7'! J y“′”j′8*
.

It consists of the action of incorporating into. To carry out this process, robots, vision systems, workpiece control computers,
A Y-table and cut/clinch mechanism are all utilized. This process is operator-initiated and requires downloading the physical characteristics of the parts and kit configuration exception data. The part insertion process is monitored and controlled by the workstation control computer and generally consists of the following process:

A. Element data is downloaded to the robot control computer by the work unit control.

B. The robot obtains the parts from the device kit. C3. The control computer and the TV camera below the circuit board perform the hole center measurement technique to position the device lead insertion holes.

In steps D, B, and C, the robot and TV system coordinate the lead insertion process and insert the device leads individually.

E. Cut/clinch the element lead.

Additionally, for radial and can-shaped elements, a lead identification process is performed prior to lead insertion. This allows the robot to move the x-
The y-z position can be determined.

Specifically, first, the elements to be inserted into the circuit board are obtained. As already mentioned, there may be exceptions for element kits, so
The first thing the work station controller 56 checks is whether the required part is included in the kit or included in the parts exception list. This process is illustrated at reference numeral 280.

If the required element is included in the element kit, the workstation control computer 56 sends all data regarding the element to the robot controller 36. In other words, the type, dimensions, and other necessary data of the part become clear.

Reference number 282 transmits element data to the robot controller.
Illustrated in.

Following the check of the element data, the stop switch is checked and, if ON, a list of functions selected by the operator is displayed on the work station screen 61. This process is illustrated by reference numbers 281, 283, 285. This feature list allows you to return to optical calibration, return to circuit board calibration,
Detach from the circuit board, cut off power supply, and stop processing.

Instructs to select X-Y table homing and restart processing. This selection is illustrated at reference numbers 287-289.

As circuit board assembly progresses, the work section controller moves the X-Y table 46 to a position below the hole into which the first device lead is inserted. At this position, the vision system forms an image of the hole and calculates the actual offset of the hole relative to the initial position. The two stages are illustrated with reference numerals 284 and 286.

Following the calculation of the hole offset amount, the calculation results are examined to determine whether the results are acceptable. If the calculation results are available, calculate the actual offset within the vision system window. The work unit control computer 56 that calculated the offset value is
The Y table 48 is instructed to move by the amount necessary to correct this calculated offset value, and to repeat this cycle as many times as necessary to reduce the offset amount to the required value. This function has reference numbers 288, 290, 2
Conversely, as illustrated at 92, if the error check 288 reveals the presence of an error, the system checks to see if this is the first attempt to locate a particular hole in the rotating plate. If this is the first attempt, the first work section controller 5B commands the X-Y table 46 to return to its initial position, and a second cycle is attempted. This process is designated by reference numbers 294 and 296.

If this is not the first attempt, an error message will be generated and the robot will receive 1. Indicates that this device is to be rejected and that this particular device is one that cannot be incorporated into a circuit board using the system. The robot controller then sends data to translate a statistical data file regarding the circuit board being assembled. This process is referred to as 298, 30
Indicated by 0.

Statistical information indicating that a particular device cannot be assembled is sent to the work center controller, as indicated by reference numeral 302, and then the third
It is confirmed whether there is a centering error in the hole. If it is determined that there is no third hole centering error, the process returns to the starting point and repeats as described above.

Conversely, if this turns out to be a third hole miscentering, the system will repeat the circuit board calibration if a recalibration has not been previously performed. If the circuit board calibration proceeds normally and no errors occur, the circuit board assembly process returns to the starting point again as described above. This process is illustrated by reference numerals 304, 308, 308, 309.

If the circuit board calibration has already been repeated, the station control will notify the operator of this and generate an appropriate malfunction message. This process corresponds to reference numeral 312 in the figure.

If the check 230 indicates that the X-Y offset is within the vision system window, the workpiece controller stores the offset calculation and determines whether this is the last hole corresponding to the particular part. If it is not the last hole corresponding to a particular element, the hole centering process is returned to the starting point and the centering process is performed for the next part.

If the margin below is the last hole for this element, enter the waiting time until the robot gets the next element. This step corresponds to reference numerals 314 , 318 , 318 .

The part acquisition shown in FIG. 17 can be performed in parallel with or before or after the hole centering process shown in FIG.

It is determined whether the element to be inserted is radial or axial. This information can be obtained from the element data file already supplied to the robot from the work unit control center. A process for examining this data to determine the type of device to be assembled is designated by reference numeral 320.

If it is determined that the elements to be assembled are radial,
The robot's gripper 34 is adjusted to the awn element type, as illustrated at 322. Similarly, for axially automorphic elements, the gripper 34 is adjusted as indicated by reference numeral 323. Following adjustment of the gripper 34 to the type of device, the robot moves the gripper 34 to the device kit storage location and adjusts the fingers of the gripper 34 to pick up the device to be inserted. Following the element pickup attempt, it is determined whether the element was actually acquired. This process is indicated by reference numbers 324, 3 in the figure.
Corresponds to 26.

If the acquired element is radial, the robot operates to position the element lead in front of the lead specific TV camera 44. The three angular positions used for lead identification are then measured, typically the three angular positions are 0°, 60°, and 180°. These steps correspond to reference numerals 328, 330 and 332 in the figure, followed by the measurement of the angle used to examine the lead, and the gripper
34 is positioned in the first angular position. Gripper 3
After gripper 4 is positioned in the first angular position, a counter is set to support the current angular position of gripper 34. At this position, the visual control device 52
Camera 44 is commanded to form an image of the device lead occupying this position. These stages correspond to reference numbers 334, 338, 338 in the figure.

The robot control device 3 uses the image of the lead identification vision system and the element data already supplied from the robot.
Step 6 determines whether the device has the appropriate number of leads.

If the number of leads is appropriate, the vision system calculates the coordinates of the leads from the image. When the calculation results are sent to the robot,
Based on this, the robot determines whether the amount is within a predetermined reference amount.

If it is within a predetermined reference amount, the interval between the lead tips is recorded as data in the lead identification computer. These steps are designated by reference numbers 340, 342, 344,
Corresponds to 3413.

Coordinates are then set indicating the angle at which the immediately processed image was obtained. Since this is the first image mentioned earlier, the image counter is set to one. It is now checked whether the image counter is less than or equal to 3, and if it is less than or equal to 3, the process is returned to the gripper rotation function as illustrated at 334, the gripper is rotated to a second angular position, and the process is repeated. .

When the process completes all three angular positions, the angle check function activates and verifies that the angle counter is greater than or equal to 3, in which case 1,' the system positions the lead at the intended insertion position on the circuit board. Calculate the position needed to These steps are designated by reference numbers 348, 350, .
Corresponds to 352.

Once the calculation of M is completed, all the basic lead information necessary to insert the lead into the hole is available. however,
Before proceeding with the insertion process, the core molding process described above must be completed. A standby cycle will be entered to ensure that this preparation process is completed, if necessary.
This standby cycle corresponds to reference numeral 354 in the figure.

As already mentioned, the element to be inserted may be an axial type element, and the element type check function 32
After confirming that 8 is an axial element, the robot
It is considered that the position of the element lead can be relatively accurately determined from the information specifying the position of the finger. In this case, the lead insertion process continues to the wake-up function cycle, as illustrated at 354.

When determining whether an element has been acquired by the acquired element check 32B, the gripper 34 is moved within the field of view of the lead identification camera 44 to form an image of the gripper 34, and the obtained image is examined. This confirms whether the element actually exists. These stages correspond to reference numbers 358, 358, 3H in the figure.

If the presence of the element is confirmed, the process is referred to as reference number 35.
Proceed to the Kongshin Awakening function standby cycle corresponding to step 4. vice versa,
When it is determined that the element is not gripped by the gripper 34, a determination is made as to whether the element is in the axial or radial configuration.

If the element is found to be radial, operation continues as if the element were missing. but,
If it turns out to be an axial type element, the assumption that the element was placed incorrectly in the element holder is lost, and the pickup finger is sequentially displaced in 0.0025 inch increments, three times. Attempt to pick up the element. Each time a pick-up is attempted, it is checked whether the element is gripped by the gripper 34.

Also, set a counter to count the number of pickups. These steps are designated by reference numbers 382, 38 in the figure.
4, 3B[f, 388, 370.

Once the component presence check 368 confirms the presence of the element, the cycle proceeds to a wait cycle 354. Conversely, if the element is not gripped by the gripper fingers even after three pick-ups, the robot-computer 36 instructs the work controller 56 that the part has not been correctly picked up, and controls the first work unit. The device sends the appropriate message. This function is indicated by reference number 3 in the diagram.
72, the 18th corresponding to 374! 2 is a diagram showing the process of actually inserting the leads of the element held by the fingers of the gripper into the circuit board. The first part of this process
At this step, the work station controller 56 sends updated information to the robot defining the position of the lead insertion hole. Upon receiving this information, robot controller 3B sets an indicator indicating that this lead has never made contact with the circuit board as a result of attempted insertion into the circuit board. After setting this mark, the robot controller will select the first one to be inserted into the circuit board.
Move the device so that the leads are over the corresponding holes in the circuit board. These stages are reference number 3 in the diagram. Ei, 3
78, 380.

After positioning the leads on the circuit board, a second check is made to determine if any leads have made contact with the circuit board. If contact is found to exist, the element is
．． Raise 0 inches to ensure the leads are separated from the circuit board. Once the leads are separated from the circuit board, a check is made to determine whether the element is radial or axial. These stages are designated by reference numbers 382, 384, 390 in the figure.
, 392.

If the element is determined to be a radial element, an attempt is made to insert the leads into the holes by simply lowering the element towards the circuit board.

A check is again made to see if the leads have come into contact with the circuit board as a result of this movement. These stages correspond to reference numbers 380°392, 394.388 in the figure.

If the lead does not make contact with the circuit board as a result of attempting lead insertion, the robot controller notifies the work part controller that the lead has been correctly inserted, and the X-Y table 4B attempts to insert the lead. It is commanded to move to the position below the next hole. These stages correspond to reference number 3H, 400 in the figure.

After the first lead is properly inserted, a similar process continues with the second lead.

Since the axial type read is defined as having only two trees, this cycle is performed only twice.

If the circuit board contact detection function 386 determines that the circuit board has been touched, the element is again raised above the circuit board and an indicator is placed to indicate that a second image of the lead has not been formed by the vision system. reports the set in error, X-
Y-table 48 is commanded to move to a position below the next hole in which the lead is to be inserted. After the first lead, corresponding to numerals 398 and 400 in the figure, has been properly inserted, a similar process is continued for lead 82.

Since the axial type lead is defined as having only two leads, this cycle is performed only twice.
If contact with the circuit board is found as a result of step 86, the element is again raised above the circuit board and an indicator is set to false indicating that no second image of the lead was formed by the vision system. The vision controller then activates the vision system mounted on the X-Y table 46 and forms an image of the lead by viewing the lead through the insertion hole.

After this imaging, it is determined whether the lead has been viewed by the vision system. These stages correspond to reference numbers 402, 404, 408 in the figure.

If the lead check reveals that the lead is visible, the vision controller calculates the offset of the lead relative to the hole and sends this information to the robot. If the lead is within 0.003 inches from center, the robotic controller lowers the element for insertion. This stage corresponds to reference numbers 408, 410, 412 in the figure. On the first incremental movement of the robot for lead insertion, a counter is set to N=1 and a check is made to see if the lead has touched the circuit board. If the lead does not contact the circuit board, the N value is checked to see if N=4, and if the answer is no, the lead is lowered again by 0.005 inch and the N value is increased by 1. This cycle is repeated until N=4, indicating that the insertion process is complete, or until the leads are indicated to have made contact with the circuit board. This step is indicated by reference number 414゜4 in the figure.
18, 418, 420, and 422.

If circuit board contact detection 41B indicates that the circuit board has been touched before the insertion process is complete, the element is moved or shaken slightly relative to the circuit board and circuit board contact detection is repeated again. If the detection result is negative even after shaking the element, the process returns to the insertion process and this process is repeated until the counter value reaches 4. This process corresponds to reference number '424.426 in the figure.

If the circuit board contact detection 42B reveals that the circuit board has been touched, the hit circuit board indicator is incremented by 1, and it is checked whether the value of this counter is 3 or less. If it is less than 3, raise the element by 0.0 inch and repeat the entire lead insertion process. These stages correspond to reference numbers 428, 430, 432 in the figure.

Conversely, if the one circuit board process e-counter check 430 reveals that the circuit board has been touched three or more times, a check is made to see if this is the first lead insertion attempt. If it is the first attempt, the cycle indicator is set and
A second process is attempted, the first step 38B of which is to check whether all leads have been inserted.

Conversely, if the lead was not inserted correctly on the previous attempt, the robot controller rejects the element and sends a message to the workpiece controller, which logs the message at the work control center. These stages are designated by reference numbers 434, 438, 438, 440, 44 in the figure.
2, corresponds to 448.

As a result of the lead position check 410, the lead is o from the center.
, ooa inches, the second image indicator is checked to see if it is 2nd cycle through, and if it is not, the 2nd image indicator is set to true. The process then returns to the point where the vision controller commands the camera to image the lead through the hole (404). Conversely, if the initial lead presence check 404 reveals that the lead is not visible, the second image indicator is set to false and an incremental spiral search pattern is entered. Once the search pattern has resulted in the robotic system finally positioning the element such that the lead is visible, an offset calculation process 406 is entered. Conversely, if the search pattern completes the entire spiral pattern and no leads are visible, the element rejection cycle begins at reference numeral 440.

Once the robot has inserted all device leads into the corresponding holes in the circuit board, the lead cutting program (Figure 18) begins. Once started, the cutter program commands the robot controller to finish inserting the component leads into the circuit board. These operations are indicated by reference numbers 464 and 466 in the diagram.
corresponds to When a signal is generated during the final insertion process to indicate contact with the circuit board, the device is rejected, as indicated by reference numeral 440.

At the beginning of the cutter program, the station control computer 56 is also commanded to position the X-Y table to position the lead cutter mechanism 50 below the first lead to be trimmed. Once the proper cutter position is reached,
A cutter is rotated in a desired direction to trim the first lead. This process is indicated by reference number 488 in the figure.
Corresponds to 470.

Once the cutter is in the first position, it must be determined whether this is the first lead to be trimmed. For example, if this is the first lead to be trimmed,
The cutter controller must ensure that the insertion process is complete before trimming the lead. These two operations correspond to reference numbers 472 and 474 in the figure.

If this is not the first lead to be trimmed, but the robot commands the cutter to complete the element insertion process, the cutter controller moves the X-Y table 4B to the final trimming position and trims the cutter approximately 0.1 inch. −
E raise. This positions the cutter tip at the final cut position, and the program activates the cutter to trim the lead to the desired length. These steps correspond to reference numerals 478, 478 in the figure.

When lead cutting is completed, the cutter is lowered and returned to its initial or known position. This process corresponds to reference number 481.483 in the figure. Once the cutter returns to its initial or known position, check to see if this is the final lead to be trimmed.

If it is the last lead, the cutter controller sends a message to the station computer 56 instructing the completion of the lead trimming operation.

These stages correspond to reference numeral 485 in the figure.

If the lead just trimmed is not the last lead, a signal is generated to repeat the program to position the cutter and begin a new cutting cycle. This step is illustrated by reference numerals 488, 480.

After the robot controller fully inserts the device leads into the circuit board, it sends a signal to the work station controller indicating that the device is ready for trimming. Following this signal, which allows the start of the cutting process, the robot controller sends all previously collected statistical data to the working station device for storage.

After the data is received by the central computer, the required data is sent to the master φ computer conflict file. These steps are referenced 482, 484, 486.
Illustrated in.

After the statistics file update is complete, the station controller checks whether the next element in the trimming process is a kit a failure, ie, an element that was not actually inserted.

If the next element must be skipped because it is an exception element, a signal is generated instructing the work station controller to proceed to the next element. After a check has been made regarding the next element to be inserted, a signal identifying this element is sent to the master.
sent to the computer. This process is reference number 4
88, 490, and 492.

After the next element to be inserted is determined and the data is sent to the robot controller, a standby cycle is entered until the cutting operation is performed. Once the cutting operation has been completed, a signal is sent to the robot indicating that insertion of the next element can begin. This process corresponds to reference number 488 in the figure.

When the trimming operation for the inserted element is completed, the element insertion is also completed, and this process is repeated to insert the remaining elements.

The component kits and circuit board pallets are manually fed into the automatic assembly section and positioned.

These mechanisms can be automated. Various other changes are possible within the scope of the invention.

[Brief explanation of drawings]

FIG. 1 is a functional diagram of the apparatus utilized in the present invention. FIG. 1A is a block diagram of a software system with a hierarchical modular structure. 2 to 6 are partial views of an element rack that supports individual elements. FIG. 7 is a partial view of an axially isomorphic element directed by a gripper of a robot. FIG. 8 is an explanatory diagram illustrating a technique for detecting the position of the robot-gripper-fist fingers relative to the robot. 9-19 are flowcharts illustrating the calibration control process utilized by the system. 30-Robot arm 34-Gripper- 38-Element rack 42-Circuit board 44-1st TV camera 4 [1-X-Y table 4th-2nd TV camera 50-2 sets of cutter/crimper to tk, shi

Claims (1)

[Claims] 1. A method for assembling a printed circuit board using a sensor-based robot system, comprising: a) trimming and shaping the leads of an element to be inserted into the printed circuit board in a predetermined manner; and b) the element. placing a printed circuit board into which leads are to be inserted within the working range of the robot; c) acquiring the element to be inserted using a grip attached to the robot; and d) acquiring the element by using a position sensor. e) find the relationship between each lead and the hole in the circuit board into which the lead is inserted, and e) position the lead above the hole with a predetermined accuracy by using the data regarding the relationship between the lead and the hole into which it is inserted; f) A method for assembling a printed circuit board using a sensor-based robot system, characterized in that the lead is inserted into the hole by using the robot. 2. In a method of inserting elements into a printed circuit board using a robot, a) the leads of each element to be inserted into the printed circuit board are trimmed and shaped in advance so that all the leads of each element are approximately parallel to each other; , when trimming, each of the leads has a different length; b) each of the elements is placed at a predetermined position within the working range of the robot; c) the circuit board is placed at a predetermined position within the working range of the robot. d) controlling the robot to select and acquire an arbitrary element among the elements; e) determining the positional relationship between each lead of the element and a hole in the circuit board into which it is inserted; f) positioning the longest non-inserted lead of the element at a location calculated such that the lead tip is directly above the hole into which it is to be inserted, and g) also determining the position of the lead tip relative to the hole to be inserted. h) positioning the lead tip above the hole into which it is to be inserted by sensing and repositioning the lead tip as necessary; inserting some of the leads into the holes by moving an arm of the circuit board; i) repeating steps f) to h) sequentially until all the leads of the device are inserted into the intended holes of the circuit board; A method for inserting elements into a printed circuit board using a robot, characterized by: 3. A method for setting up a printed circuit program using a sensor-based robot system, comprising: a) obtaining an element to be inserted into the printed circuit board using a gripper attached to a robot arm; sensing the position of the lead of the element; c) placing the printed circuit board in a known position relative to the robot; d) comparing the position of the lead of the element with the position of the hole into which the lead is inserted; e) and f) moving the robot arm to insert the lead into the circuit board. - Printed circuit board assembly method using the system. 4. A method of assembling a printed circuit board using a sensor-based robotic system, comprising: a) providing the robotic system with data identifying all components to be inserted into the circuit board; and b) assembling the circuit board. c) sequentially acquiring the parts, and d) including a position detector for detecting the position of each lead with respect to the insertion hole. A method of assembling a printed circuit board using a sensor-based robot system, characterized in that each lead of the device is individually inserted using a servo loop.