JPS6348658B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a method for improving the production efficiency of parts assembly in a robot equipped with an image processing device.

[Conventional technology]

Conventionally, a robot equipped with an image processing device inputs an image of a part to be grasped to the image processing device, analyzes the image, calculates the position coordinates (XY coordinates and inclination) of the part to be grasped, and then calculates the position coordinates. In this method, the robot grabs the parts and assembles them according to the notified position coordinates.

[Problem that the invention seeks to solve]

However, the conventional method has the disadvantage that if it takes time for the image processing device to analyze the image of the part to be grasped and calculate the position coordinates, that time is directly added to the part assembly time, reducing production efficiency. .

Therefore, in order to solve these conventional drawbacks, the present invention aims to improve the speed of assembling parts without affecting the part assembly time even if it takes some time for the image processing device to analyze the image and calculate the position coordinates of the part to be grasped. The purpose is to not reduce production efficiency.

[Means to solve the problem]

In order to solve the above problems, this invention
The robot commands to the image processing device are divided into two types: processing start commands and processing result notification commands. While the robot is assembling parts, the image processing device performs image analysis and positioning of the next part to be grasped. It is now possible to perform coordinate calculations.

[Effect]

As described above, by dividing the commands into two types, the robot that has grabbed a part sends a processing start command for the next part to be grabbed to the image processing device, inputs an image of the next part to be grabbed, and detects the presence or absence of the part. The image processing device notifies you of only simple results, assembles the parts you are grasping, and then
Even after notifying the robot of a simple result, the image processing device continues to analyze the image of the part to be grabbed next and calculate the position coordinates, so when the robot has finished assembling the part, the image processing device will automatically determine which part to grab next. Since the calculation of the positional coordinates of the part has been completed, when the robot sends a processing result notification command to the image processing apparatus, the image processing apparatus can immediately notify the robot of the positional coordinates of the part to be grabbed.

〔Example〕

An embodiment in which the present invention is applied to the assembly of the hour hand of a watch will be described below with reference to the drawings.

In FIG. 1, a robot 1 has an hour hand chuck 2 at the tip of a movable part, and a camera 3 for an image processing device.
is installed. A pallet 6 on which a watch 5, which has been assembled up to the dial plate, is mounted is transported onto the transport line 4 and positioned at an hour hand assembly position 7. The tray 9 on which the hour hands 8 are arranged at regular intervals is stored in an automatic tray feeding device 10.
Accordingly, the tray is positioned at the tray fixing position 11.

At this time, the robot 1 checks the hour hand 8a using the hour hand chuck 2, and then moves the hour hand 8a to the tray 9.
The robot 1 moves and sends a processing start command to the image processing device (Me) 12 so that the image of the hour hand 8b riding on the robot can be taken into the image memory in the image processing device (Me) 12 through the camera 3. When the image processing device 12 receives this processing start command, the hour hand 8
Load the image b into the image memory. This time is
Since it is a normal TV image, it takes about 0.03 seconds. Next, the image processing device 12 uses the image memory to detect the presence or absence of the hour hand 8b. This time is about 0.1 seconds because it is a simple process of detecting the presence or absence. Next, the image processing device 12 sends the result of the processing start command to the robot, and subsequently uses the image memory to perform image analysis and position coordinate calculation of the hour hand 8b. The robot 1 then uses the image processing device 1.
2, when the result of the processing start command is notified, the hour hand 8a that is chucked in the hour hand chuck 2 is assembled into the watch 5 on the pallet 6. The assembly time is about 2 seconds, and the image processing device 1
2, the time required for image analysis and position coordinate calculation is about 1 second, so the robot 1 can move the hour hand 8a
After assembling, when a processing result notification command is sent to the image processing device 12, the image processing device 12 can immediately notify the robot 1 of the position coordinates of the hour hand 8b. Therefore, robot 1 immediately
You can go grab the hour hand 8b.

However, it is of course necessary to wait for the image processing time only when checking the hour hand for the first time.

The above operating procedure is shown in FIG.

That is, when you check one on the checkmark, a step is entered to capture the next hour hand image while moving to the assembly position, but this time is extremely short (0.03 seconds) as mentioned above, so it can be ignored. can.

FIG. 3 shows a schematic block diagram of the present apparatus, in which the robot 1, chuck 2, camera 3, and image processing device 12 correspond to those in FIG. 1, respectively. The image processing device 12 includes a camera interface 12
0, image memory 121, robot interface 122, image processing program 124 and
It is composed of a CPU 123. When the image processing program 124 first detects that there is no image in the image memory 121, it gives a no-image signal to the robot 1 via the CPU 123 and the robot interface program 122. Then, the robot 1 outputs an image capture command to the image processing device. This is the start of the flowchart in Figure 2. The robot 1 is driven by a signal from the robot interface program 124, and the camera 3 obtains an image signal of the hour hand and inputs it to the image memory 121 via the camera interface 120. If an image of the hour hand cannot be obtained at this time,
That is, when the tray 9 becomes empty, there is a step of transporting the next tray, although this is omitted in the flowchart.

The data in the image memory 121 is stored in the CPU 123.
It is analyzed by the image processing program 124 under the control of the image processing program 124 and converted into chuck coordinate position data.
This coordinate position data is stored in the robot interface program 122. Next, robot 1
A signal according to the position data is output from the robot interface program 122 in response to a command from the robot 1, and the robot 1 checks the target hour hand.
Since the positions of the camera 3 and the chuck 2 do not match, it goes without saying that the position of the robot 1 is different when capturing an image and when chucking.

In addition to the embodiments applied to the assembly of the hour hand of a watch described above, there are also embodiments to which the present invention is applied to the assembly of the dial, minute hand, second hand, etc. of a watch.
side, back cover, and watch coil block,
And so on.

〔Effect of the invention〕

As explained above, this invention divides the command from the robot into two parts, and while the robot is assembling the parts, the image processing device can calculate the positional coordinates of the part to be grabbed next. In this way, the production efficiency of parts assembly is effectively improved.

[Brief explanation of drawings]

FIG. 1 is a configuration diagram showing an embodiment of the present invention applied to the assembly of the hour hand of a watch;
The figure is a flowchart showing an embodiment of the present invention, and FIG. 3 is a block diagram of an image processing apparatus. 1...Robot, 2...Hour hand chuck, 3...
...camera for image processing device, 4...conveyance line, 5
...Clock that has been assembled up to the dial, 6... Palette, 7... Clock assembly position, 8, 8a and 8b
... Hour hand, 9 ... Tray, 10 ... Tray automatic supply device, 11 ... Tray fixing position, 12 ... Image processing device (Me).

Claims (1)

[Scope of Claims] 1. In a parts assembly robot having an image processing function, (a) a procedure for grasping a part in response to position coordinates notified from the image processing function; and (b) a part to be grasped next. (c) assembling the grabbed parts and calculating the positional coordinates of the next part to be grabbed; (d) applying the calculated positional coordinates to the image processing function; Based on the above, the robot pre-reading method is characterized by comprising the step of immediately grabbing the next part to be grabbed after the assembly of the parts is completed.