JPS6364111A - Posture determining method for robot - Google Patents

Abstract

PURPOSE:To facilitate the posture determination of a robot by storing a linking relation and a holding relation as vector information for respective parts and obtained and displaying a posture so that the vector information can be coincident at the time of determining the posture. CONSTITUTION:First, an operator generates parts information composed of linking relation information and holding relation information expressed by a vector, stores it into a memory 7, further, prepares the shape data, basic position coordinates and a basic posture vector of parts 12 and 13 and robots 10 and 11, stores shape data into a basic display data file 3, stores others into a memory 2 and inputs a necessary command to part and a robot. When the posture of a robot to the parts is determined, a CPU 1 analyzes a command inputted from a command input device 9, obtains the posture of the robots 10 and 11 so that mutual vector information of the parts 12 and 13 to be linked or the vector information given to the parts 12 and 13 and the robots 10 and 11 can be coincident and the robot shape of the posture is displayed at a three- dimensional graphic display 5.

Description

[Detailed Description of the Invention] [Summary] When teaching the robot's posture off-line on the display unit, vector information is used for each part to determine the connection relationship with other parts or the robot, and the grip relationship with the robot part. When determining the robot's posture, the robot's posture that matches the vector information is determined and displayed, thereby making it easier to determine the robot's posture.

[Industrial application field]

The present invention includes: a storage unit that stores part shape information and robot shape information; a display unit that displays the part shape and the robot shape; and an instruction device that instructs the posture of the robot with respect to the part displayed on the display unit; receiving instructions from the instruction device;
The control unit includes a control unit that displays a robot shape in a posture corresponding to the instruction on the display unit based on the robot shape information in the storage unit, and changes the robot posture with respect to the parts on the display unit. This invention relates to a method for determining the posture of a robot.

Methods for teaching robots to move can be roughly divided into online teaching methods and offline teaching methods. In the former case, it requires a skilled technique and a considerable amount of teaching time, which has the disadvantage of lowering the operating rate of the line. For this reason, the latter method, which does not reduce the operating rate of the line, is especially effective, as it displays a solid model of the same shape as the reality on a 3D display, and creates robot movement data as if it were online teaching. This method is effective.

Therefore, there is a need for a method for easily and accurately determining the posture of a robot on a three-dimensional display.

[Prior Art] FIG. 5 shows a configuration for performing offline teaching in determining the posture of a conventional robot.

In the figure, the CPU is specified by robot shape data and part shape data in a basic display data file 3, which will be described later, basic position information and basic posture vectors of the robot and parts stored in the memory 2, and a three-dimensional cursor 6. The robot posture change instruction 2 creates display data based on the position change instruction, and performs control such as changing the position information and posture vector of the robot and parts.The memory 2 stores the basic position coordinates ( An arbitrary point on the world coordinate system), a basic posture vector (information for determining the direction of the robot coordinates on the world coordinate system), and a basic position information and basic posture vector of the parts to be displayed. Display data file 3 is the robot to be displayed,
The display control unit 4 is for storing the shape data of the part in its initial state, and the display control unit 4 is for displaying the robot and part shapes on the three-dimensional graphic display 5 based on the display data created by the CPUI. Displaying the shapes of robots and parts in three-dimensional format in response to display signals sent from the display control unit 4, 3
The dimensional cursor 6 indicates the position and posture of the robot or parts to be displayed on the three-dimensional graph ink display 5, and the information converter 20 instructs the basic position information read from the memory 2 based on the instructions of the three-dimensional cursor 6. A unit that converts the basic posture vector into corresponding information, display data creation unit 2
1 creates display data of a predetermined shape based on the robot shape data and part shape data in the basic display data file 3 in accordance with the position and orientation information given from the information converter 20 under the control of the CPU 1. It is something.

In the configuration described above, the optimum posture of the robot is determined on the three-dimensional graph ink display 5 while changing the posture of the robot with respect to the parts.

In this operation, first, the user operates the three-dimensional cursor 6 and inputs posture information of the robot with respect to the displayed part with reference to the hand at the tip of the robot. Then, the basic position coordinates and basic posture vector in the memory 2 are read out, and the position coordinates and posture vector of the robot are converted based on the input posture information. Then, display data is created based on this conversion information and the robot shape data, and the robot shape in the corresponding posture is displayed on the three-dimensional graphic display 5 at the designated position.

Next, the operator determines whether the robot's posture is optimal based on the relationship between the displayed robot shape and the part shape.

If the posture of the robot is inappropriate, the three-dimensional cursor 6 is operated again to input posture information, and the above-described process is repeated.

On the other hand, if the posture of the robot is appropriate, the joint angles of each joint (in the case of an articulated robot) corresponding to the posture are stored.

[Problems to be Solved by the Invention] In the conventional robot posture determination method using a three-dimensional cursor described above, the final lif confirmation of the posture determination was determined visually on the three-dimensional graph ink display 5, so that strict There was a problem in that the robot's posture could not be determined and it took time to make the final posture determination. 1. For example, when a robot's work involves assembling parts, it is difficult to express sufficient perspective on the three-dimensional graphic display 5, even though precise positioning accuracy is required, and visual confirmation is required. During work, accurate position and posture cannot be determined.

The three-dimensional cursor 6 can be moved in the X and Y directions by tilting the stick, and can be moved in the Z direction by rotating the stick.
The rotation angle of the robot's arm, which has 6 degrees of freedom, can be determined by setting another mode and instructing the desired rotation angle of the arm for the rotation of the stick. There was a problem in that it took time to give instructions.

On the other hand, as a method of instructing the robot's posture, the hand position and posture at the tip of the robot are stored as vector information.
There is also a method of specifying using robot language (VAL), but each time the position of a part changes, it is necessary to specify the hand position and posture in robot language, which is extremely troublesome and time-consuming. was there.

SUMMARY OF THE INVENTION In view of the above-mentioned conventional problems, it is an object of the present invention to provide a method for determining a robot's posture with respect to a component by a simple operation and in a short period of time.

[Means for solving problems]

The purpose of this is to provide a storage section that stores part shape information and robot shape information, a display section that displays the part shape and robot shape, and an instruction device that instructs the posture of the robot with respect to the parts displayed on the display section. and a control unit configured to receive an instruction from the instruction device and display a robot shape in a posture corresponding to the instruction on the display unit based on the robot shape information in the storage unit, and display a robot shape on the display unit. A robot posture determination method in which the robot posture is determined by changing the robot's posture relative to A storage means is provided for storing vector information indicating the joining direction, and when determining the posture of the robot with respect to the parts, the vector information of the parts to be joined or the vector information given to the parts to be joined and the robot match. This is achieved by a method for determining a robot posture, which is characterized in that the posture of the robot is determined, and the shape of the robot having the determined posture is displayed on the display section.

〔Example〕

FIG. 1 is a diagram showing the configuration of a system for determining the posture of a robot according to the present invention.

In the figure, parts that are the same as those in FIG. The command input device 9 is used to determine the gripping relationship in advance and store it as vector information for each robot hand and each part.The command input device 9 is used to input commands, and is used to input combination commands and combinations of parts to be combined or parts and robot hands. The command analysis unit 8, which inputs the combination, determines the division method to be read from the memory 7 based on the combination command and combination information input from the command input device 9. Although not shown, a three-dimensional cursor 6 is also provided in this system configuration, and indicates the posture of the robot.

In the configuration described above, the operation will be explained in the case of assembly work by the two-arm cooperative robot shown in FIG. In FIG. 3, 10 is a first robot arm, 11 is a second robot arm, and 12 and 13 are parts, respectively.

First, an operator creates component information as shown in FIG. 2 and stores it in the memory 7. This component information consists of connection relationship information and grip relationship information, and each piece of information is expressed by a vector. In the case of connection relationship information, a vector for connection is defined for each part, and before the parts are combined, the two vectors point in opposite directions and the two parts are arranged so that their end points coincide, and after the parts are combined, The two parts are arranged so that the two vectors point in opposite directions and their starting points coincide. In addition, in the case of a gripping relationship, the same vectors as for the parts are defined for the robot hand, and the arrangement of the part and the robot hand before and after gripping is the same as in the case of a combination relationship.

In addition to inputting such part information, the operator creates shape data, basic position coordinates, and basic posture vectors for each part 12.13 and robot 10.11, and stores the shape data in the basic display data file 3. However, the basic position coordinates and basic orientation vector are stored in the memory 2.

After creating such data regarding the robot and parts, the postures of the robot 111, parts 132, parts 12 and the robot 10 are determined in this order (the order of posture determination will be explained with reference to FIG. 4).

First, the operator uses the three-dimensional cursor 6 to move the robot 1
1 is determined as shown in Fig. 4(a).

Next, the part 13 is input to the robot from the command input device 9) 11
Enter the positioning command. Additionally, it is input that the combination to be combined is the part 13 and the robot 11.

The CPU 1 determines the combination to be combined via the command analysis unit 8, and reads the gripping relationship information described by the vectors of the part 13 and the robot 11 from the memory 7.

Here, since the robot 11 is positioned in the world coordinate system (a coordinate system arbitrarily determined on a three-dimensional graphic display), the CPU (vector indicating the rotational direction of the hand) and orientation vector (vector indicating the posture in the direction of rotation of the hand).

Then, the approach vector (vector indicating the engagement direction) and orientation vector (vector indicating the attitude in the rotation direction) of the component 13 in the world coordinate system are determined so as to satisfy the following conditions.

Condition 1: The approach vector of part 13 is robot 11
Condition 2: The starting point of the approach vector of the part 13 and the starting point of the approach vector of the robot 11 match. Condition 3: The orientation vector of the part 13 and the orientation vector of the robot 11 match. Then, the basic position coordinate and basic orientation vector information stored in the memory 2 are converted from the approach vector and orientation vector in the world coordinate system obtained.

Then, display data for the component 13 is created based on this conversion amount and displayed on the three-dimensional graphic display 5.

As a result, the part 13 is displayed on the hand of the robot 11 in a positioned state as shown in FIG. 4(b).

Next, the user operates the command input device 9 to input a combination command, and also inputs information indicating that the combination of parts 12 and 13 is to be combined.

By repeating the above-described operations from now on, a state in which the part 12 is positioned on the part 13 is displayed as shown in FIG. 4(C), and the robot 10 is
The state in which the is positioned on the component 12 is displayed as shown in FIG. 4 (dl).

After positioning the robots 10 and 11 in this manner, the angles of each joint of the robots corresponding to the displayed postures are determined.

Note that the command input device 9 may be configured with a keyboard device, or input combination commands in the form of a menu on the display 5, and the combination information is written to the parts, robot hands, etc. displayed on the display 5. The input may be made by giving instructions with a pen.

〔Effect of the invention〕

As explained above, in the present invention, the coupling relationships 9 and gripping relationships between components and between components and hands are stored as vector information, and the positions of components are adjusted so that the vector information matches.
Since the position of the robot is determined, accurate robot coordinates can be taught even in combination work, etc., and can be taught in a short time.

[Brief explanation of the drawing]

Fig. 1 is a system configuration diagram for determining the posture of a robot according to the present invention, Fig. 2 is a diagram showing information on the connection relationship between the parts to be assembled and information on the gripping relationship between the robot hand and the parts, and Fig. 3 is a diagram showing the system configuration for determining the posture of the robot according to the present invention. A diagram explaining assembly work by a two-arm cooperative robot, Figure 4 is an explanatory diagram of the order of determining the robot's posture,
FIG. 5 is a diagram showing a configuration for performing offline teaching in conventional robot posture determination. In the figure, 1 is a CPU, 2.7 is a memory, 3 is a basic display data file, 4 is a display control unit, 5 is a three-dimensional graphic display, 6 is a three-dimensional cursor, 8 is a command analysis unit, and 9 is a command input device. It is. 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 300 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 100 x 1,000, 2. Name of the invention 1. t'T) (P-su-? 3 Relationship with the case of the person making the amendment Patent applicant address 10158 Kami Elementary School III Junior High School, Nakahara-ku, Yozaki City, Kanagawa Prefecture (522) Name: Fujitsu Ltd. 4, Agent Address: Kanagawa Prefecture Kamiodanaka + OtS address, Nakahara-ku, Kenyosaki City #'l'4f1.+-i17

Claims (1)

[Claims] a storage unit that stores part shape information and robot shape information;
a display unit that displays the part shape and the robot shape; an instruction device that instructs the posture of the robot with respect to the part displayed on the display unit; The robot posture determining method comprises: a controller for displaying a robot shape in a posture corresponding to the instruction on the display section, and determining the robot posture by changing the posture of the robot with respect to the parts on the display section. A storage means is provided for storing vector information assigned to each part indicating the direction of connection with other parts or the robot, and vector information indicating the direction of connection of the robot with the part. When making a decision, find the posture of the robot such that the mutual vector information of the parts to be joined or the vector information given to the parts to be joined and the robot match, and display the robot shape of the found pose on the display. A method for determining the posture of a robot, characterized in that the posture of the robot is displayed.