JPH012891A - Vision robot 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 Summary of the Invention In a robot system with vision, when a camera is attached, the number of the link or arm to which the camera is attached is memorized, and calibration calculation is performed according to the link or arm. Calibration parameters between the camera coordinate system and a predetermined coordinate system of the robot (for example, the work coordinate system) are determined and stored in the program. This makes it possible to attach the camera to any position on the robot arm, etc.

It also facilitates the use of multiple cameras.

BACKGROUND OF THE INVENTION TECHNICAL FIELD This invention relates to a robot system equipped with a visual sensor (camera).

Conventional technology and its problems In conventional visual robot systems, the locations where cameras are installed are fixed, most of which are on the hand H, and the number of cameras is one or at most two. It is.

However, when changing hands one after another (for example, auto tool change), it is necessary to mount the camera in a place that does not interfere with hand changing.
When realizing stereoscopic vision using a plurality of cameras or when it is desired to simultaneously observe an object from different angles (for example, during assembly), it is desirable to be able to mount cameras at arbitrary positions. Conventional robots in which the camera mounting position is fixed are unable or difficult to perform the above-mentioned tasks.

SUMMARY OF THE INVENTION OBJECTS OF THE INVENTION It is an object of the present invention to provide a visual robot 0 system in which a camera can be mounted at any position and the use of a plurality of cameras is relatively easy.

Structure and effects of the invention The optical aid robot φ system according to this invention is:

means for manually marking the identification code of the member to which the camera is attached;
means for calculating calibration parameters between the coordinate system of the camera and a predetermined coordinate system of the robot by a calibration calculation program according to the member to which the camera is attached;
and the identification code of the installed camera,
The present invention is characterized by comprising means for storing the identification code of the member to which the camera is attached and the calibration parameters based on the above input results and calculation results.

The members to which the camera is attached generally include movable arms and links that constitute a robot, but also include a fixed base, a ceiling, a wall, etc. where the robot is placed.

According to this invention, after the camera is attached to an arbitrary location on a predetermined member, the identification code of the attached member is stored in the memory in correspondence with the identification code of the camera. Then, using a calibration calculation program according to the installed components.

Coordinate system calibration parameters between the coordinate system of the camera and a predetermined coordinate system (for example, a work coordinate system) of the robot are calculated, and these parameters are stored in a memory in correspondence with the identification code of the attached camera. Ru.

Therefore, no matter where the cameras are installed or how many cameras there are, when these cameras are used, the data stored in the memory mentioned above can be used to perform the prescribed processing. Predetermined robot work using a camera becomes possible. That is, according to the present invention, a camera can be attached to any arbitrary position on a robot arm, etc., and it is also easy to use two or more cameras, thereby realizing a robot system with vision that is easier to use.

DESCRIPTION OF THE EMBODIMENTS FIG. 1 shows an example of a flat working robot. The robot 1 includes a first arm 11. which is pivotally supported on a base 2 and rotates in a horizontal plane. A second arm 12 is pivotally supported by the tip of the first arm 11 and rotates in a horizontal plane. This second
A rotary direct-acting link 13 that is pivotally supported at the tip of the arm 12 and rotates once and moves in the axial direction. and a hand 14 fixed to the rotary translation link 13. A visual camera (two-dimensional visual recognition camera) can be installed at any position, and in Figure 1, the second
Cameras 7A and 7B are attached to the tip of the link 12 and to the sides of the first link 11 at approximately the center of its length. The camera can also be attached to the base 2 or the ceiling or wall of the room where the robot 1 is placed. Camera 7A, 7
The video signals representing the object obtained by B and the like are respectively sent to the image processing device 8 and subjected to predetermined processing.

The image processing device 8 is connected to a robot and a controller 9. The robot 1 is controlled and driven by the robot controller 9.

When operating a robot using a visual camera, it is necessary to determine in advance the relative positional relationship and posture relationship between the camera coordinates and the robot's work coordinates (
This is called calibration), and there are various known methods for this. Calibration is performed using one of the known methods and the data is stored in the memory of the robot controller 9. This calibration φ data is the camera coordinate system and the work coordinate system (
or other robot arm coordinate system.

For example, it is used to exchange coordinates with the coordinate system of the link 13), and is also called a coordinate system calibration parameter. These parameters include a translation machine (X-axis direction, Y-axis direction) ΔX, Δy2 rotation amount θ, and a focal length f of the camera between the camera coordinate system and the work coordinate system. During robot operation, work information (position, posture, etc.) and work information captured in the camera coordinate system are converted to information in the work coordinate system using these calibration data and used to perform two tasks. .

As shown in FIG. 3, the memory within the robot controller 9 has an area in which coordinate system calibration parameters are stored corresponding to each camera. The camera Nα (camera identification code) is used to distinguish each of the plurality of cameras. The members to which the camera is attached (arm 11, 12, link 13) correspond to this camera.
．． Penis 2. The k (identification code) of the ceiling, etc.) is also stored. For example, the ceiling etc. is ho, the arm 11 is kl, the arm 12 is NQ, 2. Identification numbers are assigned to these members, such as 3 for link 13.

Ru.

Referring to FIG. 2, first, when a camera is attached to a mounting member, the number of the mounting member is entered manually together with the camera number of the attached camera, and this is stored in the memory (step 21). . A calibration calculation program corresponding to the mounting member is prepared in advance (the calculation method is publicly known), and the program corresponding to the input mounting member is used.

A calibration between the mounted camera coordinate system and the working coordinate system is performed and the results are stored in memory (step 22.23). These processes are performed each time a camera is installed.

As described above, when a robot works, the coordinate system calibration parameters for the camera used in the work are read from the memory, and the information captured by the camera is converted into the work coordinate system using these parameters.

As described above, according to the present invention, a camera can be attached to an arbitrary position on an arbitrary member, and a plurality of cameras can be attached and used.

In the above embodiment, a planar working robot composed of rotary joints is described, but the same applies to other types of planar working robots such as a rectangular coordinate type and 9 cylindrical coordinate type, as well as three-dimensional working robots, etc. This invention can be applied.

[Brief explanation of the drawings] Figure 1 is a configuration diagram showing an example of a flat work type robot;
The figure is a flow chart showing the calibration processing procedure when a camera is attached. FIG. 3 shows an example of calibration data stored in memory. 1...Robot, 2...Base. 7A, 7B... Camera. 8... Image processing device. 9...Robot controller. 11...First arm, 12...Second arm. 13...Rotating direct link, 14...Hand. Above Figure 1 1.1

Claims (1)

[Claims] Means for inputting the identification code of the member to which the camera is attached, and a calibration calculation program according to the member to which the camera is attached, are used to calculate the distance between the coordinate system of the camera and the predetermined coordinate system of the robot. means for calculating calibration parameters;
and means for storing the identification code of the member to which the camera is attached and the calibration parameters based on the input results and calculation results, corresponding to the identification code of the attached camera. ·system.