JPS5877473A - Visual recognizing handling system - 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 (1) Technical field of the invention The present invention is directed to visual ij! This invention relates to a visual recognition handling system that handles a target object by driving four pots based on information obtained from -.

(2) Prior Art Conventionally, a system is well known in which a sample component, etc., supplied by a belt conveyor or the like is handled by operating a robot hand based on visual perception. The grasping operation in this case is suitable when the objects to be grasped are arranged at such intervals that they can be considered as a single object. Figure 1 shows an example of a handling system using such a robot. In the same figure, the robot 1 has joints 1-1.1-2.1-3.1-4.1-5 that can be rotated and positioned.
．． 1-6, and can be freely positioned within the working space by the control section 4. There are also two claws at the tip) hand 1-
It has 71.

The control unit 4 controls the hand 1- of the robot 1 based on visual perception from the TV left camera placed vertically above the target object 3.
7 position control and grasping control are performed. In this case, when the target object 3 is close to another object and is superimposed, the TV left camera can see it from vertically above. Intimidate hand 1-
7 is set to the target object, the hand position overlaps with another nearby object and cannot be grasped. Of course, it is possible to separate the target object from other objects and realign the objects, but if there are a large number of various minute parts, the time and effort required to align them becomes extremely large. Therefore, it is desirable to use the hand itself to move the obstacle away.

(3) Purpose of the Invention The purpose of the present invention is to use a robot that has a vision gage function to easily move the target object away even if there is an obstacle in the vicinity of the target object. - To provide a visual weaving hand vunging system that is fastened by -.

(4) Structure of the Invention The structure of the visual gd handling system of the present invention is to achieve the above-mentioned object.
In a visual-textile handling system that uses acupuncture needles to grasp and handle objects from above with the robot's hand, when multiple objects are to be worked on, if an object is detected that is an obstacle for the robot's hand to grasp the object as a whole, the grasp The object is characterized in that it includes means for scanning between two points on the perpendicular bisector of a line segment connecting the center of gravity of the target object and the center of gravity of the obstructing object, so that two approaching objects are moved apart.

Excellent) Examples of the invention FIG. 2 is an explanatory diagram of the configuration of an embodiment of the present invention.

The difference from the conventional robot 1 is that the controller 4 of the robot 1 is provided with a means for detecting the presence of an obstacle and scanning to move it away. Figure 3(a)
~(f) is an explanatory diagram of the operation.

In FIG. 2, a TV camera 2 inputs an image of a target object 3, and an interface unit 11 converts the video signal into a digital signal and stores it in an image memory 12.

The image memory 12 is composed of 9 pixels, for example 256×2561j, and the value of each 1ilii pixel can be referenced using the address of that pixel. The position/orientation meter side part 15 is the image memory 1
2, the position and posture necessary for the robot 1 to grasp the target object 3 are determined. For example, if the target object image is a rectangle as shown in FIG. The hand area calculation unit 14 is connected to the hand information storage unit 15.
Input the gripping width W of the robot hand and the dimension R (bXj) of the bottom of the hand shown in Fig. 3(b) from
From the position and orientation of the target object in &), the area of the hand on the image when the hand as shown in FIG. 3(c) grasps the target object from vertically above is determined. The obstacle detection unit 16 refers to the image memory 12 and checks whether there is a point on another object within the hand area. If not, it is determined that there is no obstructing object, and handling is performed based on the needle-side position and posture. If there is an obstacle, it is determined that there is an obstacle), and a rounding process is started to approach the object and move it away from the object. Figure 3 (
d) An example of the latter case is shown.

That is, as shown in FIG. 5(e), the obstacle center of gravity detection unit 17 determines the center of gravity S of the obstacle object, a portion of which is included within the hand area. The scanning data calculation unit 18 connects the center of gravity P of the object to be grasped and the center of gravity S of the obstacle object shown in the same figure (-), and sets arbitrary two points A and B on the line segment psota. B is the one closest to the center of the visual field.

The robot control unit 19 selects jll! As shown in FIG. 5(f), a control signal is sent to the robot drive unit 20], and both hands are lowered to a position where they can be grasped together and scanned, thereby performing an operation to move two objects that are close to each other apart. After performing the scanning operation, image input and recognition are performed again, and complete handling is performed this time.

(6) As described in detail, according to the present invention, even if there is an obstacle close to a target object in a visual recognition robot,
#, and the robot itself moves to move away from the target object, allowing it to grasp the target object in perfect condition. In this way, it has autonomy without using human hands or other equipment. Both provide a highly versatile visual recognition handling system.

[Brief explanation of drawings]

FIG. 1 is a general explanatory diagram of a handling system, and FIG. 2 is a configuration i1 of an embodiment of the present invention! Figure 6 (&) ~ (f
) is an explanatory diagram of the operation of the main part of the embodiment shown in FIG.
1 is logotsu), 1-7 is the hand, 2 is the TV left camera 5 is the target object, 4 is part of the system, 11 is the interface part, 1
2 is an Iij image memory, 13 is a position/center needle side part, 14 is a hand area calculation unit, 15 is an I/ND information storage unit, 16 is an obstacle detection unit, 17 is an enlarged obstacle center of gravity detection unit, and 18 is a scanning unit. A data calculation section, 19 a robot control section, and 20 a robot drive section. Patent applicant Fujitsu Ltd. Sub-agent Patent attorney 1) Yoshikazu Saka 1 Figure 2 Figure 3 (a) (b)

Claims (1)

[Claims] In a visual recognition handling system in which a target object is visually recognized from vertically above and then handled by the robot's hand from above, objects that obstruct the robot's hand from grasping the target object when working with numerical objects. is detected, the hand RtA scans between two points on the perpendicular bisector of the line segment Ofl connecting the center of gravity of the object to be grasped and the center of gravity of the obstructing object, and the two approaching objects are moved away from each other. Visual recognition handling system.