JP4942672B2 - Robot trajectory control apparatus and robot trajectory control method - Google Patents

Description

  The present invention relates to a robot trajectory control apparatus and a robot trajectory control method for automatically controlling a trajectory that is an operation path from a start point to an end point of a robot.

  In general, when determining the trajectory that is the movement path from the start point to the end point of an industrial robot, a waypoint is designed in addition to the start point and end point that are the destination points in order to avoid interference between the robot and the surrounding environment. There is a scene to do. Here, depending on how the waypoints are taken, the working time is unnecessarily extended, and therefore, a suitable waypoint design is required. However, there are problems that it takes time to design an appropriate waypoint, and that an operator is restrained to design a waypoint.

  As a conventional technique for solving the above problem, for example, as disclosed in Patent Document 1, a region including an obstacle (obstacle) is divided at a predetermined interval, and the division point is used as a via point candidate. There is an intermediate point teaching data creation method that employs a via point that has the shortest movement time among via point candidates that do not interfere with an interferer during movement.

  As another conventional technique, for example, as disclosed in Patent Document 2, a work target given from the outside is compared with the operation of the robot to change the waypoint, thereby achieving the target work. A robot work learning apparatus that employs the learning technique as described above also attempts to solve the above problem.

Japanese Patent No. 3134653 JP-A-8-314522

  The intermediate point teaching data creation method disclosed in Patent Document 1 does not cause interference even for via-point candidates whose movement time is longer than the movement time when using a via-point candidate that does not cause any interference that is currently found. There was a problem that the number of unnecessary investigations increased and the number of calculations was increased.

  The robot work learning device disclosed in Patent Document 2 also provides a work target at the beginning of learning because it is difficult to obtain in advance a trajectory with the shortest execution time to avoid interference with the surrounding environment. There is a problem that it is impossible to improve automatically because it is necessary to give a work target from the outside many times during the learning process.

  The present invention has been made to solve the above-described problems, and an object thereof is to provide a robot trajectory control apparatus and a robot trajectory control method that can automatically improve an operation path of a robot with a small number of calculations. To do.

A robot trajectory control device according to claim 1 of the present invention is a robot trajectory control device that controls a trajectory that is an operation path from a start point to an end point of a robot having a plurality of drive units, and the start point includes at least one route. A robot that outputs an operation command value that defines the contents of a plurality of interpolation operations that are operations of the plurality of drive units between the plurality of orbit coordinates based on orbit position information that defines a plurality of orbit coordinates including a point and an end point Each of the plurality of orbital coordinates is defined by the operation content of the plurality of drive units, and each state of the plurality of drive units realized when the robot is operated based on the operation command value. Operation information collecting means for collecting operation information including drive unit information indicating the position of the robot and surrounding information that defines a positional relationship between the robot and the surrounding environment; Based on the motion information collected by the collecting means, when the robot is operable by the motion command value, the motion executable determination means determines whether the robot is operable by the motion executable determination means. One of the at least one via points is a selected via point, and one of the plurality of interpolation operations including the selected via point is a selective interpolation operation, and in the selective interpolation operation, a drive unit which determines that the greatest impact on execution speed of the plurality of driving portions and a bottleneck drive unit, based on the operation information, definitive the bottleneck driver to the extent that the execution time of the entire robot is shortened the A route point candidate determining unit that calculates a first route point candidate by changing the coordinates of the selected route point, and outputs first route point candidate information that defines the first route point candidate; When the robot is determined to be inoperable by the operation executability determination means, based on said operation information, other than the bottleneck driver of the plurality of drive unit to the extent that the execution time of the entire robot is shortened of calculating a second via point candidates definitive the drive unit, and operable via point candidate search means for outputting a second via point candidate information specifying a route point candidate of the second, the serial second via-points If the second waypoint candidate defined by the candidate information exists, the second waypoint candidate is determined as a change waypoint, and the second waypoint candidate does not exist and the first waypoint candidate information If the first waypoint candidate defined by is present, the first waypoint candidate is determined as the change waypoint, and the selected waypoint is replaced with the change waypoint among the plurality of trajectory coordinates. As the new orbital position information And a waypoint changing means to be given to the robot control means.

A robot trajectory control method according to a third aspect of the present invention is a robot trajectory control method for controlling a trajectory that is an operation path from a start point to an end point in a robot having a plurality of drive units, and (a) at least a start point, Obtaining trajectory position information defining a plurality of trajectory coordinates including one waypoint and an end point, each of the plurality of trajectory coordinates being defined by the operation content of the plurality of driving units, and (b) the at least one trajectory. One of the via points is a selected via point, and one interpolation operation including the selected via point among a plurality of interpolation operations that are operations of the plurality of drive units between the plurality of trajectory coordinates. and selecting interpolation operation, the drive unit it is determined that most affect the execution speed of the plurality of drive unit in said selected interpolation operation is a bottleneck driver, of the entire robot real Calculating a first route point candidates by changing the coordinates of the selected waypoint to definitive the bottleneck driver to the extent that time is reduced, (c) among the plurality of track coordinates, wherein step ( This is executed when the robot is inoperable when the information obtained by replacing the first waypoint candidate obtained in step b) with the selected waypoint is operated with the trajectory position information, and the execution time of the entire robot is reduced. calculating a second via point candidates definitive the drive unit other than the bottleneck driver among the plurality of driving portions to the extent, said if there is; (d) second via point candidate 2 waypoint candidates are determined as change waypoints, and if the second waypoint candidate does not exist and the first waypoint candidates exist, the first waypoint candidate is determined as the change waypoint. , The selected via point among the plurality of trajectory coordinates Further comprising the step of setting the information obtained by replacing the change via point as the new trajectory position information until the calculation of the first and second via point candidates becomes impossible. Repeat (d).

  The robot trajectory control device according to claim 1 of the present invention is executed by the waypoint candidate determining means or the actionable waypoint candidate searching means based on the action information obtained from the action information collecting means while actually operating the robot. The first or second waypoint candidate is obtained within a range in which the time is shortened. Then, the waypoint changing means determines the first or second waypoint candidate as the change waypoint.

  Therefore, the robot trajectory control apparatus according to claim 1 can obtain a small number of calculation procedures by automatically and sequentially obtaining information obtained by replacing the change via point determined by the via point changing unit with the selected via point as new trajectory position information. Therefore, the trajectory of the robot can be improved without causing interference with the surrounding environment.

  The robot trajectory control method according to claim 3 of the present invention is the first or second waypoint in which the execution time is reduced by executing step (b) and step (c) while actually operating the robot. I'm getting a candidate. In step (d), the first or second waypoint candidate is determined as a change waypoint.

  Therefore, in the robot trajectory control method according to the third aspect, information obtained by replacing the selected via point with the changed via point among a plurality of orbit coordinates is automatically obtained sequentially as new orbit position information by executing step (d). Thus, the trajectory of the robot can be improved without causing interference with the surrounding environment with a small number of calculation procedures.

<Embodiment 1>
(Constitution)
FIG. 1 is a block diagram showing the configuration of the robot trajectory control apparatus according to Embodiment 1 of the present invention.

  As shown in the figure, the robot control means 1 receives initial information I1 which is trajectory position information defining a plurality of trajectory coordinates including a start point coordinate, at least one initial via point coordinate, and an end point coordinate. The plurality of orbit coordinates are defined by the operating states of the plurality of driving units of the robot apparatus 2.

  Then, the robot control means 1 outputs an operation command value C1 defining the contents of a plurality of interpolation operations between a plurality of trajectory coordinates to the robot apparatus 2 based on the initial information I1. The contents of the plurality of interpolation operations are determined according to a predetermined interpolation operation type.

  The robot apparatus 2 operates in accordance with the operation command value C1 output from the robot control means 1, and detects an internal sensor that detects the state of the driving section of the robot apparatus including the additional axis and the position of the robot and the surrounding environment. An external sensor that detects peripheral information that defines the relationship is provided. The internal sensor indicates an encoder or a sensor that detects a current value of a motor provided in each driving unit of the robot apparatus 2, and the external sensor indicates a visual sensor, a distance sensor, a tactile sensor, or the like. As an additional axis, for example, a rail on which a robot is mounted can be considered.

  Note that the external sensor does not need to accurately recognize the position and size of the obstacle, and that the robot and the obstacle are more than a certain distance or that the robot body contacts the obstacle and is damaged. It is enough to prevent it.

  For example, in the case of a tactile sensor, it is only necessary (at a minimum) to know which robot's posture is in contact, and information on which part of the robot is in contact is not necessary. In the case of a visual sensor or a distance sensor, a margin is set in advance (determined from sensor error or the like), and if the margin is used, the robot and the obstacle need to be accurate enough to avoid unintended contact.

  The motion information collecting means 3 collects the motion information I3 including the drive unit information detected by the internal sensor provided in the robot device 2 and the peripheral information which is the output of the external sensor in time series.

  Based on the motion information I3 collected by the motion information collection unit 3 described above, the motion feasibility determination unit 4 determines whether or not the robot apparatus 2 is operable based on the motion command value C1, and can perform the motion that is the determination result. Information I4 is output.

  As a determination method by the operation feasibility determination unit 4, a collision detection determination, an overspeed determination, an overtorque determination, a singular point determination, an operation limit range determination, and the like using a known internal sensor can be considered. Further, an external sensor such as a distance sensor, a vision sensor, a tactile sensor, or a contact sensor may be used as a material for determining whether the distance between the robot and the surrounding environment is maintained above a certain level. Further, it may be used as a material for determining whether the robot is obstructing the recognition work of the external sensor, such as whether the robot is positioned within the recognition range and obstructing the recognition work, such as a distance sensor or a vision sensor.

  The route point candidate determination unit 5 inputs the operation information I3 obtained from the operation information collection unit 3 when the operation enable information I4 indicates that the operation is possible, and selects a route point candidate whose execution time is shortened based on the operation information I3. decide. At this time, if the trajectory is composed of a plurality of via points, a via point whose coordinates are changed is also selected, and an end determination is made as to whether there is a via point where the trajectory can be improved.

  When the waypoint candidate determining means 5 finds a waypoint where the improvement of the trajectory can be found, the waypoint candidate determining unit 5 specifies the improvement information I50 instructing the improvement and the coordinates of the waypoint candidate (first waypoint candidate) whose execution time is shortened. If the point candidate information I51 (first waypoint candidate information) is output and it is determined that there is no prospect of improvement, the improvement information I50 indicating the improvement failure (possible) is output.

  The operable route point candidate searching means 6 receives the operable information I4, and enters the operating state when the operable information I4 instructs the inoperability. The purpose of the operable waypoint candidate searching means 6 is to change other waypoint candidates that can be operated within the execution time shortened by the waypoint candidate determining means 5 by changing the coordinates of the waypoints when the operation is impossible (second way) Is to find a waypoint candidate).

  Further, when the second waypoint candidate searching means 6 cannot find the second waypoint candidate, the operable waypoint candidate searching means 6 gives the route point candidate determining means 5 a bottleneck coordinate change command C6 instructing to try to improve the trajectory.

  When the waypoint candidate determining means 5 receives the bottleneck coordinate change command C6 from the operable waypoint candidate searching means 6, the waypoint candidate determining means 5 discards the first waypoint candidate most recently indicated by the waypoint candidate information I51, and creates a new one. Processing for determining one waypoint candidate is performed.

  As will be described in detail later, the waypoint candidate determination unit 5 uses one of the waypoints as a selection waypoint, and passes through the range where the execution time is reduced with respect to the bottleneck driving unit in the selection interpolation operation including the selection waypoint. An operation for determining point candidates is performed. On the other hand, the operable waypoint candidate searching means 6 performs an operation of determining a second waypoint candidate in which the robot apparatus 2 can operate within a range in which the execution time is reduced with respect to the drive units other than the bottleneck drive unit.

  The waypoint changing means 7 inputs the waypoint candidate information I52 from the operable waypoint candidate searching means 6 in addition to the improvement information I50 and the waypoint candidate information I51 obtained from the waypoint candidate determining means 5.

  Then, when the improvement information I50 indicates that improvement is possible, the waypoint changing unit 7 determines the first waypoint candidate indicated by the waypoint candidate information I51 as the change waypoint, and changes the coordinates of the selected waypoint. Information consisting of a plurality of trajectory coordinates replaced with the coordinates of the via points is output to the robot control means 1 as changed trajectory position information I7 (new trajectory position information). Then, the robot control unit 1 outputs an operation command value C1 based on the changed trajectory position information I7 to the robot apparatus 2.

  However, when there is a second waypoint candidate indicated by the waypoint candidate information I52, the waypoint changing unit 7 determines the second waypoint candidate as the change waypoint and changes the coordinates of the selected waypoint. Information including a plurality of trajectory coordinates replaced with the coordinates of the via points is output to the robot control means 1 as changed trajectory position information I7. This is because when the second waypoint candidate indicated by the waypoint candidate information I52 exists, the robot device 2 cannot execute the first waypoint candidate indicated by the waypoint candidate information I51. It is because it was determined by.

  On the other hand, when the improvement information I50 indicates that improvement is not possible, the waypoint changing means 7 outputs in the past including the first and second waypoint candidates designated by the waypoint candidate information I51 and the waypoint candidate information I52 in the past. Of the changed trajectory position information I7, the changed trajectory position information I7 that defines a plurality of trajectory coordinates that are determined to be operable by the operation feasibility determination unit 4 and that has the shortest execution time is finally determined.

  Thereafter, the waypoint changing means 7 performs the trajectory control of the robot apparatus 2 based on the finally determined changed trajectory position information I7.

  FIG. 2 is an explanatory diagram showing a specific apparatus configuration (first apparatus configuration) for realizing the robot trajectory control apparatus according to the first embodiment. As shown in the figure, a robot control device 22 is used as the robot control means 1, and a robot 21 corresponding to the robot device 2 operates according to an operation command value output from the robot control device 22. Outputs of internal sensors such as drive unit information indicating the state of each drive unit of the robot 21 and execution time are output to the electronic computer 24 via the robot control device 22. The output (peripheral information) of the external sensor provided in the robot 21 is output from the robot 21 to the electronic computer 24 via the sensor cable 25.

  The electronic computer 24 performs a calculation for determining a waypoint candidate for improving the trajectory. Specifically, the operation information collecting means 3, the action feasibility determining means 4, and the waypoint candidate shown in FIG. The functions of the determining means 5, operable route point candidate searching means 6, and route point changing means 7 are implemented.

  The teaching box 23 is a device in which an operator gives initial information I1 to the robot control device 22. That is, the teaching box 23 is used for teaching the initial information I1 including the start point coordinates, the end point coordinates, and the initial via point coordinates, or for an emergency stop during the operation of the robot 21. Needless to say, the teaching data may be input from the electronic computer 24.

  FIG. 3 is an explanatory diagram showing a specific example of the robot 21. The robot 21 shown in the figure operates according to an operation command value from a robot control device 22 including a command defining a plurality of interpolation operation contents between a trajectory coordinate that is a start point, at least one waypoint, and an end point coordinate. It corresponds to.

  As shown in the figure, the robot 21 includes driving units D1 to D6, a base unit K0, joints K1 to K5, and a gripping unit K6. The positions of the joints K1 to K5 are changed by driving (rotating or the like) by the driving units D1 to D5, and the gripping unit K6 performs a gripping operation by the driving unit D6. For convenience of explanation, the base portion K0 and the grip portion K6 may be simply referred to as joint portions K0 to K6.

  In the following description, the position / orientation and teaching data of the robot 21 are described by the respective positions of the driving units D1 to D6. For example, in the case of a 6-axis vertical articulated robot as shown in FIG. 3, the position of the drive unit Di (i = 1 to 6) is expressed as Ji, and the position and orientation of the robot and the coordinates of teaching data ( (Orbit coordinates) is represented by (J1, J2, J3, J4, J5, J6).

  As an interpolation command for instructing the content of the interpolation operation that connects the teaching data, an example of a command for performing interpolation in units of the driving units D1 to D6 will be considered. Specifically, there is an interpolation operation type in which each drive unit Di starts to move at the same time, moves as fast as possible during the operation, and each drive unit Di finishes the operation at the same time. In addition, the robot 21 does not pause at the via point, but performs an operation of smoothly connecting two interpolation operations sandwiching the via point (interpolation operation from the start point to the via point, and interpolation operation from the via point to the end point). To do. Interpolation commands, which are operation commands related to these two interpolation operations, are frequently used particularly in scenes where improvement of the trajectory is required or where there is much room for improvement of the trajectory.

(Operation)
FIG. 4 is a flowchart showing the processing contents of the robot trajectory control method in the robot trajectory control apparatus of the first embodiment. Hereinafter, the robot trajectory control method by the robot trajectory control apparatus according to the first embodiment will be described with reference to FIG.

  First, as shown in step S11, initial information I1 defining a plurality of trajectory coordinates including a start point coordinate, an end point coordinate, and at least one initial via point coordinate is input to the robot control means 1.

  Based on the initial information I1, the robot control means 1 outputs an operation command value C1 that defines the contents of the interpolation operation between a plurality of trajectory coordinates to the robot apparatus 2 to drive-control the robot apparatus 2. The plurality of initially set orbital coordinates may be any number as long as the operation can be performed, and any number of via points may be used.

  When the robot apparatus 2 performs the operation, the operation information collection unit 3 collects the operation information including the drive unit information and the peripheral information of the robot apparatus 2 in time series, and outputs it as the operation information I3.

  The operation executable determination unit 4 determines whether the operation can be executed based on the operation information I3 and outputs the operation executable information I4. If it is determined that the operation cannot be performed during the operation of the robot apparatus 2, the operation start state is stopped by stopping the operation execution and tracing the operation in reverse to prevent the robot apparatus 2 from being damaged. Return to. On the other hand, if there is no risk of damage to the robot apparatus 2, there is no need to stop the operation execution. Naturally, the interruption of the operation may be performed by the robot controller 22 using a known technique.

  When the operation executable determination unit 4 determines that the operation can be executed (when the operation enable information I4 indicates that the operation can be performed), the waypoint candidate determination unit 5 selects each drive unit Di collected by the operation information collection unit 3. Based on the operation information I3 (driving unit information), a waypoint candidate determination process for determining a waypoint candidate whose execution time is shortened is executed.

  Hereinafter, the waypoint candidate determination unit 5 will be described with respect to a waypoint candidate determination process. The waypoint candidate determination processing corresponds to the processing in steps S14 to S17 in FIG.

  First, the concept that is the premise of step S14 will be described. In the plurality of interpolation operations described above, the execution time of each interpolation operation is determined by the execution time of the drive unit that takes the longest time when each drive unit Di is moved independently.

  Therefore, this drive unit that determines the execution time of the interpolation operation of the robot is defined as a bottleneck in this interpolation operation and is denoted as “bottleneck drive unit BD”. The waypoint candidate determination means 5 has a function of shortening the execution time from the start point to the end point by reducing the execution time of each bottleneck drive unit BD in each interpolation operation. At this time, when a plurality of drive units corresponding to the bottleneck drive unit BD exist simultaneously in one interpolation operation, one is appropriately selected as the bottleneck drive unit BD.

  On the basis of the above-mentioned premise, the waypoint candidate determination means 5 appropriately selects the waypoint to be changed (selected waypoint) and the interpolation operation to be noticed (selected interpolation operation) in step S14.

  Thereafter, in step S15, the waypoint candidate determination means 5 selects the drive unit bottleneck drive unit BD that becomes the bottleneck in the selective interpolation operation from the operation information of each drive unit Di at the time of executing the operation, and based on the operation information I3. For the bottleneck drive unit BD, a waypoint candidate determination process is executed to determine the first waypoint candidate within a range where the execution time is shortened. Here, an example in which the bottleneck driving unit BD is determined based on the speed information of each driving unit Di will be described.

  FIG. 5 is a flowchart showing a method of selecting the bottleneck driving unit BD by the waypoint candidate determining means 5. Hereinafter, the selection processing contents of the bottleneck driving unit BD will be described with reference to FIG.

  First, in step S1, an arrival speed Vi that is the maximum speed reached by the drive unit Di during the interpolation operation is obtained for each drive unit Di.

  Subsequently, in step S2, a coefficient value Pi obtained by dividing the arrival speed Vi by the rated speed VMi that is the maximum speed of each drive part Di previously determined for the robot is obtained for each drive part Di.

  Finally, in step S3, the coefficient values Pi of the respective drive units Di are compared, and the drive unit having the largest coefficient value Pi is set as the bottleneck drive unit BD in this interpolation operation.

  After determining the bottleneck driving unit BD, the waypoint candidate determining unit 5 determines a first waypoint candidate whose execution time is shortened from the speed change pattern of the bottleneck driving unit BD.

  Returning to FIG. 4, in step S <b> 16, the waypoint candidate determination unit 5 displays the speed change pattern of the bottleneck driving unit BD in the selection interpolation operation which is one of the two interpolation operations sandwiching the selected waypoint in cases A to A. Sort by Case C.

  Hereinafter, cases A to C will be described with examples. In the series of movement from the start point to the via point (× 1 (the number of via points is “1”)) to the end point, a drive unit (hereinafter, description) that becomes the bottleneck drive unit BD when “start point → via point” is selected and interpolated. For convenience, the speed change patterns of “the first half bottleneck drive unit BDF” may be abbreviated in the following three cases.

  Case A: The movements in the two interpolation operations “start point → route point” and “route point → end point” are in the same direction, and the first half bottleneck drive unit BDF remains unchanged even in the interpolation operation “route point → end point”. When it becomes a bottleneck.

  Case B: The two interpolation operations “start point → route point” and “route point → end point” move in the same direction (including stop), and in the “route point → end point” interpolation operation, the first half bottleneck When another drive unit different from the drive unit BDF becomes a bottleneck.

  Case C: The moving direction of the first half bottleneck drive unit BDF differs between two interpolation operations “start point → route point” and “route point → end point”.

  Similarly, a speed change pattern of a drive unit that becomes a bottleneck drive unit BD (hereinafter, may be abbreviated as “second half bottleneck drive unit BDR” for convenience of explanation) in the interpolation operation “route point → end point”. Think.

  In this case, as a case equivalent to case B (hereinafter referred to as “case B ′”), the movement from “start point → route point” and “route point → end point” is in the same direction (including stop). In addition, in the “start point → route point” interpolation operation, another driving unit different from the latter half bottleneck driving unit BDR may be a bottleneck. In addition, about the latter half bottleneck drive part BDR, since case A and case C are substantially the same as the case of the first half bottleneck drive part BDF, description is abbreviate | omitted.

  FIG. 6 is an explanatory diagram showing characteristic speed change patterns corresponding to Case A to Case C for the first half bottleneck drive unit BDF described above.

  As shown in FIG. 5A, the single drive unit operates at the maximum speed in one direction by the movement of “start point → end point”, and therefore corresponds to case A. Therefore, even if the coordinates of the via point G10 are changed, the execution time of “start point → end point” cannot be further shortened. Therefore, in the case A, the waypoint candidate determination unit 5 determines that the waypoint operated this time is optimal, does not output a new waypoint candidate, and outputs the improvement information I50 indicating that improvement is impossible.

  As shown in FIG. 5B, the overall movement amount of “first point → end point” of the first-half bottleneck drive unit BDF of interest is biased toward the movement side of “start point → route point”. Applicable. Therefore, the waypoint candidate determination means 5 is the coordinate part of the first half bottleneck driving unit BDF in the waypoint coordinate G20 (for example, when the first half bottleneck driving unit BDF is the driving unit D3, J3 among J1 to J6 is applicable). Is determined as a new route point candidate by passing the route point coordinates close to the coordinates of the first half bottleneck drive unit BDF in the start point coordinates. At this time, the waypoint candidate determination means 5 outputs the improvement information I50 instructing improvement and the waypoint candidate information I51 instructing the new waypoint candidate.

  Note that when the latter half bottleneck drive unit BDR is focused on, if the speed change pattern is reversed left and right around the via point coordinates G20, the whole of “via point → end point” will be shown. Is shifted to the moving side of “route point → end point”, which corresponds to case B ′. In the case B ′, the waypoint candidate determination means 5 newly uses the waypoint coordinates obtained by bringing the coordinate portion of the latter half bottleneck driving unit BDR in the waypoint coordinates G20 closer to the coordinates of the latter half bottleneck driving unit in the end point coordinates. The point candidate will be determined.

  As shown in FIG. 6C, the first half bottleneck driving unit BDF changes the moving direction at the waypoint G30 and performs a detouring operation around the circuit, and corresponds to Case C. Therefore, a via point coordinate obtained by bringing the coordinates of the first half bottleneck drive unit BDF in the via point coordinates close to the coordinates of the first half bottleneck drive unit BDF in the start point coordinate is determined as a new via point candidate. At this time, even if it approaches the coordinates of the first half bottleneck drive unit BDF in the end point coordinates, the same effect can be obtained.

  Returning to FIG. 4, in step S <b> 17, the waypoint candidate determination unit 5 finally determines a waypoint candidate to replace the selected waypoint based on the cases A to C classified in step S <b> 16.

  That is, the waypoint candidate determining means 5 outputs the improvement information I50 indicating that improvement is impossible when the classification pattern in step S16 is case A, and indicates that improvement is possible when the classification pattern is case B or case C. And the route point candidate information I51 indicating the first route point candidate determined as described above is output.

  As described above, in the case B or C, the waypoint candidate determination unit 5 sets the coordinates that exclude the biased movement or useless movement of the first half bottleneck drive unit BDF that determines the execution time of the interpolation operation. It is assumed that 1 waypoint candidate. As a result, the execution time of the robot apparatus 2 can be shortened when the robot apparatus 2 operates from “start point → end point”.

  Here, a method of selecting at random as a method of selecting the selection via point to be changed and the selection interpolation operation in step S14 can be taken. In this case, an improvement in average improvement can be expected by making improvements several times and selecting the best one.

  Further, the number of via-point changes may be selected to be uniform (except when the execution time cannot be shortened). In this case, since the optimization progresses on average, it can be expected that the number of changes until the improvement is completed is reduced. Alternatively, an interpolation operation having a large influence on the entire execution time, such as an interpolation operation with a long execution time or a case C interpolation operation prior to the case B interpolation operation, may be selected. In this case, since the room for changing the waypoint is prioritized to those that can greatly reduce the execution time, an improvement effect can be expected.

  When the waypoint candidate information I51 instructing the first waypoint candidate is output by the waypoint candidate determining means 5, the waypoint changing means 7 changes the first waypoint candidate designated by the waypoint candidate information I51. The changed trajectory position information I7, which is determined as a point and is composed of a plurality of new trajectory coordinates in which the coordinates of the selected via point is replaced with the coordinates of the changed via point, is output to the robot control means 1.

  As a result, the robot control means 1 gives an operation command value C1 based on the changed trajectory position information I7 to the robot apparatus 2, and controls the trajectory of the robot apparatus 2. The operation state of the robot apparatus 2 at this time is collected by the operation information collecting means 3 and output as operation information I3.

  As a result, in step S18, the action execution possibility determination means 4 determines whether or not the movement point candidate (first via point candidate) indicated by the via point candidate information I51 is operable, and outputs the operable information I4. To do. If the operable information I4 indicates that operation is possible, the process returns to step S12.

  In step S12, if it is determined by the waypoint candidate determination means 5 that the execution time cannot be shortened for all the waypoints and the interpolation operation (YES), the process proceeds to step S13. For example, if the classification speed change pattern in step S16 indicates case A in all the waypoints and interpolation operations, YES is determined in step S12.

  In step S13, the learning is finished, and the waypoint changing means 7 uses the changed trajectory position information I7 that has been determined to be operable among the changed trajectory position information I7 output in the past and that has the shortest execution time. Final decision.

  Thereafter, the transit point changing means 7 only has to output the changed trajectory position information I7 with the finally determined transit point as trajectory coordinates to the robot control means 1, and it is not necessary to change the contents of the changed trajectory position information I7 thereafter. In this way, the trajectory improvement process by the robot trajectory control apparatus according to the first embodiment ends. However, the operation state may be continuously monitored, and learning may be performed again when the robot speed pattern changes due to aging or the like.

  In step S12, when it is determined that there is a possibility that the execution time can be shortened by the waypoint and the interpolation operation (NO), the waypoint candidate determination process in steps S13 to S17 is performed again. Note that if the via-point candidate determination process by the via-point candidate determining unit 5 has never been performed and immediately after the initial information I1 is input, the determination in step S12 is likely to be NO.

  Next, the search process by the operable route point candidate searching means 6 when it is determined in step S18 that the operation cannot be executed (NO) will be described.

  The search processing by the operable route point candidate searching means 6 avoids interference based on the action information I3 collected by the action information collecting means 3 and using the action information of each drive unit Di to the extent that the execution time is shortened. This is done by determining via point candidates.

  When the waypoint candidate determination unit 5 changes the waypoint candidate in order to shorten the execution time, the execution time is calculated from the operation information of each drive unit Di based on the operation information I3 obtained from the operation information collection unit 3. Whether it is shortened can be calculated. Similarly, a change in execution time when the coordinates corresponding to other driving units of the waypoints are changed can be calculated based on the operation information I3. Therefore, the operable waypoint candidate searching means 6 can derive a range in which the execution time is shortened from these calculation results and determine a second waypoint candidate that avoids interference within the range. Hereinafter, the search process of the operable waypoint candidate searching means 6 will be described in detail with reference to FIG.

  First, in step S33, a drive unit other than the bottleneck drive unit BD calculated in the latest step S15 is selected as another selection drive unit ED.

  Then, in step S34, the second waypoint candidates obtained by using the following methods (1) to (3) for the other selection drive unit ED within the range where the execution time is not extended, It is derived as a waypoint candidate instead of (reflecting the contents of the first waypoint candidate made by the waypoint candidate determining means 5), and the waypoint candidate information I52 defining the second waypoint candidate is output.

  (1) When the other selection drive unit ED does not correspond to the bottleneck drive unit BD in any of the interpolation operations before and after the via point, the selection is made only for the amount of the range that does not become the bottleneck drive unit BD regardless of the direction of change. The coordinate part of the other selection driving unit ED at the point is changed.

  (2) When the other selection drive unit ED corresponds to the bottleneck drive unit BD in at least one of the interpolation operations before and after the via point, other via points are determined in the same manner as the via point candidate determination process by the via point candidate determining unit 5. The coordinates of the selection drive unit ED are changed.

  (3) If it is not determined that the operation can be executed even with the two changes (1) and (2) above, the extended execution time is shorter than the execution time shortened by changing the bottleneck drive unit BD. To change the coordinates of another selection drive unit ED of the waypoint.

  When the route point candidate information I52 indicating the second route point candidate is output by the operable route point candidate searching means 6, the route point candidate indicated by the route point candidate information I52 is changed by the route point changing means 7. The changed trajectory position information I7 including a plurality of new trajectory coordinates in which the coordinates of the selected via points are replaced with the coordinates of the changed via points is output to the robot control means 1.

  The robot controller 1 controls the trajectory of the robot apparatus 2 by giving the robot apparatus 2 an operation command value C1 based on the changed trajectory position information I7. The operation status of the robot apparatus 2 is collected by the operation information collecting means 3 and output as operation information I3.

  As a result, in step S35, the action execution possibility determination means 4 determines whether or not the movement point candidate indicated by the passage point candidate information I52 is operable (no interference, etc.) and outputs the movement availability information I4. To do. When the operable information I4 indicates that operation is possible (YES), the process returns to step S12.

  On the other hand, when the operation executable determination unit 4 determines that the operation is impossible (NO in step S35), in step S36, the second waypoint candidate indicated by the waypoint candidate information I52 is discarded, and the process returns to step S31.

  If it is determined in step S31 that even if all the drive units Di other than the bottleneck drive unit BD are changed, the execution time cannot be shortened (YES), the route point candidate information I51 is instructed in step S32. The first waypoint candidate is discarded, and the process returns to step S12. That is, in step S32, the operable waypoint candidate searching means 6 outputs a bottleneck coordinate change command C6 that instructs the waypoint candidate determining means 5 to determine another waypoint candidate.

  If it is determined in step S31 that if any one of the drive units Di other than the bottleneck drive unit BD is changed, the execution time can be expected to be shortened (NO), the operable via point candidate searching means 6 in steps S33 to S34 performs. Further search processing is performed. In addition, when the search process by the operable route point candidate searching unit 6 has never been performed, the determination in step S31 is likely to be NO.

  Thereafter, in steps S33 to S34, the search process is performed with another drive unit ED that has not yet been tried.

  As described above, the robot trajectory control apparatus according to the first embodiment performs the waypoint candidate determination unit 5 (steps S12 to S14) based on the motion information I3 obtained from the motion information collection unit 3 while actually operating the robot device 2. ) Or operable waypoint candidate searching means 6 (steps S33 and S34), the first or second waypoint candidates are obtained within a range where the execution time is shortened. Then, the waypoint changing unit 7 determines the first or second waypoint candidate as the change waypoint.

  Therefore, the robot trajectory control apparatus according to the first embodiment automatically and sequentially obtains the information obtained by replacing the change via point determined by the via point changing unit 7 with the selected via point as the changed trajectory position information I7, thereby reducing the number of calculations. The trajectory of the robot can be improved without causing interference with the surrounding environment in the procedure.

  Here, the change amount of the via point coordinates when determining the via point candidate (step S17, step S34), but when changing by a predetermined value, it can be improved by adjusting according to the scene where the work is performed. It is possible to appropriately adjust the time required for improvement and the improvement effect. In the case of the speed change pattern represented by case A to case C of the driving unit during the process of determining the waypoint candidate, the amount of change in the waypoint coordinates is set so that the speed change pattern just changes to another case. You may make it adjust. In this case, there is an effect of preventing the two via point coordinates from being alternately set as via point candidates within the same speed change pattern. In addition, when the interference with the obstacle does not occur so much, the change amount is increased. When the interference occurs frequently, the change amount is decreased.

(First example of via point candidate determination process)
7 to 16 are diagrams showing a first example of the waypoint candidate determination process by the waypoint candidate determination means 5. 7, 9, 11, 13, and 15 are explanatory views schematically showing changes in the hand position from the start point to the end point including two via points. FIG. 8, FIG. 10, FIG. 12, and FIG. FIG. 16 is a graph showing a change in speed of two driving units that move parallel to the x-axis direction and the y-axis direction in the movement of the hands of FIGS. 7, 9, 11, 13, and 15.

  In order to simplify the description of the waypoint candidate determination process, it is assumed that there are two drive units that move parallel to the x-axis and the y-axis, respectively. In addition, it is assumed that both the drive unit that moves parallel to the x-axis direction and the drive unit that moves parallel to the y-axis direction have the same rated speed VMi. This also applies to the second to fourth examples described below.

  7, 9, 11, 13, and 15, the hand position P11 that is the start point coordinate passes through the hand position P12 that is the transit point VP1 and the hand position P13 that is the transit point VP2, and the hand that is the end point coordinate. It shows moving to the position P14. That is, the hand of the robot apparatus 2 moves in the order of P11 → P12 → P13 → P14. The object to be moved by the hand is the workpiece W1, and an obstacle E1 exists between the hand positions P11 to P14.

  Hereinafter, the route point candidate determination process by the route point candidate determination means 5 when the initial trajectory shown in FIGS. 7 and 8 is given by the initial information I1 will be described in association with the flowchart shown in FIG.

  In the initial trajectory shown in FIGS. 7 and 8, attention is paid to the interpolating operation of the via point VP1 (hand position P12) and the starting point (hand position P11) → the via point VP1 (step S14 in FIG. 4). That is, VP1 is the selection via point, and “start point → route point VP1” is the selection interpolation operation. Based on the motion information (speed change) shown in FIG. 8 about the initial trajectory, it is recognized that the bottleneck drive unit BD is the drive unit in the y-axis direction in the selective interpolation operation (step in FIG. 4). S15).

  Here, the speed change pattern at the starting point of the driving unit in the y-axis direction → the via point VP1 → the via point VP2 corresponds to the case B described above (step S16 in FIG. 4). Therefore, the y-axis coordinate of the via point VP1 is changed in the starting point coordinate direction (step S17 in FIG. 4), and the second trajectory shown in FIGS. 9 and 10 is obtained as the changed trajectory.

  Since this second trajectory does not interfere with the obstacle E1, YES is obtained in step S18 of FIG. Further, since there is a possibility of improvement of the trajectory from the second trajectory, NO is returned in step S12, and the process returns to step S14 in FIG.

  When the second trajectory is obtained, attention is paid to the interpolation operation of via point VP1 and via point VP1 → route point VP2 (hand position P13) (step S14 in FIG. 4). In other words, the via point VP1 is the selected via point, and “via point VP1 → the via point VP2” is the selective interpolation operation. Based on the operation information shown in FIG. 10 for the second trajectory, it is recognized that the bottleneck drive unit BD is the drive unit in the x-axis direction in the selective interpolation operation (step S15 in FIG. 4).

  Here, the speed change pattern at the start point of the driving unit in the x-axis direction → the transit point VP1 → the transit point VP2 corresponds to the case B ′ (step S16 in FIG. 4). Accordingly, the x-axis coordinate of the via point VP1 is changed to the coordinate direction of the via point VP2 (step S17 in FIG. 4), and the third trajectory shown in FIGS. 11 and 12 is obtained as the changed trajectory.

  Since this third trajectory does not interfere with the obstacle E1, YES is obtained in step S18 of FIG. Further, since there is a possibility of improvement in the trajectory from the third trajectory, NO is determined in step S12, and the process returns to step S14 in FIG. Hereinafter, the same explanation about correspondence with Drawing 4 is omitted suitably.

  When the third trajectory is obtained, attention is paid to the interpolation operation of the via point VP2 and the via point VP2 → the end point (hand position P14). That is, the via point VP2 is the selection via point, and the “via point VP2 → end point” is the selective interpolation operation. Based on the operation information of FIG. 12 about the third trajectory, it is recognized that the bottleneck drive unit BD is the drive unit in the y-axis direction in the selective interpolation operation.

  The speed change pattern at the via point VP1 → the via point VP2 → the end point of the driving unit in the y-axis direction corresponds to the case C. Therefore, the fourth axis shown in FIGS. 13 and 14 is obtained by changing the y-axis coordinate of the via point VP2 to the coordinate direction of the via point VP1.

  When the fourth trajectory is obtained, attention is paid to the interpolation operation of the waypoint VP2 and the waypoint VP1 → the waypoint VP2. In other words, the via point VP2 is the selected via point, and “via point VP1 → the via point VP2” is the selective interpolation operation. Based on the operation information of FIG. 14 regarding the fourth trajectory, it is recognized that the bottleneck drive unit BD is the drive unit in the x-axis direction in the selective interpolation operation.

  The speed change pattern from the via point VP1 → the via point VP2 → the end point of the driving unit in the x-axis direction corresponds to the case B. Therefore, the x-axis coordinate of the via point VP2 is changed to the coordinate direction of the via point VP1, resulting in the fifth trajectory shown in FIGS.

  Since the fifth trajectory does not interfere with the obstacle E1, YES is determined in step S18 in FIG. 4, and the process proceeds to step S12.

  From the operation information shown in FIG. 16, the bottleneck drive unit BD is the drive unit in the x-axis direction in all the interpolation operations from the start point → route point VP1 → route point VP2 → end point, and the speed change pattern is the case. Since it is A, no further reduction in execution time can be expected, so the improvement ends (YES in step S12 in FIG. 4).

  As a result, the learning of the trajectory improvement of the robot apparatus 2 is finished, and the waypoint changing means 7 is executed while it is determined that the past teaching data (the trajectory coordinates adopted as the changed trajectory position information I7) is operable. Change to the coordinates of the waypoint candidate with the shortest time. (Step S13 in FIG. 4).

(Second example of via point candidate determination processing)
FIGS. 17 to 20 are diagrams showing a second example of the waypoint candidate determination process by waypoint candidate determination means 5. FIGS. 17 and 19 are explanatory views schematically showing a hand position change from the start point to the end point including two via points. FIGS. 18 and 20 show the x-axis in the movement of the hand of FIGS. It is a graph which shows the speed change of two drive parts which move in parallel to a direction and a y-axis direction.

  17 and 19, the hand position P21 which is the start point coordinate is moved to the hand position P24 which is the end point coordinate via the hand position P22 which is the via point VP1 and the hand position P23 which is the via point VP2. Yes. That is, the hand of the robot apparatus 2 moves in the order of P21 → P22 → P23 → P24. The object to be moved by the hand is the workpiece W2, and an obstacle E2 exists between the hand positions P21 to P24.

  FIG. 37 is a flowchart showing the processing contents of the robot trajectory control method in the robot trajectory control apparatus according to the first embodiment, corresponding to the second and third examples.

  As shown in the figure, the via point deletion process of step S21 executed after YES in step S18 and after YES in step S35 is added. That is, in step S21, the waypoint candidate determination unit 5 determines that the waypoint after the trajectory is changed when the coordinates of the waypoint after the trajectory change are on the trajectory connecting the teaching point coordinates before and after the waypoint. Is deleted. After executing step S21, the process proceeds to step S12. Note that step S21 is executed even after step S32 is executed, but usually does not meet the condition of step S21. The other processes are the same as those in the flowchart shown in FIG.

  Hereinafter, the waypoint candidate determination processing by waypoint candidate determination means 5 when the initial trajectory shown in FIGS. 17 and 18 is given by the initial information I1 will be described in association with the flowchart shown in FIG. 37 as appropriate.

  When the initial trajectory shown in FIGS. 17 and 18 is given, the waypoint candidate determination means 5 performs the waypoint candidate determination process in the same manner as in the first example to obtain the improved trajectory shown in FIGS. 19 and 20. In other words, the waypoint candidate determination means 5 performs a waypoint candidate determination process by paying attention to the interpolation operation of the waypoint VP1 (hand position P22) and the waypoint VP1 → the waypoint VP2 (hand position P23). “Intermediate point VP2 and via point VP1 → interpolation operation of via point VP2” are performed, and via point candidate determination processing is performed to obtain the improved trajectory.

  When the improved trajectory is obtained, attention is paid to the interpolation operation of the via point VP1 and the start point → the via point VP1 (step S14 in FIG. 37). That is, the via point VP1 is the selected via point, and “start point → via point VP1” is the selective interpolation operation. Based on the operation information shown in FIG. 20, it is recognized that the bottleneck drive unit BD in the selective interpolation operation is a drive unit in the y-axis direction (step S15 in FIG. 37).

  The speed change pattern from the starting point of the driving unit in the y-axis direction → the via point VP1 → the via point VP2 corresponds to case B (step S16 in FIG. 37). Therefore, the y-axis coordinate of the waypoint VP1 is changed in the direction of the starting point (step S17 in FIG. 37) to obtain a changed improvement trajectory.

  However, since the change improvement trajectory interferes with the obstacle E2, the operation cannot be executed (NO in step S18 in FIG. 37).

  Therefore, a search process is performed by the operable via point candidate searching means 6, and a via point of the driving unit in the x-axis direction which is a driving unit (other selective driving unit ED) other than the bottleneck driving unit BD in order to avoid interference. An attempt is made to change the coordinates in VP1 (step S33 in FIG. 37).

  Here, the driving unit in the x-axis direction is the bottleneck driving unit BD by the interpolation operation of the via point VP1 → the via point VP2, and the speed change pattern of the starting point → the via point VP1 → the via point VP2 corresponds to the case B. . Therefore, the coordinates of the driving unit in the x-axis direction at the via point VP1 are changed to the direction of the via point VP2, and an improved trajectory after search is obtained (step S34 in FIG. 37). However, interference cannot be avoided in the improved trajectory after the search (NO in step S35). The same applies to other via points and other interpolation operations, and since the execution time cannot be further shortened, the improvement ends (YES in step S31 in FIG. 37 → NO in step S32 → NO in step S12). S13).

  Note that when the coordinates of a via point, which is a process of improvement, is on a trajectory connecting teaching point coordinates before and after the via point by an interpolation operation, the via point coordinates are deleted. That is, step S21 in FIG. 37 is executed for the coordinate point after the change after YES in step S18 or after YES in step S35. If the above condition is satisfied, the via point after the change is deleted. Then, it returns to step S12.

  As described above, by executing the process of step S21, when the changed waypoint coordinates are on the trajectory connecting the teaching point coordinates before and after the waypoint by the interpolation operation, the waypoint coordinates are deleted. This has the effect of reducing the number of calculations.

(Third example of route point candidate determination processing)
FIGS. 21 to 28 are diagrams showing a third example of the waypoint candidate determination process by waypoint candidate determination means 5. 21, FIG. 23, FIG. 25, and FIG. 27 are explanatory diagrams schematically showing changes in the hand position from the start point to the end point including two via points. FIG. 22, FIG. 24, FIG. 26, and FIG. FIG. 28 is a graph showing changes in speed of two drive units that move parallel to the x-axis direction and the y-axis direction during movement of the hands of FIGS. 21, 23, 25, and 27;

  In FIG. 21, FIG. 23, FIG. 25, and FIG. 27, the hand position P31, which is the start point coordinate, passes through the hand position P32, which is the via point VP1, and the hand position P33, which is the via point VP2. Indicates moving. That is, the hand of the robot apparatus 2 moves in the order of P31 → P32 → P33 → P34. The object to be moved by the hand is the workpiece W3, and an obstacle E3 exists between the hand positions P31 to P34.

  The route point candidate determination process by the route point candidate determination unit 5 when the initial trajectory shown in FIGS. 21 and 22 is given by the initial information I1 will be described below in association with the flowchart shown in FIG.

  In the initial trajectory shown in FIGS. 21 and 22, attention is paid to the interpolation operation of the via point VP1 (hand position P32) and the via point VP1 → the via point VP2 (hand position P33) (step S14 in FIG. 37). In other words, the via point VP1 is the selected via point, and “via point VP1 → the via point VP2” is the selective interpolation operation. Based on the operation information shown in FIG. 22 regarding the initial trajectory, it is recognized that the bottleneck drive unit BD is the drive unit in the x-axis direction in the selective interpolation operation (step S15 in FIG. 37).

  Here, the speed change pattern at the start point of the driving unit in the x-axis direction → the via point VP1 → the via point VP2 corresponds to the case B ′ (step S16 in FIG. 37). Accordingly, the x-axis coordinate of the via point VP1 is changed to the coordinate direction of the via point VP2 (step S17 in FIG. 37), and the second trajectory shown in FIGS. 23 and 24 is obtained as the changed trajectory.

  Since the second trajectory does not interfere with the obstacle E3, YES is obtained in step S18 of FIG. Then, the process returns to step S12 without deleting the waypoint in step S21. Since there is a possibility of improvement in the trajectory from the second trajectory, NO is returned in step S12, and the process returns to step S14 in FIG. Hereinafter, the same explanation about correspondence with Drawing 37 is omitted suitably.

  When the second trajectory is obtained, attention is paid to the interpolation operation of the transit point VP2 and the transit point VP1 → the transit point VP2. In other words, the via point VP2 is the selected via point, and “via point VP1 → the via point VP2” is the selective interpolation operation. Based on the operation information shown in FIG. 24 for the second trajectory, it is recognized that the bottleneck drive unit BD in the selective interpolation operation is the drive unit in the x-axis direction.

  Here, the speed change pattern in the drive unit in the x-axis direction from the transit point VP1 → the transit point VP2 → the end point (hand position P34) corresponds to the case B. Therefore, the x-axis coordinate of the via point VP2 is changed to the coordinate direction of the via point VP1, and the third trajectory shown in FIGS. 25 and 26 is obtained as the changed trajectory.

  Since the third trajectory does not interfere with the obstacle E3, YES is obtained in step S18 of FIG.

  In step S21, the transit point VP1 and the transit point VP2 overlap, that is, the transit point VP2 exists on the trajectory obtained by connecting the coordinates of the start point to the transit point VP2, which are trajectory coordinates before and after the transit point, by an interpolation operation. The via point VP2 (hand position P33) in the third trajectory is deleted, and the third trajectory is corrected. Then, it returns to step S12. Since there is a possibility of improvement in the trajectory from the third trajectory, NO is returned in step S12, and the process returns to step S14 in FIG.

  When the corrected third trajectory is obtained in step S21, attention is paid to the interpolation operation of the via point VP1 and the start point → the via point VP1. That is, the via point VP1 is the selected via point, and “start point → via point VP1” is the selective interpolation operation. Based on the operation information shown in FIG. 26 for the third trajectory, it is recognized that the bottleneck drive unit BD in the selective interpolation operation is the drive unit in the y-axis direction.

  Here, the speed change pattern from the start point of the driving unit in the y-axis direction → the via point VP1 → the end point corresponds to the case C. The y-axis coordinate of the via point VP1 is changed in the starting point coordinate direction, and the fourth trajectory shown in FIGS. 27 and 28 is obtained as the trajectory after the change.

  When the fourth trajectory is obtained, based on the operation information shown in FIG. 28, it is the drive unit in the x-axis direction that becomes the bottleneck drive unit BD in all the interpolation operations from the start point → the via point VP1 → the end point. Since the speed change pattern corresponds to case A, no further reduction in the execution time can be expected, and the improvement ends (NO in step S12 in FIG. 37). As a result, the learning of the trajectory improvement of the robot apparatus 2 ends (step S13 in FIG. 37).

  In this way, by executing the process of step S21, when the changed via point coordinates are on the trajectory connecting the teaching point coordinates before and after the via point by the interpolation operation, the via point coordinates can be deleted. This is effective in reducing the number of calculations.

(Fourth example of route point determination process)
FIGS. 29 to 36 are diagrams showing a fourth example of the waypoint candidate determination process by waypoint candidate determination means 5. 29, 31, 33, and 35 are explanatory diagrams schematically showing changes in the hand position from the start point to the end point including two via points. FIGS. 30, 32, 34, and 36 FIG. 36 is a graph showing changes in speed of two drive units that move parallel to the x-axis direction and the y-axis direction in the movement of the hand of FIGS. 29, 31, 33, and 35. FIG.

  29, 31, 33, and 35, the hand position P41 that is the start point coordinate moves to the hand position P44 that is the end point coordinate via the hand position P42 that is the via point VP1. That is, the hand of the robot apparatus 2 moves in the order of P41 → P42 → P44. The object to be moved by the hand is the workpiece W4, and an obstacle E4 exists between the hand positions P41 to P44.

  FIG. 38 is a flowchart showing the processing contents of the robot trajectory control method in the robot trajectory control apparatus according to the first embodiment, corresponding to the fourth example.

  As shown in the figure, a waypoint addition process of step S22 executed after step S32 is added. That is, in step S22, when the speed change pattern corresponds to case C but no improvement is expected due to interference with an obstacle, the waypoint candidate determination unit 5 performs a process of adding a waypoint.

  After execution of step S22, the process proceeds to step S12. Note that step S22 is executed after YES in step S18 and also after execution of YES in step S35, but step S22 is substantially bypassed because the condition of step S22 is not satisfied. . The other processes are the same as those in the flowchart shown in FIG.

  Hereinafter, the waypoint candidate determination processing by waypoint candidate determination means 5 when the initial trajectory shown in FIGS. 29 and 30 is given by the initial information I1 will be described in association with the flowchart shown in FIG. 38 as appropriate.

  In the initial trajectory shown in FIG. 29 and FIG. 30, attention is paid to the interpolation operation of the via point VP1 and the start point → the via point VP1 (step S14 in FIG. 38). That is, the via point VP1 is the selected via point, and “start point → via point VP1” is the selective interpolation operation. Based on the operation information shown in FIG. 30, it is recognized that the bottleneck drive unit BD in the selective interpolation operation is a drive unit in the y-axis direction (step S15 in FIG. 38).

  Here, the speed change pattern from the start point of the driving unit in the y-axis direction → the via point VP1 → the end point corresponds to the case C (step S16 in FIG. 38). Accordingly, the y-axis coordinate of the via point VP1 is changed in the starting point coordinate direction to obtain the second trajectory as the changed trajectory.

  However, in the second trajectory, since it interferes with the obstacle E4 (NO in step S18), the execution time cannot be shortened as it is, and the improvement ends (finally YES in step S31, execution of step S32). . Therefore, the process of step S22 is executed after the execution of step S32.

  In step S22, since the speed change pattern corresponds to case C and improvement could not be expected due to interference with the obstacle E4, a via point is added to obtain the corrected second trajectory shown in FIGS. As shown in FIGS. 31 and 32, a hand position P43 is added as a via point VP2 between the hand positions P42 to P44.

  Then, it returns to step S12. Since there is a possibility of the trajectory improvement from the corrected second trajectory, NO is determined in the step S12, and the process returns to the step S14 in FIG. Hereinafter, the same explanation about correspondence with Drawing 38 is omitted suitably.

  In the modified second trajectory shown in FIGS. 31 and 32, attention is paid to the interpolating operation of the via point VP1 and the start point → the via point VP1. That is, the via point VP1 is the selected via point, and “start point → via point VP1” is the selective interpolation operation. Based on the operation information shown in FIG. 32, it is recognized that the bottleneck drive unit BD in the selective interpolation operation is a drive unit in the y-axis direction.

  Here, the speed change pattern at the start point of the driving unit in the y-axis direction → the transit point VP1 → the transit point VP2 corresponds to the case C. Accordingly, the y-axis coordinate of the via point VP1 is changed in the starting point coordinate direction, and the third trajectory is obtained as the changed trajectory.

  However, since the third trajectory in which only the y-axis coordinate is changed interferes with the obstacle E4 (NO in step S18 in FIG. 38), the x-axis coordinate of the via point VP1 is also changed from the start point coordinate in order to avoid interference. (Steps S33 and S34 in FIG. 38), the corrected third trajectory shown in FIGS. 33 and 34 is obtained.

  Thereafter, the improvement proceeds in the same manner, and after obtaining the fourth trajectory finally shown in FIGS. 35 and 36, the improvement ends.

  As described above, when the step S22 is executed and interference cannot be avoided by the changed via point coordinates, and the speed change pattern corresponds to the case C, the execution time is further shortened by adding the via point coordinates. The trajectory can be improved.

(Specific examples of robots)
FIG. 39 to FIG. 43 are explanatory diagrams showing specific postures of a robot that is a target for trajectory improvement. 39 to 41 and 43, (a) is a front view and (b) is a side view. In FIG. 42, (a) is a top view, (b) is a side view, and (c) is a front view. In addition, in order to clarify the correspondence with FIG. 3, in FIG. 39-FIG. 43, the position of the drive parts D1-D6, etc. are shown suitably.

  39 shows initial coordinates represented by the posture of the robot 21, end point coordinates represented by the posture of the robot 31 in FIG. 40, and initial via point coordinates represented by the posture of the robot 21 in FIG. The result of applying the waypoint candidate determination process of the first embodiment to the arrangement of the obstacle E1 as shown in FIG. 42 is the waypoint coordinates represented by the posture of the robot 21 in FIG.

  Teach point data such as start point coordinates, end point coordinates, and via point coordinates coincides with the position and orientation of the robot when expressed by the angle of the robot drive unit. 39 to 41 and 43, each coordinate is expressed not by the angle of the drive unit but by the position and orientation appearance of the robot realized at that time.

  In the first to fourth examples described above, for simplification of description, the coordinates are expressed in two dimensions, each coordinate is expressed by the hand position of the gripper K6, and interference with the obstacle E1 is also considered only by the hand portion. . Actually, since it is necessary to avoid interference between the entire robot and the obstacle, FIGS. 39 to 43 show the appearance of the robot.

  In the example of FIGS. 39 to 42, when the operation is performed without taking the via point from the start point coordinate to the end point coordinate, the arm of the robot interferes with the obstacle E1 because the arm moves so as to extend. Therefore, in order to avoid interference with the obstacle E1 and move from the start point coordinates to the end point coordinates, the hand position of the robot needs to pass on the near side (robot side) of the obstacle E1 or above the obstacle E1. There is.

  FIG. 44 is a graph showing the relationship between the number of operations of the robot apparatus 2 (robot 21) and the execution time. In the figure, the operation execution time is normalized by setting the operation execution time of the initial trajectory to “1”. When it is determined that the obstacle and the robot interfere with each other, it is determined that the operation cannot be executed, and the execution time is set to “0”.

  Since the initial via point coordinates shown in FIG. 41 are located sufficiently away from the obstacle, there is no interference with the obstacle due to the movement from the start point → the via point → the end point, but the execution time is long (see FIG. 43). Equivalent to the first).

  Therefore, the robot trajectory control apparatus according to the first embodiment performs trajectory control for changing the waypoint (in the example of FIGS. 39 to 43, there is one waypoint). Then, the robot trajectory control apparatus according to the first embodiment ends the trajectory improvement control after executing the 60th operation. That is, after the 60th execution, it is determined that a via point candidate that shortens the execution time and does not interfere with the obstacle cannot be found, and finally determines the coordinates (FIG. 43) executed at the 55th execution time that is the shortest execution time. Learning is terminated by determining the coordinates as via points.

  As a result, by changing the waypoint from the initial robot posture shown in FIG. 41 to the robot posture shown in FIG. 43, the execution time between the start point and the end point can be minimized without interfering with the obstacle E1. it can.

  As shown in FIG. 44, the robot trajectory control apparatus according to the first embodiment can finally obtain a trajectory whose execution time is improved to less than 40% from the initial trajectory. In addition, since the robot trajectory control apparatus according to the first embodiment uses only the waypoint coordinates for which the operation execution time is shortened, the trajectory can be improved with a small number of calculations.

  As described above, since the robot trajectory control apparatus according to the first embodiment uses the motion information I3 of the actual robot apparatus 2 to improve the trajectory, there is no influence of modeling errors in the robot apparatus or the surrounding environment, and high accuracy. This has the effect of improving the trajectory. That is, the robot trajectory control apparatus according to the first embodiment has an effect that the robot trajectory can be automatically improved with a small number of calculations.

(Supplementary method for calculating bottleneck driver BD)
(First case)
During one interpolation operation, one drive unit keeps trying to move in the shortest time, but each drive unit may start moving at the same time and the trajectory may be defined by an interpolation operation that does not necessarily end the operation at the same time. . In the first case, it is assumed that one drive unit continues to be a bottleneck and the speed pattern of the drive unit is trapezoidal or triangular. The trapezoid is a speed pattern in which a stable period due to the maximum speed exists, and the triangle means a speed pattern in which the speed immediately decreases when the maximum speed is reached.

  First, the waypoint candidate determination unit 5 calculates a drive unit that becomes a bottleneck at a certain time t as follows (first calculation method). The speed of the drive unit Di at time t obtained from the operation information I3 of the robot apparatus 2 is Vit, and the acceleration is Ait.

  In addition, the maximum speed (rated speed) of the drive unit Di predetermined for the robot apparatus 2 is defined as VMi, and the maximum acceleration (rated acceleration) is defined as AMi. The absolute value of the value obtained by dividing the velocity Vit by VMi is the coefficient Pit, the absolute value of the value obtained by dividing the acceleration Ait by the acceleration AMi is the coefficient Qit, and the coefficient Pit and the 1/2 power of the coefficient Qit are compared for each drive unit. The larger one is the coefficient Rit. The drive unit having the largest coefficient Rit among all the drive units is defined as a bottleneck drive unit BD at time t.

  As described above, since the drive unit that is the bottleneck at each time is known, the drive unit that is always the bottleneck during the interpolation operation can be calculated, and this drive unit is used as the bottleneck in the interpolation operation. . Subsequent track improvement is performed in the same manner as the processing described with reference to FIG.

  In this way, during one interpolation operation, one drive unit keeps trying to displace in the shortest time, but each drive unit does not necessarily start moving at the same time, and the trajectory is defined by the interpolation operation. Even in the first case, the robot trajectory control apparatus of the first embodiment can automatically improve the trajectory by adopting the first algorithm.

  As another method (second algorithm) of the first case, the interpolation operation is divided according to the time when the acceleration of one of the drive units changes, and the bottleneck at the intermediate time of the divided section is representative. Consider the method of calculating. This second algorithm has the effect of reducing the number of calculations compared to finding a bottleneck at regular intervals.

  Furthermore, there is the following method (third calculation method) as another bottleneck calculation method in the first case. In one interpolation operation, the displacement amount of the drive unit Di is set to ΔHi. However, when the moving direction of the drive unit Di changes during the interpolation operation, the displacement amount ΔHi is set to “0”. A coefficient value obtained by dividing the displacement amount ΔHi by the speed VMi is compared for each drive unit, and the drive unit having the largest count value is set as a bottleneck in the interpolation operation.

  This third algorithm also has the effect of reducing the number of calculations compared to the method of calculating bottlenecks at several times during the interpolation operation.

(Second case)
Since some of the drive units are overloaded, such as when a heavy object is gripped, special limits may be imposed on the speed and torque of the drive units. At this time, the drive unit that is actually limited becomes the drive unit that determines the overall execution time, but the speed pattern may deviate from the pattern in which one drive unit keeps moving in the shortest time. . That is, in the second case, one drive unit continues to be a bottleneck, but when the speed pattern of the drive unit is other than a trapezoid or a triangle, or in the above first to third calculations, other drive units are not used. It is assumed that it is a bottleneck.

  The operation information collecting means 3 also collects as operation information I3 whether the operation is a result of the operation of the drive unit being restricted due to the above-described restriction. If the operation of the drive unit is actually restricted, the waypoint candidate determination unit 5 determines the drive unit that is actually restricted as a bottleneck and determines the next route point candidate.

  As described above, the robot trajectory control apparatus according to the first embodiment is also applied to the second case in which speed and torque are limited in order to prevent an excessive load from being applied to some driving units such as when holding a heavy object. Can improve the trajectory.

(Third case)
There is an interpolation operation in which the trajectory of the hand position (hand portion) draws a straight line or a certain curve in the work space. In the case of such an interpolation operation, a drive unit that becomes a bottleneck may be changed during one interpolation operation.

  If the start point → route point (× 1) → end point is operated by the above-described interpolation operation, and the interpolation operation of “start point → route point” is the selected interpolation operation, the route point candidate determining means 5 is “start point → route point”. A drive unit that is a bottleneck in the immediate vicinity of the via point during the interpolation operation is regarded as a bottleneck of this interpolation operation. The bottleneck calculation method at this time is the same as the first to third calculation methods described in the first case. Then, based on the speed change pattern of the nearest part before and after the waypoint, the waypoint candidate is determined in the same way as the waypoint determination process described above.

  As described above, even in the third case where there is a trajectory defined by the interpolation operation in which the hand position draws a straight line or a certain curve in the work space, the robot trajectory control device according to the first embodiment automatically performs the trajectory. Improvements can be made.

(Fourth case)
In the robot trajectory control apparatus according to the first embodiment, there is a method of improving a waypoint candidate determination process and adding a waypoint when a drive unit that becomes a bottleneck is changed during one interpolation operation. In the interpolation operation of interest, it is conceivable to calculate a bottleneck at each time point as described in the first case, and add a new via point to the coordinates where the bottleneck is changed. In this way, the effect of improving the trajectory can be improved.

  Corresponding to the first to fourth cases described above, the normal operation by at least the robot trajectory control apparatus of the first embodiment can be maintained by appropriately changing the selection contents of the bottleneck driving unit BD.

  FIG. 45 is a flowchart showing the processing contents of the robot trajectory control method in the robot trajectory control apparatus according to the first embodiment when the waypoint addition processing is used to cope with the fourth case.

  As shown in the figure, the via point addition process of step S23 executed after YES in step S18 is added. That is, in step S23, via points are added to the coordinates where the bottleneck is changed in the selective interpolation operation. After executing step S23, the process proceeds to step S12. Note that step S23 is executed after YES in step S35 and also after step S32. The other processes are the same as those in the flowchart shown in FIG.

(Second apparatus configuration)
FIG. 46 is an explanatory diagram showing a specific device configuration (second device configuration) of the robot trajectory control apparatus according to the first embodiment of the present invention. As shown in the figure, a simulation for virtually operating a robot is implemented on an electronic computer 26. That is, the robot control means 1, robot apparatus 2, motion information collection means 3, action execution possibility determination means 4, via point candidate determination means 5, operable via point candidate search means 6 and via point change means 7 shown in FIG. Is realized on software processing implemented by the electronic computer 26.

  The second device configuration is a specific device configuration that realizes the robot operation itself by simulation. The simulation used here is a robot control device 22 that is a known technique, a robot model that takes into account dynamic characteristics, and a surrounding environment model that are reproduced on the computer 26 and visually displayed to the user. . In the simulation executed by the electronic computer 26, the internal sensor output and external sensor output of the robot can be reproduced, and the interference between the robot and the surrounding environment can also be reproduced.

  In the second apparatus configuration, the robot control means 1 and the robot apparatus 2 are also mounted on the electronic computer 26, and the robot 21, the robot control apparatus 22, and the teaching box 23 that are necessary in the first apparatus configuration shown in FIG. And the cable 25 for sensors becomes unnecessary.

  In the case of the second device configuration, since the robot is operated on the simulation mounted on the electronic computer 26 and it is not necessary to operate the actual robot, the trajectory is being improved as compared with the first device configuration. In addition, there is an effect that there is no danger of damaging the robot due to interference. In addition, there is no need to prepare an external sensor for improving the trajectory, so that the present invention can be easily applied.

(Third device configuration)
FIG. 47 is an explanatory diagram showing a specific device configuration (third device configuration) of the robot trajectory control device according to the first embodiment of the present invention.

  As shown in the figure, a robot control device 32 including a robot control means 1 is used, and a robot 31 corresponding to the robot device 2 operates according to a command value output from the robot control device 32. Outputs of internal sensors such as the operation status and execution time of each drive unit of the robot 31 are output to the robot control device 32. The output of the external sensor provided in the robot 31 is output to the robot control device 32 via the sensor cable 35.

  The robot control device 32 also performs calculation for determining a waypoint candidate for improving the trajectory. Specifically, the action information collecting means 3, the action feasibility determining means 4 shown in FIG. The functions of the point candidate determining means 5, the operable via point candidate searching means 6, and the via point changing means 7 are implemented.

  The teaching box 33 is a device that gives the initial information I1 to the robot control device 32 by the operator, like the teaching box 23 of the first device configuration shown in FIG.

  Since the load of calculation processing such as the waypoint candidate determination processing for applying the present invention is low and can be processed even by a robot controller that is inferior in numerical calculation processing capability compared to a dedicated electronic computer, the robot 31 It is sufficiently possible to provide the robot control device 32 itself for which the above control is performed with a function such as a waypoint determination process.

  In the third apparatus configuration, the operation information collection unit 3, the operation execution possibility determination unit 4, the via point candidate determination unit 5, the operable via point candidate search unit 6, and the via point change unit 7 are also mounted on the robot control device 32. Therefore, the configuration corresponding to the electronic computers 24 and 26 having the first and second device configurations is unnecessary, and the device configuration can be simplified.

  Further, since the sensor cable 35 is connected to the robot control device 32, when the robot trajectory control is performed, processing such as replacement of the sensor cable to the electronic computer is not required as in the first device configuration. Become. Therefore, the third apparatus configuration has an effect that labor can be reduced when the present invention is applied to a large-scale production facility in which a large number of robots exist.

<Embodiment 2>
FIG. 48 is a block diagram showing the configuration of the robot trajectory control apparatus according to the second embodiment of the present invention. As shown in the figure, when compared with the first embodiment shown in FIG. 1, the trajectory definition conversion means 8 is added, and the teaching point data I10 that is input from the outside is given to the trajectory definition conversion means 8. The point is different.

  The permutation of the teaching point data I10 is represented by X = {x0, x1, x2,..., Xn}, and the trajectory is such that the robot device 2 moves in sequence for each teaching point every unit time such as 7 msec. Is defined. Such a method of defining a trajectory is a known technique. Note that the permutation of the input teaching point data I10 is always a trajectory position where the robot can operate.

  Note that, in the permutation X of the teaching point data I10, the final purpose is to reach the end point coordinate xn from the initial coordinate x0, and intermediate waypoint coordinates (x1 to x (n-1)) (at least one initial waypoint) ) Is not necessary.

  A permutation X of the teaching point data I10 is input to the trajectory definition conversion means 8, and based on the teaching point data I10, conversion initial information I8 defining a plurality of trajectory coordinates including a start point coordinate, a predetermined number of conversion via points, and an end point coordinate. Is output. The conversion initial information I8 may include a type of interpolation operation for connecting a plurality of trajectory coordinates.

  The robot control means 1 operates the robot apparatus 2 based on the converted initial information I8. Then, based on the operation information I3 obtained from the operation information collection unit 3, the operation execution possibility determination unit 4 determines whether the operation can be executed, and outputs operation available information I4 indicating the determination result.

  When the robot control unit 1 outputs the operation command value C1 based on the conversion initial information I8 and causes the robot apparatus 2 to execute, the operation enable information I4 indicates that the initial operation can be executed. The trajectory is improved as in the first embodiment when the information I8 is the initial information I1.

  On the other hand, when the robot apparatus 2 is executed by the robot control means 1 based on the converted initial information I8, if the operable information I4 indicates that the initial operation cannot be performed, the operable information I4 indicating that the operation is disabled It is output to the definition conversion means 8.

  When the operable information I4 indicates that the operation of the robot apparatus 2 cannot be executed based on the converted initial information I8, the trajectory definition converting means 8 calculates (x1 to x (n) from the via point coordinates in the permutation X defined by the teaching point data I10. -1) Adds one via point coordinate as a new conversion via point and outputs new conversion initial information I8 again. The conversion initial information I8 is changed until the robot apparatus 2 based on the conversion initial information I8 can execute an operation.

  Then, the converted initial information I8 that is finally determined to be executable by the operation executable determination unit 4 is handled in the same manner as the initial information I1 of the first embodiment, and the operation executable determination is performed as in the first embodiment. The trajectory improvement processing by means 4 and waypoint candidate determination means 5 is performed.

  Hereinafter, a specific operation of the trajectory definition conversion means 8 in the robot trajectory control apparatus according to the second embodiment will be described.

  The trajectory definition conversion means 8 outputs, from the permutation X of the input teaching point data I10, conversion initial information I8 that defines the start point coordinates, the initial via points, the end point coordinates, and, if necessary, the interpolation operation type connecting them.

  Hereinafter, the conversion initial information I8 initially output by the trajectory definition conversion means 8 will be described. First, the first data x0 of the permutation X of the teaching point data I10 is set as the start point coordinate, and the last data xn of the permutation of the teaching point data is set as the end point coordinate.

  Next, assuming that each drive unit moves at a constant speed in the movement of each teaching point xi → x (i + 1), a speed change pattern when the robot apparatus 2 operates from x0 to xn is calculated. Then, the coefficient Rit shown in the first case of the first embodiment is calculated from the speed change pattern, and the bottleneck drive unit BD at each time point is determined.

  The driving unit that becomes the bottleneck is observed along the time series, and the teaching point data at which the bottleneck driving unit is changed is adopted as the conversion via point.

  Finally, the trajectory definition conversion means 8 defines the type of interpolation operation that connects the points with fear of necessity. As the interpolation operation type, the interpolation operation type used by the robot control means 1 when the permutation X of the teaching point data I10 is given by a known technique, for example, the drive unit Di starts to move simultaneously, and moves at the same speed as possible during the operation. Interpolation operation types in which the driving unit Di simultaneously ends operation are conceivable. Of course, other types of interpolation operation may be used. As a result, data having the same type as the initial information I1 of the first embodiment can be output to the robot control means 1 as the converted initial information I8.

  When the trajectory based on the conversion initial information I8 output by the trajectory definition conversion means 8 is determined by the motion feasibility determination means 4 to be unexecutable, the trajectory definition conversion means 8 performs the teaching point data as follows: One via point selected from the initial via points defined in I10 is added as a new conversion via point, and information including the added change via point is redetermined as new conversion initial information I8.

  It is assumed that the route point immediately before the position where it is determined that the operation cannot be performed is xi, and the route point immediately after is xj. At this time, “j> i + 1” is established. This is because, when “j = i + 1”, the trajectory of the portion of xi → xj coincides with the trajectory defined by the permutation X of the input teaching point data, so that the operation can always be executed. Therefore, the trajectory definition conversion means 8 selects teaching point data x [(i + j) / 2] (initial waypoint) that is intermediate between xi and xj and adds it as a change waypoint. [K] means the smallest integer equal to or greater than k.

  Thereafter, until the robot apparatus 2 based on the conversion initial information I8 determines that the operation can be executed by the operation execution capability determination unit 4, the trajectory definition conversion unit 8 adds conversion via points. Eventually, the coordinates by the permutation X of the input teaching point data are equivalent to the coordinates indicated by the conversion initial information I8, so that the operation can always be executed.

  Another trajectory conversion method by the trajectory definition conversion means 8 is the following method (second method).

  First, the first data x0 of the permutation X of the teaching point data I10 is set as the start point coordinate, and the last data xn of the permutation X of the teaching point data I10 is set as the end point coordinate.

  Next, teaching data x [n / 2] between x0 and xn is set as an initial via point as an initial conversion via point. The interpolation operation type is the same as in the first embodiment.

  When the trajectory of the robot apparatus 2 based on the converted initial information I8 output by the trajectory definition conversion means 8 is determined by the motion executable determination means 4 to be unable to execute the motion, the trajectory definition conversion means 8 is as follows. A new waypoint is added and an initial waypoint is determined.

  It is assumed that the route point immediately before the position where it is determined that the operation cannot be performed is xi, and the route point immediately after is xj. Therefore, the trajectory definition conversion means 8 adds the teaching point data x [(i + j) / 2] between xi and xj as a conversion via point. Thereafter, the trajectory definition conversion means 8 adds conversion via points until it is determined that the operation can be executed.

  The conversion contents of the conversion initial information I8 by the trajectory definition conversion means 8 by the robot trajectory control device of the above-described embodiment are summarized as follows.

  First step: Receive teaching point data I10 defining the coordinates of a starting point (x0), at least one initial via point (x1 to x (n-1)), and an ending point (xn).

  Second step: The teaching point data I10 is converted into conversion initial information I8.

  Third step: When the robot apparatus 2 is inoperable when operated by the trajectory definition conversion means 8 obtained in the second step, any one of the at least one initial waypoint is additionally selected as a waypoint for conversion. It is set as new conversion initial information I8.

  This third step is repeated until it is determined that the robot is operable when it is operated with the new conversion initial information I8.

  For example, a scene in which a robot is operated as a master and slave and the movement is taken as a trajectory, a movement that is performed by a human being is observed with a vision sensor, and a robot is imitated. The scene that defines the trajectory of can be considered.

  In the robot trajectory control apparatus according to the second embodiment, even if the teaching point data I10 is given from the outside and the trajectory is defined by the permutation of the teaching point data at regular intervals (several msec) as in the various scenes described above. Then, the number of via points in the teaching point data I10 can be converted in advance into the converted initial information I8 after being reduced to an operable number.

  As described above, the robot trajectory control apparatus according to the second embodiment improves the teaching point data I10 before the trajectory improvement process is performed by the waypoint candidate determination unit 5 and the operable waypoint candidate search unit 6. The conversion initial information I8 can be automatically created.

  As a specific device configuration for realizing the robot trajectory control device of the second embodiment, the first device configuration shown in FIG. 2 (the trajectory definition conversion means 8 is mounted in the robot control device 22 or the electronic computer 24). ) Is considered. 46 (the trajectory definition conversion means 8 is mounted in the electronic computer 26), or the third apparatus configuration shown in FIG. 47 (the trajectory definition conversion means 8 is in the robot controller 32). May be a specific device configuration of the robot trajectory control device of the second embodiment.

It is a block diagram which shows the structure of the robot track control apparatus which is Embodiment 1 of this invention. It is explanatory drawing which shows the specific apparatus structure (1st apparatus structure) which implement | achieves the robot track control apparatus of Embodiment 1. FIG. It is explanatory drawing which shows the specific example of the robot shown in FIG. 3 is a flowchart showing processing contents of a robot trajectory control method in the robot trajectory control apparatus according to the first embodiment. It is a flowchart which shows the selection method of the bottleneck drive part by the waypoint candidate determination means shown in FIG. It is explanatory drawing represented with the characteristic speed change pattern about a bottleneck drive part. It is a figure which shows the 1st example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 1st example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 1st example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 1st example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 1st example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 1st example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 1st example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 1st example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 1st example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 1st example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 2nd example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 2nd example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 2nd example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 2nd example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 3rd example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 3rd example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 3rd example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 3rd example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 3rd example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 3rd example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 3rd example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 3rd example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 4th example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 4th example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 4th example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 4th example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 4th example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 4th example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 4th example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a figure which shows the 4th example of the waypoint candidate determination process by the waypoint candidate determination means shown in FIG. It is a flowchart which shows the processing content of the robot track control method in the robot track control apparatus of Embodiment 1 corresponding to the 2nd example and 3rd example of a waypoint candidate determination process. It is a flowchart which shows the processing content of the robot track control method in the robot track control apparatus of Embodiment 1 corresponding to the 4th example of a waypoint candidate determination process. It is explanatory drawing which shows the specific attitude | position of the robot used as a track improvement object. It is explanatory drawing which shows the specific attitude | position of the robot used as a track improvement object. It is explanatory drawing which shows the specific attitude | position of the robot used as a track improvement object. It is explanatory drawing which shows the specific attitude | position of the robot used as a track improvement object. It is explanatory drawing which shows the specific attitude | position of the robot used as a track improvement object. It is a graph which shows the relationship between the frequency | count of operation | movement of a robot apparatus, and execution time. It is a flowchart which shows the processing content of the robot track control method in the robot track control apparatus of Embodiment 1 at the time of using a via point addition process. It is explanatory drawing which shows the specific apparatus structure (2nd apparatus structure) of the robot track control apparatus of Embodiment 1. FIG. It is explanatory drawing which shows the specific apparatus structure (3rd apparatus structure) of the robot track control apparatus of Embodiment 1. FIG. It is a block diagram which shows the structure of the robot track control apparatus which is Embodiment 2 of this invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 Robot control means 2 Robot apparatus 3 Motion information collection means 4 Action execution possibility determination means 5 Via point candidate determination means 6 Operable via point candidate search means 7 Via point change means 8 Trajectory definition conversion apparatus 21, 31 Robot, 22, 32 Robot controller, 23, 33 Teaching box, 24, 26 Computer, 25, 35 Sensor cable, I1 initial information, I8 conversion initial information, I10 Teaching point data.

Claims (4)

A robot trajectory control device that controls a trajectory that is an operation path from a start point to an end point of a robot having a plurality of drive units,
Based on trajectory position information that defines a plurality of trajectory coordinates including a start point, at least one via point, and an end point, an operation that defines the contents of a plurality of interpolation operations that are operations of the plurality of drive units between the plurality of trajectory coordinates Comprising a robot control means for outputting a command value, each of the plurality of trajectory coordinates is defined by the operation content of the plurality of drive units,
Based on the operation command value, includes drive unit information indicating a state of each of the plurality of drive units realized when the robot is operated, and peripheral information that defines a positional relationship between the robot and the surrounding environment. Operation information collecting means for collecting operation information;
Based on the motion information collected by the motion information collection means, motion executable determination means for determining whether the robot is operable at the motion command value;
When it is determined that the robot is operable by the operation executable determination unit, one via point of the at least one via point is selected as a selection via point, and the selected via point among the plurality of interpolation operations One interpolation operation including a selected interpolation operation, a drive unit determined to have the most influence on the execution speed among the plurality of drive units in the selected interpolation operation is a bottleneck drive unit, and a robot based on the operation information calculating a first transit point candidates to the extent that the total execution time is shortened by changing the coordinates of the bottleneck the selected route point to definitive to the drive unit, the first which defines a transit point candidates of the first Route point candidate determining means for outputting route point candidate information;
When the robot is determined to be inoperable by the motion executable determination unit, based on the motion information, a range other than the bottleneck driver among the plurality of drive units within a range where the execution time of the entire robot is shortened of calculating a second via point candidates definitive the drive unit, and operable via point candidate search means for outputting a second via point candidate information specifying a route point candidate of the second,
If the second waypoint candidate specified by the second waypoint candidate information exists, the second waypoint candidate is determined as a change waypoint, and the second waypoint candidate does not exist and the second waypoint candidate does not exist. If the first waypoint candidate specified by one waypoint candidate information exists, the first waypoint candidate is determined as the change waypoint, and the selected waypoint among the plurality of trajectory coordinates is changed. A route point changing unit that provides the robot control unit with the information replaced by the route point as new trajectory position information;
Robot trajectory control device. The robot trajectory control device according to claim 1,
It further includes trajectory definition conversion means for receiving teaching point data defining coordinates of a starting point, at least one initial via point, and an ending point, and converting the teaching data into conversion initial information, wherein the conversion initial information includes the starting point, the at least one A plurality of trajectory coordinates including a predetermined number of conversion via points and end points selected from two initial via points are defined, and the robot is operated by the robot control unit based on the conversion initial information, and obtained from the operation information collecting unit. Whether or not the robot can be operated is determined by the operation execution determination unit based on the operation information.
The trajectory definition conversion means, when the operation feasibility determination means determines that the operation is not possible, additionally selects any one of the at least one initial via point as the predetermined number of conversion via points to obtain new conversion initial information. Sequentially output, and the conversion initial information at the time when the operation executable determination unit determines that the operation is possible, as the trajectory position information,
Robot trajectory control device. A robot trajectory control method for controlling a trajectory that is an operation path from a start point to an end point in a robot having a plurality of drive units,
(a) comprising a step of obtaining trajectory position information defining a plurality of trajectory coordinates including a start point, at least one waypoint, and an end point, each of the plurality of trajectory coordinates being defined by the operation contents of the plurality of driving units,
(b) One of the at least one waypoints is set as a selected waypoint, and the selected waypoint among a plurality of interpolation operations that are operations of the plurality of driving units between the plurality of trajectory coordinates. and selecting interpolation operation an interpolation operation including, a drive unit which determines that the greatest impact on execution speed of the plurality of drive unit in said selected interpolation operation is a bottleneck driver, reduces execution time of the entire robot calculating a first route point candidates by changing the coordinates of the selected waypoint to definitive the bottleneck driver in a range that,
(c) Among the plurality of trajectory coordinates, the robot operates when information obtained by replacing the first via point candidate obtained in step (b) with the selected via point is operated with the trajectory position information. is executed if the inability calculating a second via point candidates definitive the drive unit other than the bottleneck driver of the plurality of drive unit to the extent that the execution time of the entire robot is shortened,
(d) If the second waypoint candidate exists, the second waypoint candidate is determined as a change waypoint, and the second waypoint candidate does not exist and the first waypoint candidate exists. For example, determining the first via point candidate as the change via point, and setting the information obtained by replacing the selected via point with the change via point among the plurality of trajectory coordinates as new trajectory position information. Prepared,
Repeating steps (a) to (d) until both the first and second waypoint candidates cannot be calculated;
Robot trajectory control method. The robot trajectory control method according to claim 3,
(e) receiving teaching point data defining coordinates of a starting point, at least one initial waypoint, and an end point;
(f) converting the teaching data into conversion initial information, wherein the conversion initial information includes a plurality of trajectories including a start point, a predetermined number of conversion via points selected from the at least one initial via point, and an end point. Define the coordinates,
(g) When the robot is inoperable when operated with the conversion initial information obtained in step (f), any one of the at least one initial via point is added as the predetermined number of conversion via points. Selecting the new conversion initial information to select,
Step (g) is repeated until it is determined that the robot is operable when operated with the new conversion initial information,
The steps (e) to (g) are executed prior to the steps (a) to (d), and the final converted initial information becomes the trajectory position information.
Robot trajectory control method.

Images