JP2011212837A - Robot system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enhance response of direct teaching of a 6-axis robot while improving its safety.SOLUTION: In direct teaching of a 6-axis vertical articulated robot, a brake of a motor sequentially corresponding to each axis is canceled in a predetermined order for an arbitrary time. At this time, a dynamic brake is applied to the motors of the other axes. The arbitrary time is, for example, 1 ms, so that a worker can get a feeling of good response for all axes without caring about switching per axis unit in a teaching operation. Until the direct teaching is completed, brake canceling operations are repeated for all axes.

Description

  The present invention relates to a robot system capable of direct teaching with respect to a robot having a plurality of axes.

  As a robot having a plurality of axes, for example, in the case of a 6-axis robot, there is a method of setting by a data input or manual operation of a robot arm using a teaching pendant when teaching a position and orientation. On the other hand, there is direct teaching corresponding to the case where the position and orientation are set by moving the robot arm directly by the operator.

  In this case, when performing direct teaching, it is necessary to release the brakes of the motors of all axes in order to set the robot arm to a desired position and orientation. However, if the brakes of all the motors are released simultaneously, there is a possibility that a situation such as dropping the hand of the robot arm may occur. As a result, there is a risk of injury to the worker, so it is necessary to take sufficient safety measures when performing direct teaching.

  In this regard, for example, in the technique disclosed in Patent Document 1, when direct teaching is performed for a six-axis robot, a technique is disclosed in which the brakes for all axes are not released simultaneously. That is, in the thing of patent document 1, two groups of the basic axis (1st axis | shaft-3rd axis) and a hand axis | shaft (4th axis | shaft-6th axis) of 6 axes | shafts are formed, and only one of these groups is formed. The brake is released and direct teaching work is performed. Thereby, it is possible to suppress the occurrence of the situation such as the drop of the hand of the robot arm as described above.

Japanese Patent Laid-Open No. 11-77571

  However, in the one disclosed in Patent Document 1 described above, switching of the brake release of the two groups of the basic axis (first axis to third axis) and the hand axis (fourth axis to sixth axis) is artificially performed by an operator. In this respect, the responsiveness when moving the robot arm during direct teaching deteriorates, making it difficult for the operator to perform sensuous teaching, which is one of the features of original direct teaching. Met.

  The present invention has been made in consideration of the above circumstances, and the purpose of the present invention is to provide a direct teach response while preventing direct hand dropping during work when performing direct teaching in a robot having a plurality of axes. It is an object of the present invention to provide a robot system that can ensure sufficient performance.

  According to the robot system of the first aspect, when performing direct teaching of the robot, the controller releases the brake of the motor for each predetermined axis in a preset order for the plurality of axes of the articulated robot. A brake release step is performed for a set predetermined time. Then, the brake release step is repeatedly performed until the teaching operation by the user is completed. Here, the predetermined axis may be one axis, or there may be another axis having a condition for releasing the brake in accordance with the release of the brake of one axis. Further, the set time may be a time that is fixedly set for each axis, or may be a time that is variably set according to the operation state of each axis.

  As a result, when direct teaching is being performed, the brake is always released only for the specified axis, and the axis to be released is switched after the set time has elapsed. It is possible to prevent the shaft that has not been moved and dropped at a stretch, and to ensure safety during work. In addition, when the set time has elapsed for the axis to be released, all the axes will move little by little due to the time difference, but the movement of the hand in direct teaching was originally slow because it was a human work Since movement is a prerequisite, the operator can operate with the same feeling as releasing all the brakes simultaneously, and as a result, responsiveness can be ensured.

  According to the robot system of the second aspect, the brake releasing step is repeatedly executed by the controller as one cycle in which the preset order is rotated once, and within one cycle, the brake of the motor is performed for each axis. The set release time is a variable release time that is calculated in proportion to the movement amount so that the fixed opening time that is fixedly secured becomes longer as the movement amount of the axis in the previous cycle increases. Thus, a time is set by adding the variable opening time calculated so that the total time calculated for each axis is always a predetermined time.

  As a result, the brake release of the motor is performed for each of the plurality of axes, and the set time in this case is set to a time obtained by adding the variable release time to the fixed release time, thereby improving safety. In addition, the workability of teaching can be improved. For each axis, for example, even if the movement amount is small or the axis is not moving, only the fixed release time is secured as the set time for releasing the brake, so that a movable state is ensured every time. . In addition, for axes with a large amount of movement, time that is almost directly proportional to the amount of movement is added by the variable opening time so that movement can be prioritized. can do.

  According to the robot system of the third aspect, the controller executes the following first variable opening time setting process as the calculation of the variable opening time. In the first variable opening time setting process, first, the movement amount and average movement amount of each axis in the previous cycle are calculated, and then the deviation of each axis obtained by subtracting the average movement amount from the movement amount of each axis. The amount of movement is calculated. Next, calculate the distributed denominator coefficient by adding the values of the deviation movement amount of each axis with the same sign, and then add “1” to the value obtained by dividing the deviation movement amount of each axis by the distribution denominator coefficient. The calculated value is calculated as a change coefficient for each axis. Then, a variable opening time is set by multiplying the reference time preset as the initial value of the variable opening time by the coefficient of change of each axis.

  As a result, when setting the variable opening time of each axis, it is possible to set a long time for the movement amount of the corresponding axis in the previous cycle, and of course, it is distributed to the setting of the variable opening time. Since the denominator coefficient is calculated and distributed, the actual amount of movement will be evaluated after being compressed to an increase of several percent to several tens of percent. Appropriate time can be allocated to all the axes while suppressing the occurrence of this. For this reason, when each axis is desired to be moved finely, such as when finely adjusting the position of the hand, each axis is easily operated, and the operability is excellent.

  According to the robot system of the fourth aspect, the controller executes the following second variable opening time setting process as the calculation of the variable opening time. In the second variable opening time setting process, first, the movement amount and the average movement amount of each axis in the previous cycle are calculated, and then the deviation of each axis obtained by subtracting the average movement amount from the movement amount of each axis. The amount of movement is calculated. Then, a value obtained by dividing the deviation movement amount of each axis by a predetermined reference movement amount is set as an increase rate, and a value obtained by adding “1” to the increase rate is calculated as a change coefficient of each axis, and an initial value of the variable opening time As the variable opening time, a value obtained by multiplying the preset reference time by the change coefficient of each axis is set.

  As a result, when setting the variable opening time of each axis, it can be set to be a long time for the movement amount of the corresponding axis in the previous cycle, but the setting of the variable opening time is applicable. Since a large amount can be distributed almost directly proportional to the amount of movement of the shaft, the variable opening time of the shaft with a large amount of movement can be preferentially distributed, and the workability of direct teaching when the amount of movement is large is given priority. Can be made. For this reason, if you want to move your hand greatly toward the teaching position, the variable opening time can be distributed long for the axis that moves greatly due to the large movement of the hand, so move the axis you want to move greatly. The hand can be easily moved to near the teaching position, and the operability is excellent.

  According to the robot system of the fifth aspect, the controller determines that the movement amount in the previous cycle of a specific axis among the plurality of axes is equal to or less than a preset threshold as the calculation of the variable opening time. Sometimes the first variable opening time setting process is executed, and when it is determined that the movement amount of the specific axis in the previous cycle exceeds the threshold value, the second variable opening time setting process is executed. In this case, all of the plurality of axes may be targeted as the specific axis, or an axis required for direct teaching workability may be targeted.

  Thus, according to the direct teaching work, the first variable opening time setting process is performed based on whether the movement amount of the axis set as the specific axis exceeds the threshold value or the second variable is set. It is possible to automatically switch whether to perform the opening time setting process, and depending on the movement method of a specific axis, whether it is a work with a large movement amount in the work, or a detailed work with a small movement amount Since the method of allocating the variable opening time can be switched and set according to the difference in the work content of whether or not the work is being performed, it is suitable for work where the movement amount is drastically changed.

  According to the robot system of the sixth aspect, when the second variable opening time setting process is executed by the controller, if the deviation movement amount of each axis is less than “−1”, the increase is the smallest. The reciprocal of the absolute value of the rate is calculated as a compression coefficient. In calculating the change coefficient for each axis, “1” is added after the increase rate of the deviation movement amount of each axis is multiplied by the compression coefficient.

  As a result, when the second variable opening time setting process is performed, the deviation movement amount is relatively small for an axis with a very small movement amount with respect to an axis with a large movement amount. Is less than “−1”, by setting the compression coefficient, it is possible to prevent the variable opening time from being excessively allocated to the shaft with a large amount of movement, and at the same time, within one cycle determined for a certain period of time, It is possible to secure a fixed release time as a minimum brake release time even for an axis with a very small amount of movement. As described above, an appropriate variable opening time can be allocated to an axis with a large amount of movement in the previous period while suppressing the occurrence of a problem that an axis with a very small amount of movement in the previous period cannot be released without releasing the brake. it can.

  According to the robot system of the seventh aspect, when performing direct teaching of the robot, the controller releases the brake of the motor for each axis in a predetermined order with respect to a plurality of axes of the articulated robot. Perform the brake release step to be performed for hours. Then, the brake release step is repeatedly performed until the teaching operation by the user is completed.

  As a result, when direct teaching is being performed, the brake is always released only for one axis, and the axis to be released is switched at regular intervals, so only a specific axis is dropped at once. It is prevented and safety during work can be secured. In addition, when the axis to be released is switched at regular intervals, all axes will move little by little due to the time difference, but the movement of the hand in direct teach is supposed to be a slow movement from the point of human work As a result, the operator can operate with the same feeling as releasing all the brakes at the same time, and as a result, responsiveness can be ensured.

  According to the robot system of the eighth aspect, when performing direct teaching of the robot, the controller sets the release target axes in a preset order for the plurality of axes of the articulated robot. It is determined whether or not there is an axis corresponding to the orthogonal order sequentially with respect to the axis after the release target axis among the remaining axes with respect to the release target axis. Set as the simultaneous release target axis, perform the brake release step for releasing the motor brake for a certain period of time for the release target axis and the simultaneous release target axis, and perform the brake release step until the teaching operation by the user is completed.

  As a result, only the axes that are in a perpendicular or orthogonal relationship that does not hinder even if the brakes are released at the same time will be released by releasing the brakes for a certain period of time. In this work, it is possible to improve the workability by increasing the degree of freedom of arm movement while suppressing the occurrence of an arm drop accident.

  According to the robot system of claim 9, the controller determines the simultaneous release target axes, and the remaining order of the remaining release axes with respect to the set release target axes follows the axis behind the release target axes. As a result, it is determined whether there is an orthogonally equivalent axis in the order of the axis that is earlier than the release target axis, so that the probability of occurrence of an axis that can be released at the same time can be increased. Workability in teaching can be further improved.

Electrical block diagram showing the first embodiment of the present invention External configuration diagram of the robot system Flow chart showing the flow of control for one cycle of the brake release step of direct teaching A diagram schematically showing the starting position and teaching position of the robot Explanation of direct teaching operation (1) Explanation of direct teaching operation (2) FIG. 3 equivalent view showing the second embodiment of the present invention FIG. 3 equivalent view showing a third embodiment of the present invention Flowchart of direct teaching control showing the fourth embodiment of the present invention 3 equivalent figure Flow chart of control for setting variable opening time proportional to movement amount The figure which shows the concrete setting example of the movement amount and opening time of each axis FIG. 11 equivalent diagram showing the fifth embodiment of the present invention The figure which shows the specific setting example of the movement amount of each axis | shaft, and opening time (the 1) FIG. 2 is a diagram showing a specific setting example of the movement amount and opening time of each axis (No. 2) FIG. 11 equivalent view showing the sixth embodiment of the present invention The figure which shows the concrete setting example of the movement amount and opening time of each axis Graph showing the transition of opening time of each axis

(First embodiment; claims 1 and 7)
Hereinafter, a first embodiment of the present invention will be described with reference to FIGS.
FIG. 2 shows an external configuration of the robot system 1. The robot system 1 includes a robot 2, a controller 3 that controls the robot 2, and a teaching pendant 4 connected to the controller 3.

  The robot 2 is configured as a vertical 6-axis articulated robot. The robot 2 includes a base 5, a shoulder portion 6 supported by the base 5 so as to be able to turn in a horizontal direction around the first axis L 1 of the rotational joint axis, and a second rotational joint axis supported by the shoulder portion 6. A lower arm 7 supported so as to be pivotable in the vertical direction about the axis L2, and a first upper arm 8 supported to be pivotable in the vertical direction about the third axis L3 of the rotary joint axis. A second upper arm 9 supported at the tip of the first upper arm 8 so as to be able to twist and rotate about the fourth axis L4 of the rotary joint axis, and a rotary joint on the second upper arm 9 The wrist 10 is supported so as to be pivotable in the vertical direction about the fifth axis L5 of the shaft, and the flange 11 is supported on the wrist 10 so as to be able to twist and rotate about the sixth axis L6 of the rotary joint shaft. ing. An end effector (not shown) such as a hand for gripping a workpiece or a camera used for visual inspection is attached to the flange 11 which is the tip of the robot.

  The base 5, the shoulder portion 6, the lower arm 7, the first upper arm 8, the second upper arm 9, the wrist 10, and the flange 11 function as arms in the robot 2, and each arm excluding the base 5 that is a fixed arm Reference numerals 6 to 11 are fixedly connected to the first axis L1 to the sixth axis L6, which are rotary joint axes rotatably supported by the previous stage arm.

  Next, in FIG. 1 showing the electric block configuration, the robot 2 is provided with a servo for rotating around the first axis L1 to the sixth axis L6 that are the rotary joint axes of the arms 6 to 11 described above. Motors M1 to M6 including motors are provided. The arms 6 to 11 are configured to be rotationally driven by a rotation of the motors M1 to M6 via a speed reduction mechanism (not shown). The motors M1 to M6 are provided with encoders S1 to S6 connected to the first shaft L1 to the sixth shaft L6, and are configured to output a rotation detection signal to the controller 3.

  The controller 3 that controls the operation of the robot 2 has a microcomputer as a main component, and includes a CPU 12, a ROM 13 and a RAM 14 as storage means. The ROM 13 includes a memory such as an electrically rewritable flash memory (a kind of EEPROM) in addition to a general read-only memory. The ROM 13 stores various control programs for driving and controlling the robot 2 including a control program related to direct teaching described later.

  The controller 3 is provided with a drive circuit 15 for driving the motors M1 to M6 of the robot 2. The controller 3 is provided with a rotation position detection circuit that detects the rotation angle of each of the motors M1 to M6 based on detection signals from the encoders S1 to S6 on the robot 2 side. Based on this, a control signal for controlling the position and orientation is output to the drive circuit 15.

  Next, the first axis L1 to the sixth axis L6 that are the six rotational joint axes of the arms 6 to 11 of the robot 2 will be described. Three-dimensional coordinates are defined for the base 5 and the arms 6 to 11. Among these, the coordinate system of the base 5 (indicated by coordinate axes XYZ in FIG. 2) installed on the floor surface is a stationary coordinate system and is set as a reference coordinate (robot coordinate). The coordinates of the arms 6 to 11 are set on the rotation center axes of the first axis L1 to the sixth axis L6 of the rotary joint axes of the arms 6 to 11, and the rotation of the arms 6 to 11 causes the coordinates on the robot coordinates. The position and posture change.

  The ROM 13 stores the coordinate position of the shoulder section 6 on the robot coordinates, the coordinate position of the lower arm 7 on the coordinates of the shoulder section 6, the coordinate position of the first upper arm 8 on the coordinates of the lower arm 7, and the first upper position. The coordinate position of the second upper arm 9 on the coordinates of the arm 8, the coordinate position of the wrist 10 on the coordinates of the second upper arm 9, the coordinate position of the flange 11 on the coordinates of the wrist 10, and the arms 6 to 11 Various parameters such as length are stored. In the following, the position of the coordinates of the arms 6 to 11 on the robot coordinates is simply referred to as the position of the arms 6 to 11.

  The origin of the coordinates of the flange 11 that is the tip of the robot is determined at the center of rotation of the tip surface of the flange 11. The CPU 12 has a coordinate conversion calculation function. When the position of the robot tip (including the posture; the same applies hereinafter) is given, the rotation of the arms 6 to 10 such that the robot tip takes the given position. The angle can be calculated (reverse conversion), and when the rotation angles of the arms 6 to 11 are given, the positions of the arms 6 to 11 can be calculated (forward conversion).

  An operation program for setting the operation of the robot 2 by teaching is created by the teaching pendant 4 and stored in the ROM 13 or the RAM 14. The controller 3 calculates the movement trajectory of the robot tip based on the control program stored therein, and the speed pattern of the robot tip is changed to a trapezoid pattern or a triangle while the robot tip moves from the start position to the teaching position. The position of the tip of the robot at every sampling time is calculated so that a predetermined speed pattern such as a pattern is obtained.

  During the operation control of the robot 2, the controller 3 calculates the rotation angles of the arms 6 to 10 from the rotation angles of the motors M1 to M6 given from the encoders S1 to S6 at each sampling time. The current position is detected by calculating the position of the flange 11 from the rotation angle. Then, the rotation angle of each arm 6-11 at the position next to the current position of the robot tip is acquired from the memory, the rotation angle of each motor M1-M6 is calculated from the rotation angle of each arm 6-11, and each rotation is performed. The angle is output to the drive circuit 15 as the target rotation angle of each of the motors M1 to M6.

Next, a case where direct teaching of the robot 2 is performed in the robot system configured as described above will be described with reference to FIGS.
When the direct teaching mode is set from the teaching pendant 4 by the operator, the controller 3 executes the direct teaching control program according to the procedure shown in FIG. In direct teaching, the operator moves the arms 6 to 11 of the robot 2 directly to the teaching position for setting. For example, as shown in FIG. 4, the case of moving from the start position shown in (a) to the teaching position shown in (m) will be described. Further, in this embodiment, the setting order of the first axis L1 to the sixth axis L6, which are rotary joint axes, is set so as to shift from the first axis L1 to the sixth axis L6 in order of numbers.

  This order does not necessarily have to be set in the order of the axis number from the first axis L1, and can be set as desired in accordance with the work. Moreover, about the 6th axis | shaft L6 which twists the flange 11, depending on the end effector attached, it may be able to be made into a free state by always releasing a brake at the time of teaching operation.

  First, the controller 3 releases the brake of the motor M1 of the first axis L1 to make it free, and also applies dynamic brakes to the motors M2 to M6 of the second axis L2 to the sixth axis L6 that are other rotary joint axes. To lock (steps A1 and A2 in FIG. 3). In this state, the controller 3 waits for 1 ms as an arbitrary time, for example, as a predetermined time, and enables the direct teaching operation around the first axis L1 (step A3). The brake release operation for one rotary joint axis is used as a brake release step.

  In the above operation, since the turning movement around the first axis L1 is free from the state shown in FIG. 4A, for example, as shown in FIG. A portion above the shoulder portion 6 can be rotated in a horizontal plane. For convenience of explanation, it is shown that the robot 2 has been moved to the position shown in FIG. 5B in order to make the movement easy to understand. However, since the actual opening time is 1 ms, The amount of movement is small.

  In the above operation, releasing the brake of the motor M1 and setting it to the free state means that in the drive circuit 15 that drives the motor M1, the energization path is controlled to be in an open state so that self-braking is not applied to the rotation of the motor M1. It means to control. In addition, the dynamic brake applied to the motor M1 and the locked state means that the drive circuit 15 that drives the motor M1 is controlled so that the current path is short-circuited and self-braking is applied to the rotational operation of the motor M1.

  Next, the controller 3 releases the brake of the motor M2 of the second axis L2 to make it free, and at the same time, the motors M1, M1 of the first axis L1, the third axis L3 to the sixth axis L6, which are other rotary joint axes. A brake is applied to M3 to M6 (steps A4 and A5), and, in the same manner as described above, an arbitrary time (same time, for example, 1 ms) is awaited in this state (step A6). In this state, since it is a state of a brake release step in which the turning around the second axis L2 is free, the operator can perform the direct teaching by turning the arm 7, that is, the lower arm 7 in the vertical direction.

  In the same manner, the controller 3 releases the motors M3 to M6 sequentially for each of the third and subsequent shafts L3 and thereafter to free the state, and applies the dynamic brake to the other rotary joint shafts. Control is performed so that only one axis is in a free state (steps A7 to A18). As described above, since the first axis L1 to the sixth axis L6 are sequentially released in units of a short time (1 msec), direct teaching work can be performed at predetermined intervals for all the rotary joint axes. The operation of performing this series of brake release steps for all axes is defined as a brake release cycle.

  If the position of the robot 2 cannot be moved from the start position (FIG. 4 (a)) to the target teaching position (FIG. 4 (m)) in one brake release cycle (NO in step A19). In the same manner, the brake release cycle is repeatedly executed (steps A1 to A18). When the position of the robot 2 can be moved to the teaching position and the end of the direct teaching operation is input from the teaching pendant 4 (in step A19) YES), the controller 3 ends the direct teaching mode.

  When the position of the robot 2 by direct teaching is set to the teaching position shown in FIG. 4 (m) (or FIG. 6 (m)), the variable data of this position is taken in by the teaching pendant 4, thereby Can be incorporated as data.

  According to the first embodiment as described above, when direct teaching is performed, the controller 3 fixes the six rotational joint axes from the first axis L1 to the sixth axis L6 one by one in a predetermined order. Since the control is performed so that the brake is released every time, it is possible to prevent only a specific rotating joint shaft from dropping at a stretch, and to improve the safety in teaching work.

  Further, since the rotary joint axes to be released from the brake are sequentially switched at regular intervals, in other words, all the axes can be moved at regular intervals in sequence, and the hand of the robot 2 in the teaching operation as in direct teaching is performed. In the operation in which the movement of the 10 parts is relatively slow, the operator can operate with the same feeling as when all the axes are opened at the same time. It can be almost equivalent to sex.

  In the above embodiment, the case where 1 ms is set as the fixed time (arbitrary time) is shown. However, in consideration of the safety of the worker who performs the direct teaching work, the following principle is applied for the fixed time. ing.

  That is, when the brakes of the motors M1 to M6 of the rotary joint shafts L1 to L6 are sequentially released, the upper limit is the time during which the safety of the worker can be ensured even if the released motors run away. Assuming a range where the speed at which the tip of the robot 2 moves increases so that no accident occurs even if it collides with an operator when the brake-released motor becomes in an abnormal rotation state is doing.

  In this case, although depending on the type and specifications of the robot 2, if the dangerous speed of movement of the hand of the robot 2 is, for example, 250 mm / s or more, the time until the critical speed is reached immediately after the brake is released is assumed. And can be set as a certain time. Further, here, a time during which the critical speed is not reached is set even when the end effector is not attached to the hand of the robot 2, that is, when the tip of the robot 2 moves fastest in the lightest state. For this reason, in actuality, it may still take some time to reach the critical speed when the end effector is attached.

  If a certain time is calculated for each rotary joint axis in accordance with the above principle, it is expected that a different time is obtained as a certain time for each axis. Therefore, different fixed times can be set for each rotary joint axis, and the shortest time among them can be set as a common fixed time. In this respect, it is preferable to set the fixed time in consideration of the responsiveness at the time of direct teaching.

  In the above embodiment, the end of direct teaching is determined in units of a brake release cycle. However, the end of direct teaching may be determined for each brake release step.

(Second Embodiment; Claim 8)
FIG. 7 shows a second embodiment of the present invention. Hereinafter, parts different from the first embodiment will be described. In this embodiment, the number of rotary joint axes for releasing the brake is increased to 2 axes under conditions that do not hinder the operation, and the degree of freedom in the position and orientation work during direct teaching is increased. It is a thing.

  FIG. 7 is a diagram showing a flow of control by the controller 3 when performing direct teaching. First, the controller 3 sets the release axis 1 variable to “1” (step B1), and then controls all the axes L1 to L6 to the locked state (step B2). In this case, locking of each axis is controlled by applying a dynamic brake (DB) to the motors M1 to M6 that drive the axes L1 to L6.

  Next, forward conversion is performed on the axes L1 to L6 to calculate the positions of the arms 6 to 11 (step B3). Subsequently, the controller 3 sets a value obtained by adding “1” to the value of the release axis 1 variable, that is, “2” as a loop count value for setting the release target axis (step B4). The approach of the release axis 1 variable axis, that is, the first axis L1 is calculated (step B5). Next, an axis of the loop count value “2”, that is, the approach of the second axis L2 here is calculated as the loop count approach (step B6).

  Subsequently, the controller 3 calculates the angles of the vectors of both the release axis 1 approach and the loop count approach calculated in steps B5 and B6 (step B7). That is, here, the intersection angle θ between the first axis L1 and the second axis L2 is calculated. Then, it is determined whether or not the crossing angle θ between the vectors obtained as a calculation result satisfies the following expression (1) (step B8).

(90 ° −α) ≦ | θ | ≦ (90 ° + α) (1)
This is to determine whether or not the crossing angle θ between the vectors is “corresponding to orthogonality” corresponding to a state in which the intersecting angle θ is orthogonal (= 90 °) within the allowable angle range (± α). Here, if the allowable angle range (± α) is, for example, α = 45 °, whether or not the range indicated by the equation (1) specifically satisfies one of the following equations (1a) and (1b): Judgment will be made.

45 ° ≦ θ ≦ 135 ° (1a)
−45 ° ≧ θ ≧ −135 ° (1b)
The above-mentioned “equivalent to orthogonal” refers to a range of ± 45 ° with respect to an axis corresponding to one variable of the release axis, centered on an angle orthogonal to this (ie, forming 90 °). When the arm is rotated, only the rotation around the target axis can be independently performed, and the range of the angle of intersection with the axis that is not rotated in conjunction with the arm is shown. Therefore, as the intersection angle θ between the two axes, the allowable range around the orthogonal angle of 90 ° is not necessarily ± 45 °, and a narrower allowable range can be set.

  In addition, narrowing the allowable range described above will tighten the criteria for determining the simultaneous release target axis to release the brake at the same time when direct teaching is performed, but this increases the safety during direct teaching. It is in a trade-off relationship with increasing, and can be set to a desired range by the user.

  Next, when the condition of the above-described formula (1a) or (1b) is not satisfied, the controller 3 sets the loop count value to “6” or less (the loop count value is “2”). Therefore, YES in step B9), the loop count value is incremented by “1” (= “3”) (step B10), the process returns to step B6, and the axis of the loop count approach, that is, the third axis L3 is again set as the loop count approach. An approach is calculated (step B6).

  Hereinafter, similarly, the controller 3 calculates an angle θ formed by each vector of the first axis L1 and the third axis L3 (step B7), and determines whether or not it is “orthogonal equivalent” (step B8). ). In this way, the loop count is increased one after another to determine whether or not there is an “orthogonal equivalent” rotational joint axis, and if there is an axis that matches the condition (YES in step B8), the corresponding The brake of the loop count value axis to be released is released and controlled to a free state (step B11).

  Subsequently, the controller 3 releases the brake of the value of the release axis 1 variable, that is, the brake of the first axis L1, and controls it to the free state (step B12). When determining the above condition, if there is no axis that matches the condition even if the loop count value exceeds “6”, that is, if the condition is not met even if the sixth axis L6 is executed (step B9). In NO), only the axis of the value of the release axis 1 variable, that is, the first axis L1 is released (step B12).

  As described above, when either the two axes, the loop count value axis and the release axis 1 variable value axis, are released at the same time, or only the release axis 1 variable value axis is released. Also, in step B13, it is canceled only for an arbitrary time (for example, a fixed time of 1 ms). As a result, when the conditions are met, the two axes are simultaneously released, and when the conditions are not met, the brake is released for one axis and the arm related to the free axis can be moved during this period. it can. When the arbitrary time elapses, the controller 3 applies a brake to the free shaft and returns all the shafts to the locked state.

  Next, since the direct teaching operation has not been completed yet (NO in step B14), the controller 3 determines whether or not the value of the release axis 1 variable is “6” (NO in step B15). ), The value of the release axis 1 variable is incremented by “1” to become “2” (step B16). Since the value of the incremented release axis 1 variable is not yet “6” (NO in step B17), the process returns to step B2 and the same process as above is performed for the second axis L2, which is the axis of the release axis 1 variable value. carry out. Specifically, steps B2 to B14 are repeatedly performed.

  When steps B2 to B14 are repeated, the axes (the simultaneous release target axes) to be determined whether or not the brake can be released simultaneously with the second axis L2 are the third axis L3 to the sixth axis L6 (actually In the determination, the answer to YES in step C is the fourth axis L4). Similarly, when the axis of the release axis 1 variable is the third axis L3, the simultaneous release target axes are the fourth axis L4 to the sixth axis L6, and in the fourth axis L4, the simultaneous release target axis is the fifth axis L5. In the sixth axis L6 and the fifth axis L5, the simultaneous release target axis is only the sixth axis L6.

  Further, when the release axis 1 variable is “6”, when the increment is made from “5” to “6” in step B16, the determination in step B17 is YES, so the above steps are not repeated directly. Proceeding to step B12, only the sixth axis L6 is released and the free state is set.

  As a result, when there is an axis that matches the condition when releasing the brake of the second axis L2, the brake of that axis is also released at the same time, and when there is no axis, the brake is released only for the second axis L2. In this manner, when the brake release operation is sequentially performed for all the rotary joint axes L1 to L6, one brake release cycle is completed. At this time, if the direct teaching operation is completed and the position of the robot 2 is set to the teaching position (the state shown in FIG. 4 (m)), the operation ends (step B14). Perform the cycle.

  In this case, since the value of the release axis 1 variable is “6” (YES in step B15), the controller 3 sets the value of the release axis 1 variable to “1” again, and returns to step B2. The above steps B2 to B17 are repeatedly executed. When the end of the direct teaching work is input from the teaching pendant 4, the controller 3 determines YES in step B14 and ends the control program.

  According to such a second embodiment, in addition to the effects of the first embodiment, the brake can be released for two axes at the same time, so the degree of freedom of movement of the arm can be increased during direct teaching work. . Even in this case, since the axis of the release axis and the axis of which the brake is released at the same time satisfy the condition of “orthogonal equivalent”, the rotation around one axis causes the rotation around the other axis. It is possible to suppress the movement in conjunction with the movement, and thereby the movement can be performed without hindering the teaching operations.

  Also in the second embodiment, the end of direct teaching is determined in units of a brake release cycle, but the end of direct teaching may be determined for each brake release step.

(Third embodiment; claim 9)
FIG. 8 shows a third embodiment of the present invention. Hereinafter, parts different from the second embodiment will be described. In the third embodiment, with respect to the axis of the value of the release axis 1 variable, it is determined whether or not all other axes are “corresponding orthogonally” and the brake is released, as in the second embodiment. It is a different place.

  In the second embodiment, for the axis of the release axis 1 variable value, the loop count was incremented by “1” as an object to simultaneously release the brake, whereas it ended when the value reached “6”. Then, after reaching “6”, the process returns to “1” to determine the value of the release axis 1 variable.

  That is, as shown in FIG. 8, the controller 3 sets the release axis 1 variable to “1” (step C1) and controls all the axes L1 to L6 to the locked state (step C2). Next, forward conversion is performed for the axes L1 to L6 to calculate the positions of the arms 6 to 11 (step C3). Subsequently, the controller 3 sets “1” as the loop count value (step C4), and sets the value of the release axis candidate variable to a value obtained by adding “1” to the release axis 1 variable, that is, “2” (step 2). C5). The release axis candidate variable is a variable for setting the simultaneous release target axis.

  Next, the controller 3 calculates the release axis 1 variable axis, that is, the approach of the first axis L1 as the release axis 1 approach (step C6). Subsequently, the axis of the release axis candidate variable, that is, the approach of the second axis L2 is calculated as the release axis candidate approach (step C7).

  Subsequently, the controller 3 calculates the angles of the vectors of both the release axis 1 approach and the release axis candidate approach calculated in steps C6 and C7 (step C8). In the same manner as described above, here, the intersection angle θ between the first axis L1 and the second axis L2 is calculated. Then, it is determined whether or not the crossing angle θ between the vectors obtained as a calculation result is “corresponding to orthogonality”, that is, whether or not the above equation (1) is satisfied (step C9).

  Hereinafter, the controller 3 sequentially determines whether or not there is a release axis candidate that satisfies Expression (1). First, when the loop count value is “6” or less (YES in Step C10), the value of the release axis candidate variable is set to “6” until the value of the release axis candidate variable reaches “6” (NO in Step C11). 1 "is incremented (step C12) and the loop count value is incremented by" 1 "(step C13), the process returns to step C7, and the above steps C7 to C13 are repeatedly executed.

  Thereafter, if the value of the release axis candidate variable reaches “6” (NO in step C9) without determining the “joint orthogonal” rotational joint axis (NO in step C9), the value of the release axis candidate variable Is set to “1” (step C14), and it is determined through step C13 whether or not there is an “orthogonal equivalent” axis. Thus, it is determined whether or not there is an “orthogonal equivalent” axis even for an axis having a value smaller than the value of the release axis 1 variable.

  As a result of the above determination, if there is an “orthogonal equivalent” axis (YES in step C9), the controller 3 releases the brake of the axis of the release axis candidate variable that is the “orthogonal equivalent” axis. Then, the free state is set (step C15), and then the brake of the value of the release axis 1 variable is released and the free state is set (step C16). If the result of determination is that there is no "corresponding orthogonal" axis (NO in step C10), only the axis corresponding to the value of the release axis 1 variable is released to a free state (step C16).

  As a result, it is determined whether there is an “orthogonal equivalent” axis for the lower value axis even if there is no “orthogonal equivalent” axis for the higher value axis for the release axis 1 variable value axis. As a result, the probability that an axis that releases the brake at the same time is generated increases.

  The controller 3 waits for an arbitrary time (for example, a fixed time of 1 ms) in a state where the axis corresponding to the release axis 1 variable and the axis corresponding to “orthogonal” are freed (step C17), Return all axes to the locked state. When the brake release step for determining the axis corresponding to “orthogonal” for the first axis L1 that is the axis of the release axis 1 variable and the brake release is completed as described above, whether or not the direct teaching operation is completed. (Step C18), if NO, then the release axis 1 variable is incremented by "1" (step C20), the process returns to step C2 again, and the brake release steps of steps C2 to C17 described above are repeated. Execute.

  In this way, the operation until the value of the release axis 1 variable becomes “6” is executed as one brake release cycle, and whether or not there is an “orthogonal equivalent” axis sequentially for each of the axes L1 to L6. If it exists, the brake is released at the same time so that the teaching operation of the robot 2 can be performed.

  If the direct teaching operation is not completed when the brake release cycle is completed (NO in step C18), the value of the release axis 1 variable is “6” (YES in step C19). ), The controller 3 resets the release axis 1 variable to “1” (step C21), sets the release axis candidate variable to “0” (step C22), returns to step C2, and returns to the brake described above. Repeat the release step. Thereafter, when an input indicating that the direct teaching work has been completed is input from the teaching pendant 4 or the like, the controller 3 terminates the direct teaching control operation (YES in step C18).

  According to the third embodiment, the same effects as those of the first and second embodiments can be obtained, and the axis corresponding to the orthogonal value of the higher value axis of the release axis 1 variable value axis can be obtained. Even if it does not exist, it is determined whether or not there is an orthogonal equivalent axis with respect to the lower value axis, so the probability that an axis that releases the brake at the same time will increase, and the degree of freedom of movement of the arm in direct teaching will be increased. It becomes high and workability can be improved.

In the above embodiment, the end of the direct teaching work is determined for each brake release step, but it can also be determined for each brake release cycle.
(Fourth embodiment; claims 1, 2 and 3)
9 to 12 show a fourth embodiment of the present invention. Hereinafter, parts different from the first embodiment will be described. In this embodiment, as in the first embodiment, the brake release step (the predetermined axis to be released is the motor's opening position) in a preset order for a plurality of axes, that is, the first axis L1 to the sixth axis L6. In a state where the brake is released, a step of executing a brake release state in which the other remaining shafts remain in a state where the motor is braked is executed for a predetermined time. As in the first embodiment, the brake release step is repeatedly executed with one cycle (one brake release cycle) in which the preset order is rotated once. The time required for one cycle is fixed and fixed, and is set to 60 ms in this embodiment.

  Within one cycle, the brakes of the motors are released in order for each axis. This brake release time (hereinafter referred to as the release time) is a fixed time (uniformly determined for all axes in the first embodiment). However, in the present embodiment, the axis having a larger amount of movement in the previous cycle is configured to be longer in proportion to the amount of movement so that the time is longer.

  In the case of this embodiment, the opening time Tk (n) of each axis is set to a fixed opening time Tkf (1 ms in this embodiment) that is fixedly secured, and a variable opening time Tkv (n) that changes according to the amount of movement. It is set as the added time. The time required for one cycle is 60 ms, of which the fixed opening time Tkf is 6 ms for the entire six axes, so the variable opening time Tkv (n) is 54 ms in total, and this 54 ms is distributed to each axis.

  The variable opening time Tkv (n) of each axis is based on the movement amount in the previous cycle (n−1) of each axis at a reference time Tkv (0) (9 ms in this embodiment) set in advance as an initial value. It is calculated by multiplying the coefficient of change Ck (n-1) obtained in this way. The coefficient of change for each axis is determined as follows. First, the movement amount Mk (n-1) of each axis and the average movement amount Mave (n-1) of all axes in the previous cycle are calculated, and the average movement amount Mave is calculated from the movement amount Mk (n-1) of each axis. The value obtained by subtracting (n-1) is defined as the deviation movement amount ΔMk (n-1) of each axis. Next, the absolute value of the value obtained by adding the deviation movement amount ΔMk (n−1) of each axis having the same sign (plus or minus) is calculated as the distributed denominator coefficient D. Then, “1” is added to the value obtained by dividing the deviation movement amount ΔMk (n−1) by the distribution denominator coefficient D, and the value is calculated as the change coefficient Ck (n−1). As described above, the variable opening time reference time Tkv (0) (9 ms) multiplied by the change coefficient Ck (n-1) is multiplied by the variable opening time Tkv (n) of each axis in the next period (n). (The first variable opening time setting process).

  Therefore, the variable opening time of each axis has a portion (reference time × deviation moving amount / distributed denominator coefficient) that is directly proportional to the deviation of the movement amount in the previous cycle, and is obtained as the sum of this variable opening time and the fixed opening time. The open time to be set is set to a longer time as the movement amount of each axis in the previous cycle is larger. Here, the amount of movement of each axis is expressed as a rotation angle, and is expressed as a cumulative total of the rotated angles. For example, when it rotates 2 ° in one direction and 2 ° in the opposite direction, the movement amount is expressed as 4 °.

  When the direct teaching mode is set by the operator, the controller 3 executes the direct teaching control program shown in FIG. That is, in order to execute the first brake release cycle, the controller 3 first sets the fixed release time and variable release time of the brakes of each axis as initial values to preset reference times (FIG. 9). “NO” in step D1, step D2). Here, since the fixed opening time Tkf of each axis is fixed and set to 1 ms, the reference time of the fixed opening time is 1 ms, and the reference time Tkv (0) of the variable opening time is 9 ms as described above. Therefore, the opening time Tk (n) given to each axis is 10 ms. The reference time for the variable opening time is set as a value obtained by evenly distributing the entire variable opening time (54 ms) in one cycle to each of the six axes.

  When the release time Tk (n) of each axis is set, the controller 3 starts to execute the first brake release cycle. As shown in FIG. 10, this brake release cycle is performed by performing a brake release step in the order of the axis numbers from the first axis L1 to the sixth axis L6. This is the same as in the first embodiment, but the difference from the first embodiment is that the current angle of the axis is detected before entering the brake release step of each axis and stored in, for example, the ROM 13 (first Step E1 for one axis, the direction of rotation during brake release, and the rotation angle at that time in the reverse direction is stored in ROM 13 (step E4 for the first axis), and the current angle of the axis after the brake release step is completed Is detected and stored in the ROM 13 (step E5 for the first axis). Then, the cumulative rotation angle of each axis during this brake release step is calculated, and this cumulative value is stored in the ROM 13 as the movement amount of each axis (step D4 in FIG. 9).

  When the first brake release cycle is completed, the controller 3 sets the release time Tk (2) of each axis in the next second brake release cycle according to the movement amount of each axis in the previous period. Calculation to set the time is performed ("NO" in step D5 in FIG. 9, "YES" in step D1, step D3). A control program for obtaining the release time of each axis in the next brake release cycle is shown in FIG.

  In other words, the calculation of the opening time Tk (n) of each axis starts with obtaining the average movement amount Mave (n-1) of all axes in the previous period (n-1) (step F1). Next, the controller 3 subtracts the average movement amount Mave (n-1) from the movement amount Mk (n-1) of each axis in the previous period (n-1), and calculates the value as the deviation movement amount ΔMk (n- 1) (step F2). Subsequently, the controller 3 obtains the sum of deviation movement amounts ΔMk (n−1) having the same sign among the deviation movement amounts ΔMk (n−1) of the respective axes, for example, the sign is plus (+), and this is calculated as the distribution denominator coefficient D (Step F3). Even if the sum of the minus (−) deviation movement amounts is obtained and the absolute value thereof is taken, the distribution denominator coefficient D having the same value can be obtained. Thereafter, the controller 3 adds “1” to a value obtained by dividing the deviation movement amount ΔMk (n−1) by the distribution denominator coefficient D, and sets this as a change coefficient Ck (n−1) (step F4). Then, the controller 3 obtains the variable opening time Tkv (n) of each axis by multiplying the change coefficient Ck (n-1) of each axis by the reference time Tkv (0) of the variable opening time (step F5), and finally. For each axis, the fixed opening time Tkf and the variable opening time Tkv (n) are added, and this is set as the opening time of each axis (step F6). Thus, the opening time of each axis in the next cycle is calculated. Exit.

  When the release time of each axis in the next cycle is obtained in this way, the controller 3 executes a brake release cycle in the next cycle (step D4 in FIG. 9). Also during execution of this brake release cycle, the movement amount of each axis is obtained in the same manner as described above, and the release time of each axis in the next brake release cycle is obtained based on this movement amount. In this way, the release time of each axis is obtained based on the movement amount of each axis in the previous cycle, and the brake release cycle is continuously executed until the robot tip moves to the teaching position.

  FIG. 12 shows the amount of movement of each axis and the release time of each axis in the next period for each period from the first to the third brake release cycle. In FIG. 12, the first axis L1 to the sixth axis L6 are indicated by J1 to J6 (FIGS. 14, 15, and 17 are the same). With reference to FIG. 12, for example, with respect to the first axis L1 and the fifth axis L5, an example of calculating the opening time in the second cycle will be described. First, the total movement of the first axis L1 to the sixth axis L6 in the first time Since the amount is 3 ° and the average movement amount Mave (1) is 0.5 °, the deviation movement amount ΔMk (1) is +1.5 for the first axis L1, +0.5 for the second axis L2, and third. The axis L3 to the sixth axis L6 are each -0.5, and the distributed denominator coefficient D is a value 2 that is the sum of the deviation movement amounts ΔMk (1) of the first axis L1 and the second axis L2 having a plus sign. The change coefficient Ck (1) of each axis is obtained by adding “1” to the value obtained by dividing the deviation movement amount ΔMk (1) of each axis by the distributed denominator coefficient D. Therefore, the change of the first axis L1. The coefficient Ck (1) is [(1.5 / 2) +1] = 1.75, and the change coefficient Ck (1) of the fifth axis L5 is [(−0.5 / 2) +1] = 0.75. . Since the variable opening time Tkv (2) of each axis is obtained by multiplying the reference time Tkv (0) of the variable opening time by the change coefficient Ck1, the variable opening time Tkv (2) of the first axis L1 is (9 ms × 1.75) = 15.75 ms, and the variable opening time Tkv (2) of the fifth axis L5 is (9 ms × 0.75) = 6.75 ms. Since the opening time is the sum of the fixed opening time Tkf and the variable opening time Tkv (n), the opening time of the first axis L1 is (1 ms + 15.75 ms) = 16.75 ms, and the opening time of the fifth axis L5 is (1 ms + 6). .75 ms) = 7.75 ms. In this way, the brake release time for each axis in the next one cycle is obtained.

  As described above, according to the present embodiment, the brake release time (release time) of each axis is set as a time obtained by adding the fixed release time and the variable release time that changes according to the movement amount of the previous cycle. Even for an axis with a small number of shafts, only a fixed release time is secured as the release time, and a brake release state in which each axis can always move every cycle is secured by this fixed release time.

Also, for axes with a large amount of movement in the previous cycle, if the amount of movement in the previous cycle is large, the longer variable opening time is set so that movement can be prioritized in the next cycle. The opening time can reflect the situation. In particular, in this embodiment, the variable opening time is limited to 54 ms in total within 60 ms in one cycle, and the distribution denominator coefficient is calculated and distributed for setting the variable opening time of each axis. I try to do it. This distributed denominator coefficient is the value obtained by adding the deviation movement amount of each axis obtained by subtracting the average movement amount from the movement amount of each axis, so that the actual movement amount of each axis is several percent. It will be evaluated after being compressed to an increase of tens of percent. For this reason, it is possible to perform an appropriate time distribution for all the axes while suppressing a large amount of time from being distributed all at once to an axis having a large amount of movement in the previous cycle. For example, as shown in FIG. 12, the movement amount of the previous cycle is “0”, for example, the fifth axis L5 has the movement amount of “0” for the first time, the second time, and the third time. However, even on that axis, the variable opening time is not “0” in the second, third, and fourth times of the next cycle, but 6.75 ms (7.75 ms including the fixed opening time) That time is secured.
In this way, appropriate time distribution is performed for all axes while preventing a lot of time from being distributed all at once to an axis with a large amount of movement in the previous cycle, so that the hand is moved close to the teaching position. After that, when you want to move each axis finely, such as when fine-tuning the position of the hand, the brake release time for each axis is appropriately secured, and each axis can easily move. Excellent operability.

(Fifth embodiment; claims 1, 2, 4, 6)
13 and 14 show a fifth embodiment of the present invention. In this embodiment, when obtaining the variable opening time of each axis, the average movement amount Mave (n−1) of the previous cycle is obtained, and the deviation movement amount ΔMave (n−1) is obtained by subtracting the average movement amount from the movement amount of each axis. ) Is the same as in the fourth embodiment. Differences from the fourth embodiment are as follows.
In other words, after calculating the deviation movement amount ΔMave (n−1), the minimum deviation movement amount ΔMave (n−1) is divided by a predetermined reference movement amount, for example, “1 °”, and the value is increased. Rate. In the case of this embodiment, since the reference movement amount is 1 °, the value of the deviation movement amount becomes the increase rate as it is.

  Then, when the minimum value of the increase rate of each axis is “−1” or more, a value obtained by adding “1” to the increase rate of each axis is defined as a change coefficient Ck (n−1) of each axis. Further, when there is an increase rate of less than “−1” in each axis, the reciprocal of the absolute value of the smallest increase rate Ms (n−1) is calculated as the compression coefficient P (n−1), and each axis A change coefficient Ck (n−1) is obtained by multiplying the rate of increase by a compression coefficient and adding “1”. After obtaining the change coefficient as described above, the variable open time of the current cycle is set by multiplying the reference time Tkv (0) of the variable open time by the change coefficient Ck (n-1) (referred to above as the second open time). Variable opening time setting process).

  As described above, the contents of the control for setting the release time of the nth cycle based on the movement amount of each axis in the previous cycle will be described with reference to FIG. When the brake release cycle of the previous period (n-1) is completed, the controller 3 obtains the average movement amount Mave (n-1) of all the axes in the previous period (step G1), and then the deviation movement amount ΔMk of each axis. (n-1) is calculated (step G2). Then, the controller 3 takes the deviation movement amount ΔMk (n−1) as an increase rate as it is (step 3) and searches for the smallest increase rate ΔMk (n−1) (step G4). It is determined whether or not the increase rate Ms (n-1) is less than "-1" which is a preset threshold value (step G5).

  When the minimum increase rate is “−1” or more (“NO” in step G5), the controller 3 adds “1” to the increase rate ΔMk (n−1) of each axis, and the value is added to each axis. Change coefficient Ck (n-1) (step G6). Then, the controller 3 calculates the variable opening time Tkv (n) of each axis by multiplying the reference time Tkv (0) of the variable opening time by the change coefficient Ck (n-1) of each axis (step G7) and fixed. The variable opening time Tkv (n) for each axis is added to the opening time Tkf to obtain the opening time Tk (n) for each axis (step G11).

  When the minimum increase rate is less than “−1” (“YES” in step G5), the controller 3 uses the reciprocal of the absolute value of the minimum increase rate Ms (n−1) as the compression coefficient P (n−1). (Step G8), the increase rate ΔMk (n-1) of each axis is multiplied by the compression coefficient P (n-1), and a value obtained by adding "1" to the product is obtained as the change coefficient Ck (n-1) of each axis. ) (Step G9). Then, the controller 3 calculates the variable opening time Tkv (n) of each axis by multiplying the reference time Tkv (0) of the variable opening time by the change coefficient Ck (n-1) of each axis (step G10) and fixed. The opening time Tkf (n) is added to the opening time Tkf and the variable opening time Tkv (n) of each axis (step G11).

FIG. 14 shows the movement amount of each axis and the opening time of each axis in the next period when the minimum value of the increase rate is “−1” or more for each period from the first to the third of the brake release cycle. It is a thing. In this case, the concept of the compression coefficient need not be taken into account, but FIG. 14 shows the compression coefficient “1” in relation to FIG. 15 described later.
With reference to FIG. 14, for example, with respect to the first axis L1 and the fifth axis L5, an example of calculating the opening time in the second cycle will be described. First, the first axis L1 to the first axis in the first cycle that is the previous cycle. Since the total movement amount of the six axes L6 is 3 °, the average movement amount Mave is 0.5 °. The deviation movement amount ΔMk (1) is +1.5 for the first axis L1, +0.5 for the second axis L2, and -0.5 for the third axis L3 to the sixth axis L6, respectively. Since the deviation movement amount ΔMk (1) of each axis becomes the increase rate of each axis as it is, the minimum of the increase rates is “−0.5”, which is “−1” or more. Therefore, since the change coefficient Ck1 of each axis is obtained by adding “1” to the deviation movement amount ΔMk (1) of each axis, the change coefficient Ck (1) of the first axis L1 is (1.5 + 1) = 2. .5, the change coefficient Ck (1) of the fifth axis L5 is (−0.5 + 1) = 0.5. Since the variable opening time Tkv2 of each axis is obtained by multiplying the reference time Tkv (0) of the variable opening time by the change coefficient Ck1, the variable opening time Tkv (2) of the first axis L1 is (9 ms × 2.5). ) = 22.5 ms, and the variable opening time Tkv (2) of the fifth axis L5 is (9 ms × 0.5) = 4.5 ms. Since the opening time is the sum of the fixed opening time Tkf and the variable opening time Tkv (n), the opening time of the first axis L1 is (1 ms + 22.5 ms) = 23.5 ms, and the opening time of the fifth axis L5 is (1 ms + 4). .5 ms) = 5.5 ms.

  FIG. 15 shows the movement amount of each axis when the minimum value of the increase rate is less than “−1” at the second time and the third time in each period from the first to the third brake release cycle. And the opening time of each axis in the next cycle. With reference to FIG. 15, for example, with respect to the first axis L1 and the fifth axis L5, the calculation example of the third opening time will be described. First, the total movement of the axes L1 to L6 in the second time which is the previous cycle. Since the amount is 6.6 ° and the average movement amount Mave (2) is 1.1 °, the deviation movement amount ΔMk (2) is +2.7 for the first axis L1, 1.7 for the second axis L2. The third axis L3 to the sixth axis L6 are each -1.1. Since the deviation movement amount ΔMk2 of each axis becomes the increase rate of each axis as it is, the minimum ΔMs (2) among the increase rates is “−1.1”, which is less than “−1”. When the minimum increase rate is less than “−1”, the reciprocal 1 / −1.1 = 0.090909 of the absolute value of the minimum increase rate is the compression coefficient P2. Since the change coefficient of each axis is obtained by multiplying the increase rate of each axis by the compression coefficient and adding “1”, the change coefficient Ck (2) of the first axis L1 is 2.7 × 0. 909091 + 1 = 3.4454547, and the coefficient of change Ck (2) of the fifth axis L5 is −1.1 × 0.909091 + 1 = 0. Since the variable opening time Tkv (3) is obtained by multiplying the reference time of the variable opening time by the change coefficient Ck (2), the variable opening time Tkv (3) of the first axis L1 is (9 ms × 3. 454547) = 31.090923 ms, and the variable opening time Tkv (3) of the fifth axis L5 is “0”. Since the opening time Tk (3) is the sum of the fixed opening time Tkf and the variable opening time Tkv (n), the opening time Tk (3) of the first axis L1 is (1 ms + 31.090923 ms) = 32.090923 ms, the fifth The opening time Tk (3) of the axis L5 is (1 ms + 0 ms) = 1 ms.

  Thus, according to the present embodiment, when the minimum increase rate of each axis is “−1” or more, “1” is added to the increase rate (same as the deviation movement amount) of each axis. Since the variable coefficient is the change coefficient and the variable coefficient multiplied by the reference time is the variable opening time, unlike in the case of the fourth embodiment, when setting the variable opening time of each axis, For those having a relatively large amount of movement in the previous period, it can be set so as to have a long open time. In addition, when setting the variable opening time, the time to be added to the reference time can be distributed much in direct proportion to the deviation movement amount of the corresponding axis. In this case, the direct teaching workability when the movement amount is large can be prioritized. For this reason, if you want to move your hand greatly toward the teaching position at the beginning of direct teaching, you can allocate a lot of variable opening time to the axis that moves greatly according to the movement. It is possible to move the hand to the vicinity of the teaching position easily by moving the position greatly, and the operability is excellent.

  Here, if the deviation movement amount of each axis is used as the increase rate as it is, the deviation movement amount of the axis having a very small movement amount may become small and the increase rate of the axis may be less than “−1”. Then, the sign of the change coefficient becomes negative. As a result, the calculated value of the variable opening time becomes negative, and the opening time calculated as the sum of the fixed opening time and the variable opening time is the fixed opening time (minimum to be secured ( 1 ms), and the opening time may be “0”.

  In order to prevent this from occurring, in the present embodiment, when the minimum increase rate of each axis is less than “−1”, when calculating the change coefficient of each axis, The reciprocal of the absolute value of the minimum deviation movement amount is defined as a compression coefficient, and “1” is added after the increase rate of each axis is multiplied by the compression coefficient. For this reason, when the deviation movement amount is relatively small for an axis with a very small movement amount relative to an axis with a large movement amount and the increase rate is less than “−1”, the compression coefficient is set, Prevents a lot of variable opening time from being allocated to an axis with a large amount of movement. Therefore, under the circumstances where one cycle is set to a fixed time, a minimum fixed opening time ( Only 1 ms) can be secured. Thereby, even if it is an axis | shaft with very small moving amount | distance, variable open time can be appropriately allocated to an axis | shaft with much moving amount | distance, suppressing that it becomes impossible to move.

(Sixth embodiment; claims 1, 2, 3, 5, 6)
FIGS. 16 to 18 show a sixth embodiment of the present invention. The difference from the fourth embodiment is that a specific axis of each axis, for example, a main axis (basic axis) is calculated as a variable opening time calculation. ) When the movement amount of the first axis L1 exceeds a preset threshold value, the second variable opening time setting process of the fifth embodiment is executed.

  When the direct teaching mode is set by the operator, the controller 3 executes the direct teaching control program shown in FIG. When the control program is executed, first, the controller 3 selects the first variable opening time setting process for setting the variable opening time (step H1). Therefore, in the first brake release cycle, as in the case of the fourth embodiment, the release time given to each axis is set to 10 ms.

  Thereafter, the controller 3 enters the execution of the first brake release cycle, calculates the total rotation angle of each axis, and stores it in the ROM 13 as a storage means (step H2). When the first brake release cycle is completed, the controller 3 reads the movement amount of each axis (step H3), of which the movement amount of the first axis L1 as the main axis is a predetermined threshold value, for example, It is determined whether or not 3 ° is exceeded (step H4).

  If the threshold value is not exceeded (“NO” in step H4), the controller 3 calculates the variable opening time by the first variable opening time setting process, adds 1 ms to this variable opening time, The release time is set (step H5), and the second brake release cycle is executed ("NO" in step H7, step H2).

  If the threshold value is exceeded (“YES” in step H4), the controller 3 calculates the variable opening time by the second variable opening time setting process, and adds 1 ms as the fixed opening time to this variable opening time. Addition is performed to set the release time of each axis (step H6), and the second brake release cycle is executed ("NO" in step H7, step H2).

  Thereafter, as described above, the first variable opening time setting process or the second variable opening time setting process is selected depending on whether or not the movement amount of the first axis L1 in the previous cycle exceeds the threshold value. To calculate the variable release time and execute the brake release cycle.

  FIG. 17 shows a specific example in which the opening time of each axis is obtained by calculating the variable opening time by the first variable opening time setting process or the second variable opening time setting process according to the movement amount of the first axis L1. . In FIG. 17, the first and second brake release cycles are based on the first variable opening time setting process, the third and fourth times are based on the second variable opening time setting process, the fifth time Indicates the opening time of the first variable opening time setting process.

FIG. 18 is a graph showing changes in the opening time of each axis from the first time to the fifth time calculated as described above.
As described above, in the present embodiment, the first variable opening time setting process or the second variable opening time setting process is selected depending on whether or not the movement amount of the first axis L1 in the previous cycle exceeds the threshold value. To calculate the variable opening time. The first axis L1 moves greatly when the robot 2 is moved from the start position toward the teaching position during teaching, but moves only slightly when the hand is moved close to the teaching position for fine adjustment. Therefore, according to the present embodiment, when it is desired to move the robot 2 greatly, for the first axis L1 that has moved greatly, for example, the opening time of the next cycle is set to a long time necessary for a large movement (second) In the case of fine adjustment of the position of the hand, for example, the movement of the first axis L1 is reduced and the opening time suitable for all axes is different from the case where the robot 2 is moved largely. (First opening time setting process), and in general, each axis can easily operate in the entire direct teaching, and the operability is excellent.

(Other embodiments)
The present invention is not limited to the embodiments described above and shown in the drawings, and can be expanded or modified as follows.

  In the embodiment, brakes are sequentially released from the first axis L1 for all six rotary joint axes. However, depending on the type of end effector attached to the flange 11, the flange 11 may be released even if the brake is always released for the sixth axis L6. Since there are cases where there is no drop or unexpected movement of the arm of the part, it is possible to simultaneously release the brakes for the motors of a plurality of axes with the first axis L1 to the fifth axis L5 as the target of the brake release step. The workability can be improved.

In the embodiment, a six-axis robot has been described, but the present invention is not limited to six axes.
In the second or third embodiment, an example is shown in which the brake release step is executed by releasing the brake of the release axis 1 variable value axis and the other one axis. Since three axes can be the target in three dimensions, this can be determined by calculation with the same conditions set, and the brake can be released simultaneously for the three axes. In this case, since there are three axes corresponding to “orthogonal equivalent”, it is possible to prevent the other free axes from rotating in conjunction with each other when rotating around one axis.

  Although the case where the brake release time in each brake release step is set to an arbitrary time, for example, 1 ms is shown, the present invention is not limited to this, and an appropriate time can be set. In this case, since the brakes are sequentially released from the six axes L1 to L6, for the operator, if the time is increased, the movement time per axis increases while the degree of freedom of the other axes decreases, If the time is shortened, the movement time per axis is reduced, but the user feels that direct teaching can be performed smoothly by switching all axes at a high speed. If an appropriate time is set in consideration of such properties, teaching can be performed in consideration of workability while taking into consideration the subjective feeling of the worker.

  In the second or third embodiment, the determination of the axis corresponding to “orthogonal” is set within the range of ± 45 ° centered on the equation (1), that is, the orthogonal angle of 90 °. The allowable range can be appropriately set as long as it is an intersection angle that can be determined as a possible range even if the brake is released simultaneously. In this case, if the permissible range is narrowed, it is limited to a range close to orthogonal, so that it is improved in terms of safety. If the permissible range is widened, the degree of freedom is increased and the direct teaching workability is improved.

  In addition, the allowable angles for the determination of “equivalent to orthogonal” can be set for each of the combinations of the first axis L1 to the sixth axis L6. This is because, depending on the shafts to be combined, the allowable range may be widened, and conversely, in consideration of safety, a narrower one may be better.

  The movement amount of each of the axes L1 to L6 may be expressed by the movement distance (unit: mm) of the tip of each arm 6 to 11. In this case, in the fifth embodiment, when the increase rate is obtained by dividing the minimum deviation movement amount ΔMave (n−1) by a predetermined reference movement amount, the reference movement amount is set to 100 mm, for example.

  The specific axis in the sixth embodiment is not limited to the first axis L1, but may be the second axis L2 or the third axis L3, which is another main axis (basic axis), and any one of the other axes L4 to L6. In short, what is necessary is just an axis that moves greatly with a large movement of the hand when the position of the hand is largely changed.

In the second embodiment, the number of rotary joint axes for releasing the brake is increased under conditions that do not hinder, that is, the axis is later in the order than the release target axis set in a predetermined order. If there is an axis that is orthogonal to this with respect to the current posture of the release target axis, releasing the brake of the release target axis and the simultaneous release target axis with this axis as the simultaneous release target axis is You may apply to 4th Embodiment-6th Embodiment.
In the third embodiment, the release of the brake is performed by determining whether the order of the axis before the release target axis is “equivalent to orthogonal” with respect to the release target axis for the previous order as well. May be applied to the fourth to sixth embodiments.

  In the drawings, 1 is a robot system, 2 is a robot, 3 is a controller, 5 is a base, 6 is a shoulder, 7 is a lower arm, 8 is a first upper arm, 9 is a second upper arm, 10 is a wrist, 11 is a flange, 12 is a CPU, M1 to M6 are motors, and S1 to S6 are encoders.

Claims (9)

A robot system comprising an articulated robot having a plurality of motors having a plurality of axes and corresponding to each axis, and a controller for controlling the operation of the articulated robots by driving and controlling the plurality of motors. In
The controller is
A direct teaching mode in which an operator performs teaching while directly touching the articulated robot is configured,
In the direct teaching mode,
In a preset order for the plurality of axes of the articulated robot, a predetermined axis to be released is in a state where the brake of the motor is released, and the other remaining axes are in a state where the brake of the motor is applied. A brake release step is executed to execute the brake release state to be kept for a predetermined time,
A robot system, wherein the brake release step is continuously repeated in a preset order until the end of teaching operation by a user. The robot system according to claim 1,
The controller repeatedly executes the brake release step as one cycle in which the preset order goes around once, and within one cycle as a set time for releasing the brake for each axis , A variable opening time calculated in proportion to the amount of movement so that the longer the amount of movement of the axis in the previous period, the longer the amount of movement of the axis in the fixed opening time secured in a fixed manner, and the time calculated for each axis A robot system characterized in that a time obtained by adding a variable opening time calculated so that the sum of the values is always a predetermined time is set. The robot system according to claim 2, wherein
The controller is
As the calculation of the variable opening time,
Calculate the movement amount and average movement amount of each axis in the previous cycle,
Calculate the deviation movement amount of each axis by subtracting the average movement amount from the movement amount of each axis,
The value of the deviation movement amount of each axis is added with the same sign to calculate the distributed denominator coefficient,
A value obtained by adding “1” to a value obtained by dividing the deviation movement amount of each axis by the distribution denominator coefficient is calculated as a change coefficient of each axis.
A first variable opening time setting process is performed for setting, as the variable opening time, a value obtained by multiplying a reference time preset as an initial value of the variable opening time by a change coefficient of each axis. Robot system. The robot system according to claim 2, wherein
The controller is
As the calculation of the variable opening time,
Calculate the movement amount and average movement amount of each axis in the previous cycle,
Calculate the deviation movement amount of each axis by subtracting the average movement amount from the movement amount of each axis,
A value obtained by dividing the deviation movement amount of each axis by a predetermined reference movement amount determined in advance is set as an increase rate, and a value obtained by adding “1” to the increase rate is calculated as a change coefficient of each axis.
A second variable opening time setting process is performed for setting the variable opening time by multiplying a reference time preset as an initial value of the variable opening time by a change coefficient of each axis. Robot system. The robot system according to claim 3, wherein
The controller is
As the calculation of the variable opening time,
When it is determined that the amount of movement of a specific axis among the plurality of axes is equal to or less than a preset threshold value, the first variable opening time setting process is executed,
When it is determined that the amount of movement of the specific axis exceeds the threshold value,
Calculate the movement amount and average movement amount of each axis in the previous cycle,
Calculate the deviation movement amount of each axis by subtracting the average movement amount from the movement amount of each axis,
A value obtained by dividing the deviation movement amount of each axis by a predetermined reference movement amount determined in advance is set as an increase rate, and a value obtained by adding “1” to the increase rate is calculated as a change coefficient of each axis.
A second variable opening time setting process is performed for setting the variable opening time by multiplying a reference time preset as an initial value of the variable opening time by a change coefficient of each axis. Robot system. The robot system according to claim 4 or 5,
The controller is
When executing the second variable opening time setting process,
When the increase rate of the deviation movement amount of each axis is less than “−1”, the reciprocal of the absolute value of the smallest increase rate is calculated as a compression coefficient,
In calculating the change coefficient of each axis, the robot system is characterized by adding "1" after multiplying the increase rate of the deviation movement amount of each axis by the compression coefficient. The robot system according to claim 1,
The robot system is characterized in that the controller sets a predetermined time for all the axes as a set predetermined time for releasing the brake of the motor for each axis in the brake releasing step performed in the direct teaching mode. The robot system according to any one of claims 1 to 7,
The controller is
In the direct teaching mode,
Axis to be released is set in a preset order for a plurality of axes of the articulated robot, and among the remaining axes excluding the set release target axis, an axis that is behind the release target axis Sequentially determining whether or not there is an orthogonally-corresponding axis based on the current posture of the release target axis, and if there is a corresponding axis, set it as a simultaneous release target axis, Performing the brake release step of performing the brake release of the motor for the set time for the simultaneous release target and keeping the brakes of the motors of the remaining shafts applied,
A robot system, wherein the brake releasing step is repeatedly performed until a teaching operation by a user is completed. The robot system according to claim 8, wherein
The controller is
In determining the simultaneous release target axis,
Of the remaining axes with respect to the set release target axis, the axis corresponding to the orthogonality in accordance with the order is also applied to the axis subsequent to the release target axis and subsequent to the release target axis. A robot system characterized by determining whether or not an axis exists.

Images