JP2014128840A - Robot control system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a robot control system capable of controlling a three-dimensional visual sensor and a robot while connecting the three-dimensional visual sensor to the robot despite difference in a signal form and an information form between the three-dimensional visual sensor and the robot.SOLUTION: A robot control system includes: a three-dimensional visual sensor 2 for detecting three-dimensional position information and attitude information on a workpiece; a visual sensor control unit 4 for controlling the three-dimensional visual sensor and performing an information processing; a robot 7 for transporting the workpiece; a robot control unit 6 for controlling the robot; and an information mediating unit 5 for converting an information form output from the visual sensor control unit to a form that the robot control unit can handle and transmitting the converted form to the robot control unit, and converting an information form output from the robot control unit to a form that the visual sensor control unit can handle and transmitting the converted form to the visual sensor control unit.

Description

  The present invention relates to a robot control system for detecting a three-dimensional information representing a position and posture of a workpiece by a three-dimensional visual sensor and controlling the robot based on the information to convey the workpiece. The present invention relates to a robot control system that enables control by connecting these sensors regardless of the difference in information format between the sensor and the robot.

  As an automated work process such as processing, assembly, inspection, etc., work items such as unprocessed products, parts, intermediate products, and final products are stored in a tray and placed near the work position. The work on the tray is taken out and transported to the work position. Work efficiency has been improved by automatically transferring and supplying workpieces by such robots. However, at present, there are many systems that require the workpieces to be arranged and arranged on the tray, and the automation is not sufficiently efficient for the work of arranging and arranging the workpieces manually.

  Even if the workpieces are not aligned on the tray, in order to automatically transport and supply workpieces, the workpiece position information and posture information are accurately acquired, and the robot is controlled based on these pieces of information. There is a need. An example of a robot control system that obtains workpiece position information from image information captured by a camera and controls the robot based on the information is described in Patent Document 1 below. Patent Document 1 describes a robot control system that obtains workpiece position information and posture information from workpiece image information captured by a stereo camera, and controls the operation of the robot based on the position information and posture information.

JP 2009-248214 A

  Since the robot control system as in Patent Document 1 obtains the position information and posture information of the workpiece by the three-dimensional visual sensor and controls the operation of the robot by using the position information and posture information, it is necessary to arrange the workpieces in advance. There is an advantage that efficient automation work becomes possible. However, since such a robot control system needs to develop an entire system by comprehensively combining a 3D visual sensor and a robot, labor and cost for constructing the system increase, There was a problem that the cost also increased.

  In addition, there are various types of three-dimensional visual sensors and robots depending on manufacturers, and the control forms and signal input / output forms are also different. In addition, there are variations in various specifications and forms even at the same manufacturer. Although it is desirable to construct a robot control system by selecting the most suitable one among those models and variations, it is practically extremely difficult to construct such a robot control system. This is because a manufacturer that develops a robot control system preferentially uses its own system or fixes it to a model that is easy to develop the system.

  Even if the user selects the most suitable 3D visual sensor and robot and connects them to construct a robot control system, the measurement signal, control signal line, signal format, information format, etc. are the manufacturer. Because it differs depending on the model, the robot could not be connected to a 3D visual sensor of any model.

  As described above, in the conventional robot control system, the cost of the entire system increases, and it is difficult to construct a system by selecting an optimal model for each of the three-dimensional visual sensor and the robot. Furthermore, after constructing the robot control system, it has been extremely difficult to change the model of the three-dimensional visual sensor or the robot.

  Accordingly, the present invention is a robot control system for detecting a three-dimensional information representing the position and posture of a workpiece with a three-dimensional visual sensor, controlling the robot based on the information, and transporting the workpiece. It is an object of the present invention to provide a robot control system that enables control by connecting these visual sensors and robots regardless of the signal formats and information formats.

  In order to achieve the above object, a robot control system according to the present invention includes a three-dimensional visual sensor for detecting three-dimensional position information and posture information of a workpiece, and a control and information processing for the three-dimensional visual sensor. Visual sensor control unit, robot for transporting the workpiece, robot control unit for controlling the robot, and format in which the robot control unit can handle the information format output from the visual sensor control unit An information mediating unit that converts the information format output from the robot control unit into a format that can be handled by the visual sensor control unit and sends the information format to the visual sensor control unit. It is what has.

  In the robot control system, the information mediation unit preferably uses a control device for sequence control.

  In the robot control system described above, the information mediating unit may convert binary data representing numerical values and text data to each other.

  In the robot control system described above, the information mediation unit may handle binary data representing a numerical value as integer data.

  In the robot control system, the robot may be an articulated robot.

  Since this invention is comprised as mentioned above, there exist the following effects.

  In the robot control system of the present invention, the three-dimensional visual sensor and the robot can be made the most suitable models according to the work content and work environment. If a low-cost model is selected as the three-dimensional visual sensor and the robot, the overall cost of the robot control system can be reduced. Furthermore, even after constructing the robot control system, it is possible to change the model of the three-dimensional visual sensor or the robot.

FIG. 1 is a diagram showing an overall configuration of a robot control system 1 of the present invention. FIG. 2 is a perspective view showing a state of the workpiece 9 in the tray 8. FIG. 3 is a flowchart showing the operation of the information mediating unit 5 when performing calibration. FIG. 4 is a flowchart showing the operation of the information mediating unit 5 when performing automatic operation for picking up a workpiece.

  Embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing an overall configuration of a robot control system 1 of the present invention. The articulated robot 7 is installed in the vicinity of the supply position of the tray 8, holds the work 9 (see FIG. 2) in the tray 8 by the hand 71, and removes the work 9 from the tray 8 to a predetermined work position. Transport to. Here, the work position is a position for performing work such as processing, assembly, and inspection. Moreover, a workpiece | work refers to the target object of these operations.

  The tray 8 on which the workpiece 9 is placed is set at a supply position as shown by a tray changer (not shown) provided separately. When unprocessed workpieces 9 are successively taken out by the robot 7 and there are no unprocessed workpieces 9 in the tray 8, the processed tray 8 is replaced with a new tray 8 on which the unprocessed workpiece 9 is placed by the tray changer. Exchanged.

  In the robot control system 1 of the present invention, it is not necessary to align the workpieces 9 on the tray 8. The workpiece 9 is placed on the tray 8 in an unaligned state. FIG. 2 is a perspective view showing a state of the workpiece 9 in the tray 8. The work 9 is placed on the tray 8 in an unaligned state, and the position and the posture are in a disjoint state. In order to transport the workpiece 9 in such a state by the robot 7, it is necessary to accurately acquire position information and posture information of each workpiece.

  The three-dimensional visual sensor 2 is an image sensor for acquiring the position and orientation of each workpiece 9 as three-dimensional information. The three-dimensional visual sensor 2 includes two cameras 21 and 22 arranged at a predetermined distance, and can determine the position and posture of each workpiece 9 as three-dimensional information. The two cameras 21 and 22 are oriented in the direction in which the center of the visual field is substantially the center of the tray 8 so that the entire area of the tray 8 can be acquired as an image.

  In the three-dimensional visual sensor 2, the timing of image acquisition and information transmission are controlled by the visual sensor control unit 4. The visual sensor control unit 4 and the three-dimensional visual sensor 2 are connected by a normal LAN cable. A visual sensor power supply 3 is disposed at a midway position of the LAN cable connecting the visual sensor control unit 4 and the three-dimensional visual sensor 2. The visual sensor power supply 3 is for supplying power to the three-dimensional visual sensor 2 via a LAN cable. Since power is supplied to the 3D visual sensor 2 via a LAN cable, it is not necessary to provide the 3D visual sensor 2 with another power supply device.

  The visual sensor control unit 4 controls the three-dimensional visual sensor 2 and processes image information sent from the three-dimensional visual sensor 2 to obtain the position and orientation of each workpiece 9 as three-dimensional information. . The obtained position and orientation information is output to the outside in response to an external command. Generally, the three-dimensional visual sensor 2 and the visual sensor control unit 4 are sold as an integrated product.

  On the other hand, the operation of the robot 7 is controlled by the robot controller 6. The robot control unit 6 can operate the robot 7 by executing an arbitrary program in response to an external command, and can also directly operate the robot 7 in accordance with an external command. Furthermore, it is possible to output information in the robot control unit 6 to the outside. Generally, the robot 7 and the robot controller 6 are sold as an integrated product.

  As described above, if the system is developed as a robot control system as in Patent Document 1, the visual sensor control unit of the three-dimensional visual sensor and the robot control unit are directly connected to operate as a robot control system. It is possible. In addition, some products of the same manufacturer can link the visual sensor control unit and the robot control unit directly to perform a linking operation. However, in such a robot control system, the types of the three-dimensional visual sensor and the visual sensor control unit and the robot and the robot control unit are fixed, and it is possible to make these optimal models according to the work contents and the work environment. Can not.

  Therefore, in the present invention, the information mediation unit 5 is interposed between the visual sensor control unit 4 and the robot control unit 6, regardless of the difference in signal format or information format between the visual sensor control unit 4 and the robot control unit 6. These can be connected to perform robot control. The visual sensor control unit 4 and the information mediation unit 5 are connected by a normal LAN cable. Further, the information mediation unit 5 and the robot control unit 6 are connected to a general-purpose communication cable by a communication method (for example, a communication method for connecting industrial equipment / control device).

  Thus, it is not necessary to make the communication system of the visual sensor side and the robot side different from the information mediation unit 5, and both communication systems may be the same. Such a communication method depends on a communication method that can be used by the visual sensor control unit 4 and a communication method that can be used by the robot control unit 6. If the information mediation unit 5 can use various communication methods, it is possible to connect various types of visual sensors and robots.

  The information mediation unit 5 not only physically connects the visual sensor control unit 4 and the robot control unit 6, but also converts the information format output from one into a format that can be handled by the other, and the position information and control information Are mutually available. The information mediating unit 5 sends commands (commands) such as information acquisition, information setting, and control necessary for the robot control system to the visual sensor control unit 4 and the robot control unit 6. As the information mediating unit 5, a control device for sequence control (generally, a device called a sequencer, PLC: programmable logic controller, etc.) can be used.

  Next, a specific operation of the information mediating unit 5 will be described. Here, as an example, the visual sensor control unit 4 employs an information format expressed by text data, that is, a character code such as JIS code or ASCII code, and the robot control unit 6 directly inputs binary data, that is, binary code. A case where an information format used as such is adopted will be described.

  FIG. 3 is a flowchart illustrating the operation of the information mediating unit 5 when calibration is performed by the robot control system 1. Calibration means that the coordinate information (position information and angle information) on the operation of the robot 7 and the robot control unit 6 matches the coordinate information of the workpiece that the 3D visual sensor 2 and the visual sensor control unit 4 require by 3D image recognition. It is a procedure to make. Specifically, the work is placed at the calibration position, the hand 71 of the robot 7 is moved to the calibration position, and the position is taught to the visual sensor control unit 4. The visual sensor control unit 4 matches the coordinate information of the workpiece obtained by the three-dimensional image recognition with the coordinate information from the robot control unit 6.

  When calibration is started, the information mediation unit 5 first sends a command for starting a calibration program to the robot control unit 6 in step 101. The robot control unit 6 returns confirmation information that the program has started to the information mediating unit 5. Next, in step 102, a command to start continuous imaging is sent to the visual sensor control unit 4. During calibration, continuous imaging is performed by the three-dimensional visual sensor 2.

  Next, in step 103, the information mediation unit 5 sends a movement command to the robot control unit 6. The hand 71 of the robot 7 is moved to a position near the calibration position. On the robot 7 side, the position is adjusted so that the hand 71 can grip the workpiece at the calibration position. Thereafter, the robot control unit 6 stores the current coordinate information in a register, and further sends movement completion information to the information mediating unit 5.

  The coordinate information stored by the robot controller 6 is as follows. The position information is a position on each coordinate axis of orthogonal coordinates (X, Y, Z axes) with the robot origin position as the origin, and the data is a signed integer value in units of 0.1 mm. The angle information represents angles around the X, Y, and Z axes (hereinafter, these angles are referred to as RX, RY, and RZ), and the data is a signed integer value in units of 1/100 degrees. For example, if the X-axis coordinate data is 5012, the X-axis coordinate value represents a position in the positive direction 501.2 mm. If the RX data is 17000, the angle around the X axis represents an angular position of 170.00 degrees in the positive direction.

  Next, in step 104, the information mediation unit 5 confirms the completion of movement of the robot 7 and acquires coordinate information from the robot control unit 6. Note that the coordinate information (position information and angle information) of the hand 71 corresponds to the position information and posture information of the workpiece at the calibration position. The coordinate information (X, Y, Z, RX, RY, RZ) is an integer value as described above. Since coordinate information by such an integer value is binary data, it cannot be handled by the visual sensor control unit 4 as it is.

  Next, in the procedure 105, the information mediating unit 5 converts the information format of the coordinate information and creates a teaching command for the visual sensor control unit 4. That is, coordinate information composed of integer values as described above is converted into real values, and further, a teaching command (text data) including these real values as text data is created.

  In step 106, the information mediation unit 5 sends the teaching command created in step 105 to the visual sensor control unit 4. The visual sensor control unit 4 adjusts internal coefficients and constants so that the coordinate information of the workpiece obtained by the three-dimensional image recognition matches the coordinate information from the robot control unit 6.

  Thus, the calibration for one workpiece is completed. When performing calibration for a plurality of workpieces, the above steps 103 to 106 are repeated by the number of workpieces. When calibration for all the workpieces is completed, the process proceeds to the next procedure 107. In order to improve calibration accuracy, it is preferable to perform calibration on a plurality of workpieces.

  In the next procedure 107, the information mediation unit 5 sends a continuous imaging stop command to the visual sensor control unit 4. In step 108, a command to end the calibration program is sent to the robot controller 6. And the operation | movement at the time of calibration of the information mediation part 5 is complete | finished.

  When the calibration as described above is completed, the robot control system 1 can perform automatic operation for picking up the workpiece. FIG. 4 is a flowchart showing the operation of the information mediation unit 5 when the robot control system 1 is automatically operated. When automatic driving is started, the information mediating unit 5 first sends a command for starting a program for automatic driving to the robot control unit 6 in step 201. The robot control unit 6 returns confirmation information that the program has started to the information mediating unit 5.

  Next, in step 202, the information mediation unit 5 sends information for confirming that the tray is normally set to the robot control unit 6. When the robot controller 6 receives the information on the tray confirmation, if the robot 7 is in a normal ready state, the robot controller 6 sends the start instruction information to the information mediator 5 so as to start the three-dimensional image recognition of the work on the tray. send.

  Next, in step 203, the information mediating unit 5 sends a command for setting the window number to the visual sensor control unit 4. The visual sensor control unit 4 divides the image of the entire visual field of the three-dimensional visual sensor 2 into several regions or windows including the image of the workpiece. Each window is assigned one consecutive positive integer number (window number).

  The visual sensor control unit 4 performs a 3D image recognition process on the window having the window number set by the information mediating unit 5. The window number is initially set to 1. When the setting is completed, the visual sensor control unit 4 sends setting completion information to the information mediating unit 5. Next, in step 204, the information mediation unit 5 sends a 3D image recognition command to the visual sensor control unit 4 so as to perform 3D image recognition of the work on the tray.

  Next, in step 205, the information mediating unit 5 receives the 3D position / posture information as a result of the 3D image recognition of the workpiece from the visual sensor control unit 4. The three-dimensional position / posture information in the visual sensor control unit 4 is the position information and posture information of the workpiece obtained by the three-dimensional image recognition of the workpiece, and the coordinate information (X, Y, Z, RX, RY, RZ) of the workpiece. Represented as: Here, (X, Y, Z) is a coordinate value of orthogonal coordinates (X, Y, Z axes), and (RX, RY, RZ) represents an angular position around each of the X, Y, Z axes. Each element of the coordinate information is expressed as a numerical value up to 3 digits after the decimal point, the unit of the coordinate value is mm, and the unit of the angular position is degree. The three-dimensional position / posture information is sent as text data.

  Next, in step 206, the information mediation unit 5 converts the information format of the three-dimensional position / posture information into coordinate information that can be handled by the robot control unit 6. Specifically, a character string indicating the type of coordinate value is searched from text data, and the coordinate value of text data following the character string is converted into a numerical value. By repeating such processing, all coordinate values of the coordinate information (X, Y, Z, RX, RY, RZ) can be obtained.

  Further, these coordinate values are converted into integer values that can be handled by the robot controller 6. That is, each coordinate value (X, Y, Z) is multiplied by 10 and rounded such as rounding off to the decimal point to obtain a signed integer value, and each angular position (RX, RY, RZ) is multiplied by 100 to the decimal point. A rounding process such as rounding off is performed below to obtain a signed integer value.

  Next, in step 207, the information mediating unit 5 stores each coordinate value and each angular position converted to integers in a predetermined register. Then, information indicating that the storage of the coordinate information has been completed is sent to the robot controller 6. Thereby, the robot control unit 6 acquires the coordinate information of the workpiece and starts the workpiece take-out operation. When the removal and transfer of the workpiece are completed, the robot control unit 6 sends information on completion of removal to the information mediating unit 5.

  Next, after receiving the information of completion of extraction from the robot control unit 6, the information mediation unit 5 determines the next operation in step 208. If a work remains in the current window, the process returns to step 204 and steps 204 to 208 are repeated. If all the workpieces in the current window have been taken out and there are no workpieces, the process returns to step 203 to add 1 to the current window number and set it as a new window number. However, if the current window number is already the last number and there is no work, the process proceeds to the next step 209 without returning to the step 203.

  Next, in step 209, the information mediating unit 5 confirms that there is no work on the tray, and operates the tray changer to replace the tray. If all the trays have not been processed, the procedure returns to step 202 and steps 202 to 209 are repeated. When all trays have been processed, the process proceeds to the next step 210. In step 210, a command for terminating the program for automatic driving is sent to the robot controller 6. And the operation | movement at the time of the automatic driving | operation of the information mediation part 5 is complete | finished.

  As described above, by interposing the information mediation unit 5 between the visual sensor control unit 4 and the robot control unit 6, the signal formats and information formats of the visual sensor control unit 4 and the robot control unit 6 are different. However, it is now possible to control these robots by connecting them. The information mediation unit 5 not only physically connects the visual sensor control unit 4 and the robot control unit 6, but also converts the information format output from one into a format that can be handled by the other, and the position information and control information Are mutually available.

  In the robot control system as in the present invention, the three-dimensional visual sensor and the robot can be made to be the optimum models according to the work contents and the work environment. If a low-cost model is selected as the three-dimensional visual sensor and the robot, the overall cost of the robot control system can be reduced. Furthermore, even after constructing the robot control system, it is possible to change the model of the three-dimensional visual sensor or the robot. In that case, what is necessary is just to change the processing content of the information mediation part 5 as needed.

  In the above embodiment, the visual sensor control unit handles coordinate information and commands in text data format, and the robot control unit handles coordinate information in binary data format. However, this is merely an example. Any signal format or information format may be used by the visual sensor control unit and the robot control unit. By converting the information formats to each other by the information mediation unit, the visual sensor control unit and the robot control unit having different information formats can be connected to each other to form a robot control system. Further, although an articulated robot is exemplified as the robot, the present invention can also be applied to a robot of a type other than the articulated type.

  According to the present invention, there is provided a robot control system for detecting a three-dimensional information representing a position and posture of a workpiece by a three-dimensional visual sensor and controlling the robot based on the information to convey the workpiece. It is possible to provide a robot control system for controlling a robot by connecting the visual sensor and the robot regardless of the signal format and the information format.

DESCRIPTION OF SYMBOLS 1 Robot control system 2 3D visual sensor 3 Power supply for visual sensors 4 Visual sensor control part 5 Information mediation part 6 Robot control part 7 Robot 8 Tray 9 Work 21, 22 Camera 71 Hand

Claims (5)

A three-dimensional visual sensor (2) for detecting the three-dimensional position information and posture information of the workpiece (9);
A visual sensor control unit (4) for performing control and information processing of the three-dimensional visual sensor (2);
A robot (7) for transporting the workpiece (9);
A robot controller (6) for controlling the robot (7);
The information format output from the visual sensor control unit (4) is converted into a format that can be handled by the robot control unit (6) and sent to the robot control unit (6), and the robot control unit (6) A robot control system comprising: an information mediating unit (5) that converts an information format output from the visual sensor control unit (4) into a format that can be handled by the visual sensor control unit (4) and sends the information format to the visual sensor control unit (4). The robot control system according to claim 1,
The information mediating unit (5) is a robot control system using a control device for sequence control. The robot control system according to any one of claims 1 and 2,
The information mediating unit (5) is a robot control system for converting binary data representing numerical values and text data to each other. The robot control system according to claim 3,
The information mediating unit (5) is a robot control system that handles binary data representing numerical values as integer data. The robot control system according to any one of claims 1 to 4,
The robot (7) is a robot control system which is an articulated robot.

Images