JPH0463687A - Holding method for object by robot - Google Patents

Abstract

PURPOSE:To immediately confirm whether or not an object is held by the tip of a robot by providing an image sensor at the tip of the robot. CONSTITUTION:An image sensor 9 to receive the image of an object (electric wire) 10 is provided at the tip having a holding section 8 holding the object. The image output of the image sensor 9 is signal-processed by an image processor 12 to extract the distinctive image regarding the object 10. A robot 1 is driven by a computer 19 to move the holding section 8 near the object 10 so that the image regarding the object 10 is located at the specified position on a screen, and the object 10 is held by the holding section of the robot 1.

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial Utilization Public Funds) The present invention relates to a method for causing a robot to grip an object.

(Prior Art) Conventionally, in order to have a robot grasp an object such as an electric wire, a television camera is used to image the object, and the image of the object is processed with signals to confirm the position of the object. The position of the robot's tip is detected from the encoder attached to each joint of the robot, and based on both the position information of the robot's tip and the position information of the object, the tip approaches the object. 2. Description of the Related Art A method of gripping an object using a gripper is known.

(Problems to be Solved by the Invention) However, when grasping objects such as electric wires, there are tasks that must be performed in locations that are difficult for workers to see, tasks that require accurate robot operation, and operations that require precise control of the robot. Another problem is that it is complicated.

The present invention has been made in view of the above-mentioned circumstances, and its purpose is to provide a robot control method that allows easy control operations and simplifies the control process compared to conventional methods. The purpose of the present invention is to provide a method for having a robot grasp an object.

(Means for Achieving the Object) A feature of the method of making a robot grasp an object according to the present invention is that an image sensor for receiving an image of the object is provided at a tip having a grip portion for grasping the object; Signal processing the image output of the image sensor to extract a characteristic image related to the object, and controlling the robot so that the image related to the object is positioned at a specified position on the screen, move the gripping section to the object. The goal is to bring it closer to.

(Function) According to the method of having a robot grasp an object according to the present invention, since the image sensor is provided at the tip of the robot, it can be immediately confirmed whether or not the tip of the robot is grasping the object. Then, the robot is controlled so that the tip of the object is located at a designated position on the screen and so that the tip approaches the object.

(Example) A method for causing a robot to grasp an object according to the present invention will be described below with reference to the drawings.

In Fig. 1, 1 is a robot, 2 is a robot base, 3 is a rotating arm, 4 is a rotating joint, 5 is a telescopic arm, 6 is a rotating joint, 7 is a tip, and 8 is a gripping part. be. An image sensor 9 is provided at the tip 7. This image sensor 9 has the function of receiving and capturing an image of electric wires 10 as a target object.
Assume that 0' is the object to be grasped. The image sensor 9 is connected to an image processing device 12 via an optical fiber 11. An image output from the image sensor 8 is input to an image processing device 12 via an optical fiber 11. The image processing device 12 plays a role of signal processing the image output of the image sensor 9.

Suppose now that an image as shown in FIG. 2 is displayed on the screen 13. In this FIG. 2, 14 is an electric wire image corresponding to the electric wire 10' as the object to be grasped;
15 is an image of an electric wire that is not an object to be grasped, and 16 is an image of an object other than the electric wire 10. Here, the reason why the wire image 14 is displayed wider than the wire image 15 is because
This means that the electric wire 10' is closer to the tip 7 than the electric wire 10'. Furthermore, the reason why the wire images 14.15 are painted black is because the wire images 14.15 are blacker than the other objects 16.

The image 16 means, for example, a distant house or other object, but is shown schematically here. In addition,
An image 15' indicated by diagonal lines represents an object having an intermediate brightness between white and black.

The image processing device 12 performs Wl image processing according to the procedure described below.

The image processing device 12 first separates the image displayed on the screen 13 into three colors, performs A/D conversion, and stores the results in an RGB memory (not shown). Then, in order to distinguish between bright and dark parts of the image 14, 15, and 16 displayed on the screen 13, a threshold level is set to identify the dark parts, and as shown in FIG. Only video 17 is extracted.

Here, a characteristic image is that the image of electricity i[10 as an object is black.

Next, focusing on the premise that the wire images 14.15 should cross the screen 13, the wire images 614.15 are removed from the screen 13. Therefore, as shown in FIG. 4, only the image 15' other than the wire image 14.15 remains on the screen 13 (see FIG. 3). ) to erase the screen on which the image 15' is shown (see FIG. 4). With this, the fifth
As shown in the figure, only the images showing the wire images 14 and 15 are left on the screen 13 in full force. The image processing device 12 compares the widths of the wire images 14 and 15. This information is then output to the computer 19. On the premise that the wide wire image 14 is the object to be grasped, the computer 19 removes the wire image 15 related to the electric wire 10# on the far side from among the wire images 14 and 15, as shown in FIG. Only the wire image 14 is extracted and displayed on the screen 13 as shown in FIG.

Next, the image processing device 12 outputs data related to the position of the wire image 14 on the screen 13 to the computer 19. The computer 19 controls the robot 1 so that the wire image 14 is located at a designated position on the screen, in this case at the center of the screen (see WA in FIG. 7). That is, the computer 19 drives the joints, arms, etc. of the robot 1. The stepping motor (not shown) is moved so that the width of the wire image 14 becomes thicker and the wire image 14 is centered on the screen 13, so that the wire image 14 on the screen 13 is horizontal and the wire image 14 control so that the widths of both left and right end portions are equal (so that the gripping portion 8 is horizontal and perpendicular to the electric wire 10'). The computer 19 repeatedly controls the joints of the robot 1 and displays the electric wire image 1 on the screen 13.
When the thickness of the wire 4 becomes constant, the gripping arm 8 is driven to grip the electric wire 10'.

In the above embodiments, color has been used as an example of the characteristic image of the object, but if the object has a characteristic shape, the shape can also be specified as the characteristic image of the object. good.

(Effects of the Invention) The method of making a robot grasp an object according to the present invention is as follows:
The configuration as described above provides the following effects.

■The structure of the robot is simplified.

■Robot autopilot processing is simplified.

■Easier operation for robot operators.

[Brief explanation of the drawing]

FIG. 1 is a schematic diagram for explaining the method of making a robot grasp an object according to the present invention, and FIGS. 2 to 7 are schematic diagrams of images projected on the screen of the image processing device shown in FIG. 1. This is the diagram shown. DESCRIPTION OF SYMBOLS 1...Robot, 7...Tip 8...Gripping part, 10...Electric wire 9...Image sensor, 13...Screen Figure 1 Figure 2 Figure 4 Figure 6

Claims (1)

[Claims] An image sensor for receiving an image of the object is provided at the tip having a gripping part for gripping the object, and the image output of the image sensor is signal-processed to extract a characteristic image regarding the object. A method for causing the robot to grasp the object by moving the gripping section close to the object while controlling the robot so that an image related to the object is located at a specified position on the screen.