JP4741691B2 - Robot system with robot abnormality monitoring function - Google Patents

Description

  The present invention relates to a robot system having a robot abnormality monitoring function, and more particularly, to a robot system capable of detecting an abnormality in an articulated robot constituted by a servo motor having an encoder even when the abnormality occurs in the encoder. .

  Conventionally, in an articulated robot constituted by a servo motor, a single encoder is provided inside the motor, and control is performed using the command value and the position and speed of each axis that has processed the feedback value from the encoder. Further, when the difference between the command value and the feedback value exceeds the allowable value, it is detected as an abnormality and the operation of the robot is stopped.

  It is important for safe control of the robot to detect a failure of the encoder itself such as an encoder output abnormality or an abnormality related to communication between the encoder and the control device. As such a technique, for example, Patent Document 1 discloses an abnormality detection device in which an absolute encoder and an increment encoder are installed on each axis and an abnormality is determined by comparing position information obtained by each means. . Patent Document 2 discloses a safety device in which safety is improved by doubling a communication path with an encoder.

JP-A-4-235610 JP-A-1-199201

  In each of the inventions described in the above-mentioned patent documents, improvement of the safety against an encoder failure or abnormality is achieved. However, in Patent Document 1, different types of encoders are used, and in Patent Document 2, a communication path to the encoder is set. The structure related to the encoder, such as double, becomes complicated, resulting in a cost increase. Therefore, there is a demand for a technique capable of reliably and quickly detecting an abnormality related to the output and communication of the encoder while having a conventional structure related to the encoder.

  Therefore, an object of the present invention is to provide a robot system having an abnormality monitoring function capable of detecting a failure or abnormality of an encoder that could not be detected conventionally.

In order to achieve the above object, an invention according to claim 1 includes a robot mechanism unit and a control device that controls the robot mechanism unit, and the control device includes an encoder that is included in a motor that drives the robot mechanism unit. A value or signal indicating the position of each axis is received from the controller, and a command or signal is transmitted to the motor so that the position of each axis at each time becomes a value defined in a predetermined robot program, thereby controlling the motor. In a robot system, a visual target arranged at a fixed position and having known graphic features is mounted on a single movable link of the robot mechanism unit at a position and posture for capturing the visual target within its visual field. Image processing is sequentially performed on an image obtained by imaging one visual sensor and a visual target placed at the fixed position by the visual sensor at a predetermined cycle. The spatial relative position of the visual sensor and the visual target is calculated according to how the visual target appears on the screen of the visual sensor, and is always in a fixed relative positional relationship with the visual sensor. A first measurement unit that sequentially obtains at least one of a spatial position and a moving speed of the point of interest as a first measurement value, and at least one of a spatial position and a moving speed of the point of interest based on the output of the encoder A second measurement unit that sequentially obtains two measurement values, a comparison unit that selects data at the same time from the first measurement value and the second measurement value, compares the selected data, and calculates the difference; when the difference obtained by the comparison unit exceeds a preset threshold, with a, an alarm output unit for outputting an alarm as a failure or abnormality of the encoder, the robot cis To provide a beam.

  According to a second aspect of the present invention, in the robot system according to the first aspect, a collaborative work area in which a work area of the robot mechanism unit and a human work area overlap is set in advance, and the robot mechanism unit performs the collaborative work. Detecting means for detecting whether or not it exists in the area, and comparing the first measured value and the second measured value only when the robot mechanism section exists in the cooperative work area. A robot system is provided.

  According to a third aspect of the present invention, in the robot system according to the first or second aspect, the control is performed so that the brake of the robot mechanism unit is released and the robot mechanism unit maintains a constant posture. In addition, a robot system is provided in which the first measurement value and the second measurement value are compared.

The invention according to claim 4 is a robot system that includes a robot mechanism unit and a control device that controls the robot mechanism unit, and controls the motor by an output of an encoder included in a motor that drives the robot mechanism unit. A visual sensor mounted on one movable link of the robot mechanism unit, and image processing is sequentially performed on an image obtained by capturing a landscape around the robot mechanism unit with a predetermined period by the visual sensor. Then, based on the output of the encoder of the first measurement unit that detects the spatial position of the visual sensor and the robot mechanism unit, the spatial point of interest that is always in a fixed relative positional relationship with the visual sensor. a second measurement unit for obtaining position sequentially, position variation amount calculated from the spatial position of the focus point obtained by the second measuring unit is set in advance threshold Nearby, and position variation by the first measuring unit is provided with a means for outputting an alarm when it is detected to provide a robot system.

  According to the robot system according to the present invention, the control device cannot be detected by the conventional method by measuring the fluctuation amount of the robot with the encoder and measuring the fluctuation amount with a visual sensor which is a device other than the encoder and comparing them. Even when an abnormality or failure occurs in the encoder, the controller can recognize the abnormality or failure.

  By applying the present invention to a robot system having a cooperative work area or a robot system in which the brake of the robot mechanism unit is released and the robot mechanism unit is controlled to maintain a constant posture, This can be achieved more reliably.

  By processing the image of the surrounding landscape of the robot mechanism obtained by the visual sensor, the position variation of the link to which the visual sensor is attached is obtained without preparing a specific visual target, and the encoder malfunctions or fails Can be detected.

It is a figure showing an example of 1 composition of a robot system concerning the present invention. It is a figure which shows the 1st measurement example using a visual target. It is a figure which shows the 2nd measurement example using a visual target. It is a figure which shows the 3rd example of a measurement using a visual target. It is a figure which shows the 4th example of a measurement using a visual target. It is a figure explaining an example of the flow of processing in a comparison part. It is a figure explaining the other example of the flow of a process in a comparison part. It is a figure which shows the example which imaged the surrounding scenery of the robot using the visual sensor.

  FIG. 1 shows a robot system 10 according to an embodiment of the present invention. The robot system 10 includes an articulated robot mechanism unit 12 and a robot control device 14 that controls the robot mechanism unit 12. The robot mechanism unit 12 includes at least one movable link, for example, a robot arm, and a servo motor that drives each robot arm. In the illustrated example, the robot mechanism unit 12 includes first to third robot arms 16, 18, 20, and first to first robot arms. This is an articulated robot having three servo motors 22, 24 and 26. Each servo motor has an encoder (first to third encoders 28, 30, 32 in the illustrated example) for detecting its own position (in many cases, it is built in). Further, a visual sensor 34 such as a camera is attached to one of the robot arms (usually the third robot arm 20 at the most distal end side) at a position and posture where a visual target such as a marker described later can be detected or imaged.

  The robot control device 14 performs a motor control unit 36 that controls the first to third servo motors 22, 24, 26, an image processing unit 38 that processes an image obtained by the visual sensor 34, and a comparison process that will be described later. A comparison unit or comparator 40 and an alarm output unit 42 that outputs an alarm based on the processing result of the comparison unit 40 are provided. The motor control unit 36 receives a value or signal indicating the position of each servo motor (position of each axis) from the encoders 28, 30, and 32, and the position of each axis at each time is defined by a predetermined robot program or the like. A command or signal is transmitted to each servo motor so as to be a value, and each servo motor is controlled.

  The feature of the present invention is that the spatial position or movement speed of the robot is calculated by sequentially processing the images obtained by the visual sensor periodically, and at the same time the spatial position or movement of the robot obtained via the encoder. It is to calculate the speed and monitor each difference. For example, when a visual sensor is mounted on the wrist axis of a 6-axis robot, the spatial position or moving speed by the respective methods at the visual sensor mounting position and the wrist tip portion is compared.

  As a position detection method using a visual sensor, a visual target arranged in a fixed position with known graphic features is detected, and the visual sensor and the visual target are detected depending on how the visual target appears on the screen of the visual sensor. The spatial relative position is calculated. The spatial relative position of the target point at a certain relative position from the visual sensor with respect to the visual target is calculated by calculating the relative position of the target point and the visual sensor and the relative position of the visual sensor and the visual target. it can. By calculating this periodically, the moving distance of the visual point or the point of interest at a certain relative position from the visual sensor is calculated. Here, by calculating the moving distance periodically, it is also possible to calculate the moving speed of the visual point or the point of interest at a certain relative position from the visual sensor. Hereinafter, the example is demonstrated using FIGS.

  FIG. 2 is a diagram for explaining a first measurement example using a target (here, a marker 44) arranged at a fixed position in a space where the above-described visual sensor 34 can capture an image. The marker 44 in the illustrated example is obtained by adding one diameter to a circle, but applicable markers are not limited to such a shape. FIGS. 2A and 2B show images 44a and 44b obtained by imaging the marker 44 by the visual sensor 34 at times T1 and T2, respectively. First, as shown in FIG. 2A, the image processing unit 38 of the control device 14 described above performs the shortest distance X1 in the X direction and the shortest distance Y1 in the Y direction of the marker image 44a from the predetermined origin O, and the marker image. The X direction length α1 and the Y direction length β1 of 44a are calculated and stored in a suitable memory or the like (not shown) as parameters at time T1.

  Next, as shown in FIG. 2B, the image processing unit 38 includes the shortest distance X2 in the X direction between the predetermined origin O and the marker image 44b, the shortest distance Y2 in the Y direction, and the length in the X direction of the marker image 44b. Α2 and Y-direction length β2 are calculated and stored in a suitable memory or the like (not shown) as parameters at time T2. With the above parameters, it is possible to calculate the spatial position and moving speed of the point of interest of the robot that is always in a certain relative positional relationship with the camera 34. The obtained spatial position and moving speed can also be stored in a memory or the like. As for the specific calculation method, since a well-known method can be applied, detailed description is omitted, but it is not necessary to use all the above parameters, and other parameters representing the position or shape of the marker image may be used. Good.

  FIG. 3 is a diagram for explaining a second measurement example using targets (here, markers 44 and 46) arranged at fixed positions in a space where the visual sensor 34 can capture images. The marker 46 in the illustrated example has a circular shape with two diameters orthogonal to each other, but applicable markers are not limited to such shapes. FIGS. 3A and 3B show images 44a and 46a and 44b and 46b obtained by imaging the markers 44 and 46 by the visual sensor 34 at times T1 and T2, respectively. First, as shown in FIG. 3A, the image processing unit 38 of the control device 14 described above is the distance in the X direction between the representative position (for example, the center of gravity) of the marker image 44a and the representative position (for example, the center of gravity) of the marker image 46a. The distance B1 in the A1 and Y directions and the X-direction lengths α1, γ1 and the Y-direction lengths β1, δ1 of the marker images 44a and 46a are calculated and stored in a suitable memory or the like (not shown) as parameters at the time T1. To do.

  Next, as illustrated in FIG. 3B, the image processing unit 38 includes the distance A2 in the X direction and the Y direction between the representative position (for example, the center of gravity) of the marker image 44b and the representative position (for example, the center of gravity) of the marker image 46b. The distance B2 and the X-direction lengths α2 and γ2 and the Y-direction lengths β2 and δ2 of the marker images 44b and 46b are calculated and stored in a suitable memory or the like (not shown) as parameters at time T2. Based on the above parameters, it is possible to calculate the spatial position and moving speed of the target point of the robot (for example, a part of the third robot arm to which the camera 34 is fixed) that is always in a fixed relative positional relationship with the camera 34. The obtained spatial position and moving speed can also be stored in a memory or the like. As for the specific calculation method, since a well-known method can be applied, detailed description is omitted, but it is not necessary to use all the above parameters, and other parameters representing the position or shape of the marker image may be used. Good.

  FIG. 4 is a diagram for explaining a third measurement example using a marker (here, a rectangular shape 48 with two sides curved) arranged at a fixed position in a space where the above-described visual sensor 34 can capture an image. The applicable target is not limited to such a shape. FIGS. 4A and 4B show images 48a and 48b obtained by imaging the marker 48 (FIG. 1) with the visual sensor 34 at times T1 and T2, respectively. First, as shown in FIG. 4A, the image processing unit 38 of the above-described control device 14 includes the shortest distance X1 in the X direction and the shortest distance Y1 in the Y direction from the predetermined origin O, and the marker image 48a. And the inclination angle θ1 of a predetermined axis (longitudinal axis 50a in the illustrated example) of the marker image 48a are calculated and stored as parameters at time T1 in an appropriate memory or the like (not shown).

  Next, as illustrated in FIG. 4B, the image processing unit 38 includes the shortest distance X2 in the X direction and the shortest distance Y2 in the Y direction from the predetermined origin O, and the length in the X direction of the marker image 48b. α2 and the inclination angle θ2 of a predetermined axis (longitudinal axis 50b in the illustrated example) of the marker image 48b are calculated and stored in a suitable memory or the like (not shown) as a parameter at time T2. Based on the above parameters, it is possible to calculate the spatial position and moving speed of the target point of the robot (for example, a part of the third robot arm to which the camera 34 is fixed) that is always in a fixed relative positional relationship with the camera 34. The obtained spatial position and moving speed can also be stored in a memory or the like. As for the specific calculation method, since a well-known method can be applied, detailed description is omitted, but it is not necessary to use all the above parameters, and other parameters representing the position or shape of the marker image may be used. Good.

  FIG. 5 is a diagram for explaining a fourth measurement example using targets (here, markers 44, 46, and 48) arranged at fixed positions in a space where the visual sensor 34 can capture images. The applicable marker is not limited to such a shape. FIGS. 5A and 5B show images 44a, 46a and 48a, and 44b, 46b and 48b obtained by imaging the markers 44, 46 and 48 by the visual sensor 34 at times T1 and T2, respectively. ing. First, as shown in FIG. 5A, the image processing unit 38 of the control device 14 described above represents the representative position (for example, the center of gravity) of the marker image 44a, the representative position (for example, the center of gravity) of the marker image 46a, and the representative of the marker image 48a. The distances A1, B1 and C1 between the positions (for example, the center of gravity), the angle α1 formed by the line segments A1 and C1, and the angle β1 formed by the line segments A1 and B1 are calculated, and an appropriate memory or the like as a parameter at the time T1 ( (Not shown).

  Next, as illustrated in FIG. 5B, the image processing unit 38 represents the representative position (for example, the center of gravity) of the marker image 44b, the representative position (for example, the center of gravity) of the marker image 46b, and the representative position (for example, the center of gravity) of the marker image 48b. Distances A2, B2 and C2, angle A2 formed by line segments A2 and C2, and angle β2 formed by line segments A2 and B2 are calculated and stored in an appropriate memory (not shown) as a parameter at time T2. Remember. Based on the above parameters, it is possible to calculate the spatial position and moving speed of the target point of the robot (for example, a part of the third robot arm to which the camera 34 is fixed) that is always in a fixed relative positional relationship with the camera 34. The obtained spatial position and moving speed can also be stored in a memory or the like. As for the specific calculation method, since a well-known method can be applied, detailed description is omitted, but it is not necessary to use all the above parameters, and other parameters representing the position or shape of the marker image may be used. Good.

  As illustrated in FIGS. 2 to 5, the image processing unit 38 captures at least one of the visual targets 44, 46, and 48 disposed at a fixed position by the camera 34 with respect to an image acquired by a predetermined period. The image processing is sequentially performed, and at least one of the spatial position and the moving speed of the point of interest always having a fixed relative positional relationship with the camera 34 is sequentially obtained as the first measurement value, and the result is transmitted to the comparison unit 40. .

  Specific examples of the point of interest include fixing a part of the third robot arm 20 to which the camera 34 is fixed, a tool tip point (TCP), a bracket for fixing the camera 34 to the third robot arm 20, and the like. Examples include a part of a jig (not shown), etc., as long as it is always in a fixed relative positional relationship with the camera 34 (the relative positional relationship with the camera is not changed by the operation of each axis). May be used. Further, the position detection by the visual sensor described with reference to FIGS. 2 to 5 is an example, and various other marker shapes and three-dimensional shapes can be used.

  On the other hand, the motor control unit 36 of the control device 14 sequentially obtains at least one of the spatial position and the moving speed of the target point as the second measurement value based on the output of the encoder, and the result is sent to the comparison unit 40. Send. The motor control unit 36 takes in the output of each encoder at every predetermined sampling period and calculates the position and speed of each axis at each time. From this calculation result, the dimensions of the robot parts such as the arm lengths of the robot arms 16, 18, and 20 constituting each axis, and the positional relationship between the robot parts and the point of interest, the spatial position of the point of interest is obtained. It is done. Further, the moving speed can be obtained by periodically calculating the spatial position.

  The comparison unit 40 of the control device 14 includes the spatial position or moving speed of the target point of the robot obtained by the method using the visual sensor illustrated in FIGS. 2 to 5 and the spatial position of the target point obtained by the encoder. Or compare the moving speed. For comparison of position and speed, the timing of image capture by the visual sensor and the timing of capture of the encoder output need to be the same, but “simultaneous” or “same time” used in this specification is strictly In addition to the above, it is assumed that the timing of image capture by the visual sensor and the timing of capture of the encoder output are within a predetermined time difference that is regarded as substantially simultaneous in the implementation of the present invention.

  6 and 7 are diagrams for explaining the flow of processing for comparing the calculation position or calculation speed of the target point by the visual sensor and the calculation position or calculation speed of the target point by the encoder at the same time. First, in the process shown in FIG. 6, the motor control unit 36 and the image processing unit 38 include data including data acquisition times by the encoder and the visual sensor, and at least one of the position and speed of the point of interest calculated at each acquisition time. Is transmitted to the comparison unit 40. More specifically, in the motor control unit 36, based on the output from the encoder, the comparator 40 uses at least one of the position and speed of the point of interest at time Ti (i = 1, 2,...) As the second measured value. On the other hand, in the image processing unit 38, based on the output from the visual sensor, at least one of the position and speed of the point of interest at the time Tj (j = 1, 2,...) Is used as the first measurement value to the comparator 40. Send. That is, in the present embodiment, the motor control unit 36 and the image processing unit 38 function as a second measurement unit that obtains the second measurement value and a first measurement unit that obtains the first measurement value, respectively.

  The comparison unit 40 selects data in which the difference between the capture time (Ti) by the encoder and the capture time (Tj) by the visual sensor is within a predetermined time difference, and at least one of the position and speed of the point of interest by the visual sensor (first) 1) and at least one of the position and speed of the point of interest by the encoder (second measurement value). Here, if the compared position or speed difference exceeds a predetermined threshold (that is, there is a significant difference between the measured value by the encoder and the measured value by the visual sensor) Since it is considered that some abnormality such as a communication abnormality has occurred, the comparison unit 40 sends a command or signal for issuing an alarm to the alarm output unit 42, and the alarm output unit 42 issues an alarm. Further, with the occurrence of the alarm, the control device 14 can take appropriate measures such as a robot emergency stop.

  In FIG. 7, instead of the process of comparing the capture times in the case shown in FIG. 6 to determine whether the time is the same time, the data capture time by the encoder and the visual sensor is defined in advance by the capture signal or the like at the same time. It is a figure which shows a process in the case of being. In this case, since it is ensured that the data acquisition time by the encoder and the visual sensor becomes the same time T by the acquisition signal or the like, the comparison unit 40 may simply compare the data at the same time T. In the subsequent processing, as in the case of FIG. 6, when the compared position or speed difference exceeds a predetermined threshold value, it is considered that some abnormality has occurred in the encoder. A command or signal for issuing an alarm is sent to the unit 42, and the alarm output unit 42 issues an alarm. Further, with the occurrence of the alarm, the control device 14 can take appropriate measures such as a robot emergency stop.

  As a preferred application example of the present invention, there is a robot system in which a human and a robot collaborate. In such a system, in order to ensure human safety, the robot exists in a collaborative work area where the work area of the robot mechanism unit and the human work area overlap (that is, an area where both the human and the robot can enter). Detection means such as an area sensor, a proximity sensor, a surveillance camera, etc. that detect the presence of a robot in the cooperative work area is transmitted to the control device when the detection device detects the presence of a robot in the cooperative work area. Only when the robot is present in the cooperative work area, simultaneous measurement / comparison of the position or speed of the point of interest by the encoder and the visual sensor described above can be performed.

  As another preferred application example of the present invention, in a robot system in which a visual sensor is attached to a robot arm, the robot mechanism unit is stationary without using a brake. That is, even when control is performed so that the brake of the robot mechanism unit is released and the robot mechanism unit maintains a constant posture, the actual measurement is performed by comparing the first measurement value with the second measurement value. In this case, it is possible to prevent the robot from being determined to be stationary due to an encoder error even though the robot is moving, which is effective in ensuring the safety of the operator.

  FIG. 8 is a diagram for explaining a case where image processing such as contour extraction is performed on an arbitrary image (for example, a landscape surrounding the robot mechanism) taken periodically in a robot system in which a visual sensor is attached to a robot arm. is there. For example, when an image 52 as shown in FIG. 8A is obtained by a visual sensor at time T1, and then an image as shown by a solid line 54 in FIG. 8B is obtained at time T2, two images are displayed. By performing superimposition processing or the like, it is possible to detect whether the camera position has changed. In the illustrated example, as can be seen from FIG. 8B, the image position is shifted from the dotted line 56 (corresponding to the solid line 50 in FIG. 8A) to the solid line 54 (upward) from time T1 to T2. Therefore, it can be determined that there has been a change in the camera position (a change in the robot arm on which the camera is mounted). At this time, if the change in the robot position due to the encoder output is not confirmed between time T1 and time T2, it recognizes that the encoder is malfunctioning or abnormal and outputs an alarm.

  The robot system according to the present invention processes an image obtained by a visual sensor mounted on the robot without changing the number of encoders and the encoder communication form in the prior art, and based on the image processing result, The speed is calculated and compared with the position and speed of the robot calculated by the encoder. Here, if the encoder breaks down and accurate encoder information cannot be obtained, there will be a significant difference between the calculated values of each method. If the difference exceeds the allowable value, the encoder By detecting this as an abnormality, it becomes possible to recognize a failure that could not be detected in the past.

10 Robot system 12 Robot (mechanism)
DESCRIPTION OF SYMBOLS 14 Control apparatus 16, 18, 20 Robot arm 22, 24, 26 Servo motor 28, 30, 32 Encoder 34 Visual sensor 36 Motor control part 38 Image processing part 40 Comparison part 42 Alarm output part 44, 46, 48 Marker

Claims (4)

A robot mechanism unit and a control device for controlling the robot mechanism unit, wherein the control device receives a value or signal indicating the position of each axis from an encoder included in a motor that drives the robot mechanism unit, and at each time In the robot system for controlling the motor by transmitting a command or signal to the motor so that the position of each axis becomes a value defined in a predetermined robot program ,
A visual target placed in a fixed position and having known graphic features;
One visual sensor mounted on one movable link of the robot mechanism unit at a position and posture for capturing the visual target within its visual field;
How the visual target is seen on the screen of the visual sensor by sequentially performing image processing on an image acquired by imaging the visual target placed at the fixed position by the visual sensor at a predetermined cycle. To calculate a spatial relative position between the visual sensor and the visual target, and at least one of a spatial position and a moving speed of a point of interest that is always in a fixed relative positional relationship with the visual sensor is sequentially set as a first measurement value. A first measuring unit to be obtained;
A second measurement unit for sequentially obtaining at least one of a spatial position and a moving speed of the point of interest as a second measurement value based on the output of the encoder;
A comparison unit that selects data at the same time from the first measurement value and the second measurement value, compares the selected data, and calculates the difference;
An alarm output unit that outputs an alarm as a failure or abnormality of the encoder when the difference obtained by the comparison unit exceeds a preset threshold; and
Robot system equipped with.   The robot mechanism includes detection means for presetting a collaborative work area where a work area of the robot mechanism unit and a human work area overlap, and detecting whether the robot mechanism unit exists in the collaborative work area, 2. The robot system according to claim 1, wherein the first measurement value and the second measurement value are compared only when a unit exists in the cooperative work area.   The first measurement value and the second measurement value are compared while the brake of the robot mechanism unit is released and control is performed so that the robot mechanism unit maintains a constant posture. The robot system according to claim 1 or 2. In a robot system that includes a robot mechanism unit and a control device that controls the robot mechanism unit, and controls the motor by an output of an encoder included in a motor that drives the robot mechanism unit.
A visual sensor mounted on one movable link of the robot mechanism,
A first measurement unit that detects a change in the spatial position of the visual sensor by sequentially performing image processing on an image obtained by capturing the landscape around the robot mechanism unit with a predetermined period by the visual sensor. When,
A second measuring unit for sequentially obtaining a spatial position of a point of interest that is always in a certain relative positional relationship with the visual sensor based on an output of the encoder of the robot mechanism unit;
An alarm is generated when the amount of position fluctuation calculated from the spatial position of the point of interest obtained by the second measuring section is within a preset threshold value and position fluctuation is detected by the first measuring section. Means for outputting;
Robot system equipped with.

Images