CN116997441A

CN116997441A - Robot control device, robot control program, and robot control method

Info

Publication number: CN116997441A
Application number: CN202180093362.8A
Authority: CN
Inventors: 春尾七星; 冈原卓矢
Original assignee: Mitsubishi Electric Corp
Current assignee: Mitsubishi Electric Corp
Priority date: 2021-03-24
Filing date: 2021-03-24
Publication date: 2023-11-03
Also published as: DE112021007395T5; JP7438453B2; WO2022201362A1; JPWO2022201362A1

Abstract

A reference orbit (202) and a branching point (204) corresponding to the measured position and posture of a target object are acquired, the measured position and posture of the target object are set as an end point (Pe), a 1 st operation orbit, which is an operation orbit of the robot (101) from a start point (Ps) to the branching point (204), is generated directly by using the acquired reference orbit (202), a 2 nd operation orbit, which is an operation orbit of the robot (101) from the branching point (204) to the end point (Pe), is generated by performing a 1 st operation, and the robot (101) is controlled to be driven according to an operation instruction including the 1 st operation orbit and the 2 nd operation orbit.

Description

Robot control device, robot control program, and robot control method

Technical Field

The present invention relates to a robot control device, a robot control program, and a robot control method for moving an articulated robot from a start point to an end point.

Background

In a normal point-to-point control for operating a robot from a start point to an end point, in order to reduce processing during execution of a job and to increase the calculation speed, an operation command is sometimes generated before execution of the job, in consideration of restrictions concerning the robot body and the surrounding environment. In order to increase productivity, it is desirable to create a track in which an evaluation value for quantitatively expressing productivity such as operation time falls within a certain range when creating a track before execution of a job. When the position and posture of the target object in the robot work are not fixed, it is necessary to generate an operation trajectory in accordance with the actual position and posture of the target object. Since it takes a long time to generate an action track satisfying all restrictions, a measure of shortening the time taken for track generation using information of an already generated action track is widely performed.

In the technique described in patent document 1, it is assumed that an operation track for performing an operation by passing through a plurality of operation areas in a certain order is generated in advance. The operation track is divided into a plurality of operation area sections. When it is determined that a part of the operation track interferes with the obstacle due to the arrangement or shape change of the surrounding obstacle, the track is regenerated only for the operation region section where the interference occurs.

Patent document 1: japanese patent No. 6560841

Disclosure of Invention

In patent document 1, the track of a specific work area section where interference occurs can be changed, but the section to be changed is fixed in a section where interference with an obstacle is determined to occur. However, if the section to be changed is not properly set when the position and posture of the target object are different, the evaluation value may be deteriorated beyond an allowable range when the connection with the section without change is considered. For example, if the section to be changed is short when the fingertip of the robot needs to be greatly tilted as compared with the track generated in advance, the waiting time for the fingertip changing operation needs to be long.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide a robot control device capable of shortening the time for generating a trajectory in which an evaluation value falls within a certain range even when the position and posture of a target object fluctuate.

In order to solve the above-described problems and achieve the object, a robot control device according to the present invention moves an end effector of a robot from a start point to an end point, and performs a predetermined task on a target object whose position and posture are not fixed. The robot control device includes: a storage unit that stores a reference rail and a branching point in correspondence with each of a plurality of sets of positions and postures that can be obtained by the target object; a measuring unit that measures the position and posture of the target object when the task is executed; an individual trajectory generation unit that acquires a reference trajectory and a branch point corresponding to the measured position and posture of the target object from the storage unit, sets the measured position and posture of the target object as an end point, generates a 1 st operation trajectory, which is an operation trajectory of the robot from the start point to the branch point, using the acquired reference trajectory, and generates a 2 nd operation trajectory, which is an operation trajectory of the robot from the acquired branch point to the end point; and a robot control unit that controls driving of the robot in accordance with an operation command including the 1 st operation track and the 2 nd operation track. The reference trajectory is an operation trajectory of the robot from the start point to the 1 st point, which is one of a plurality of positions and postures that can be obtained from the target object, and is an operation trajectory that can avoid interference with an obstacle and satisfy the condition that the evaluation value falls within the first range. The branching point is a point on the reference trajectory, and is an operation trajectory in which the evaluation value of the operation trajectory from the branching point to the 1 st point falls within a second range that is larger than the first range.

ADVANTAGEOUS EFFECTS OF INVENTION

According to the present invention, even when the position and posture of the target object fluctuate, the time for generating the trajectory whose evaluation value falls within a certain range can be shortened.

Drawings

Fig. 1 is a conceptual diagram showing a configuration of a robot system in embodiment 1.

Fig. 2 is a plan view conceptually showing a rotational displacement in the plane of the target object in embodiment 1.

Fig. 3 is a conceptual diagram showing a plurality of candidates of the hand approach path in embodiment 1.

Fig. 4 is a block diagram showing the overall configuration of the robot control device according to embodiment 1.

Fig. 5 is a conceptual diagram for explaining the target set in embodiment 1.

Fig. 6 is a diagram for explaining the definition of the position in the target set in embodiment 1.

Fig. 7 is a diagram for explaining a plurality of modes in which rotation deviation angles of the target object are different in embodiment 1.

Fig. 8 is a diagram for explaining a plurality of modes in which the hand approach path to the target object in embodiment 1 is different.

Fig. 9 is a flowchart showing the operation of the track generation unit that adds the reference track and the branch point from the target set in embodiment 1 to the individual branch point storage unit.

Fig. 10 is a conceptual diagram showing an example of selection of the position of the target object used for generating the reference trajectory in embodiment 1.

Fig. 11 is a conceptual diagram showing another alternative example of the position of the target object used for generating the reference trajectory in embodiment 1.

Fig. 12 is a conceptual diagram showing another alternative example of the position of the target object used for generating the reference trajectory in embodiment 1.

Fig. 13 is a conceptual diagram showing a relationship between a reference trajectory and a branching point in embodiment 1.

Fig. 14 is a flowchart showing a procedure for calculating a branch point in embodiment 1.

Fig. 15 is a conceptual diagram showing a relationship between a reference rail and an obstacle in embodiment 1.

Fig. 16 is a conceptual diagram relating to the selection range of the branching point in embodiment 1.

Fig. 17 is a conceptual diagram showing a method of generating a track by the individual track generating unit in embodiment 1.

Fig. 18 is a flowchart showing a processing procedure of the individual track generation unit and the robot control unit in embodiment 1.

Fig. 19 is a flowchart showing an operation procedure of generating individual tracks in embodiment 2.

Fig. 20 is a conceptual diagram showing a reference track and individual tracks in embodiment 2.

Fig. 21 is a block diagram showing a hardware configuration of the robot control device according to embodiment 1 and embodiment 2.

Detailed Description

The robot control device, the robot control program, and the robot control method according to the embodiments will be described in detail below with reference to the drawings.

Embodiment 1.

Fig. 1 is a conceptual diagram showing a configuration of a robot system 100 according to embodiment 1. As shown in fig. 1, the robot system 100 includes: a multi-joint robot 101 having a plurality of axes and a plurality of arms; an end effector 102 mounted on the front end of the arm of the robot 101; a belt conveyor 103; and an external sensor 104. The robot 101 moves the end effector 102 from the start point Ps to the end point Pe to perform a predetermined task. The end point Pe is not fixed and is defined based on the position and posture of the target object 200 flowing on the belt conveyor 103. The external sensor 104 as a measuring unit acquires the position and posture of the target object 200 immediately before the actual operation of the robot 101.

The belt conveyor 103 performs a conveying operation in a constant direction based on the moving speed V [ mm/s ], and no jig for fixing the target object 200 is present on the belt conveyor 103. Therefore, the relative position and posture of the target object 200 with respect to the robot 101 are different each time.

Fig. 2 is a plan view conceptually showing a rotational displacement in the plane of the target object 200 in embodiment 1. Fig. 3 is a conceptual diagram showing a plurality of candidates of the hand approach route in embodiment 1.

In fig. 2, the stacked surface of the belt conveyor 103 is viewed from above. Since the belt conveyor 103 has no mechanism for fixing the target object 200, the target object 200 is rotationally offset with respect to a normal vector of the stacking surface of the belt conveyor 103, and a vector perpendicular to the paper surface is used as a rotation axis. This rotational offset is referred to as an in-plane rotation with respect to the belt conveyor 103, and its rotational angle is referred to as a rotational offset angle.

When the target object 200 is assumed to be an amorphous object, if the trajectory for bringing the end effector 102 close to the target object 200 is not properly selected according to the shape of the target object 200, the achievement of the task may become difficult. In the present embodiment, as shown in fig. 3, a plurality of hand approaching paths, which are tracks for approaching the end effector 102 to the target object 200, are prepared in advance for the end effector 102, and one hand approaching path capable of achieving a task is selected according to the shape of the target object 200. Fig. 3 shows 2 hand approaching paths including a hand approaching path HA and a hand approaching path HB.

In embodiment 1, fluctuation of the position and posture of the target object 200 with respect to the robot 101 is regarded as one pattern by a set of 3 elements including positional displacement in the three-dimensional translational direction, in-plane rotation with respect to the belt conveyor 103, and a hand approach path selected based on the shapes of the end effector 102 and the target object 200.

Fig. 4 is a block diagram showing the overall configuration of the robot control device according to embodiment 1. The robot control device includes a peripheral device information acquisition unit 301, a task information acquisition unit 302, a target set setting unit 303, a track generation unit 310 including a reference track generation unit 304 and a branch point specification unit 305, an individual branch point storage unit 306 as a storage unit, a sensor information acquisition unit 307, an individual track generation unit 308, and a robot control unit 309.

The peripheral device information acquisition unit 301 acquires peripheral device information, which is information related to the peripheral devices of the robot 101. The peripheral device information is the moving speed V of the belt conveyor 103. The task information acquisition unit 302 acquires task information, which is information related to a task for performing track generation. The task information includes at least the type of the end effector 102 used, the type of the target object 200, and the target operation of the task. The target operation of the task includes gripping, arrangement, and the like of the target object 200.

The target set setting unit 303 acquires the peripheral device information, that is, the moving speed V of the belt conveyor 103, from the peripheral device information acquiring unit 301. The target set setting unit 303 obtains the task information from the task information obtaining unit 302.

The target set setting unit 303 calculates a set of positions where the target object 200 can exist, that is, a target set 201, and a combination of the posture where the target object 200 inside the target set 201 can exist and the hand approach path, using the movement speed V of the belt conveyor 103 and the task information.

Fig. 5 is a conceptual diagram for explaining the target set 201 in embodiment 1. The method of calculating the target set 201 in the present embodiment will be described. In the present embodiment, the target set 201 shown in fig. 5 is calculated based on the moving speed V of the belt conveyor 103. If the upper limit value of the time (operation time) taken by the robot 101 for the operation of the target object 200 is set to T s, the length of the target set 201 in the conveyor moving direction can be set to t×v. The length of the target set 201 in the direction perpendicular to the conveyor moving direction can be set to, for example, the width W [ mm ] of the belt conveyor 103. The operation time of the robot 101 with respect to the target object 200 is, for example, a time from when the robot 101 moves from the start point Ps to the end point Pe to when execution of a predetermined task is completed. In the present embodiment, the fluctuation in the height direction of the conveyor is not considered, but the three-dimensional target set 201 may be generated in consideration of the fluctuation in the height direction of the conveyor.

The range of the target set 201 may be directly specified by the user irrespective of the moving speed V of the belt conveyor 103 as a limit of the work area assumed by the user of the robot system 100.

Next, a method of calculating the fluctuation pattern of the target object 200 included in the target set 201 will be described. Fig. 6 is a diagram for explaining the definition of the position in the target set 201 in embodiment 1. Fig. 7 is a diagram for explaining a plurality of modes in which rotation deviation angles of the target object 200 in embodiment 1 are different. Fig. 8 is a diagram for explaining a plurality of modes in which the hand approach path to the target object 200 in embodiment 1 is different.

As shown in fig. 6, the space in the target set 201 is divided into a plurality of pieces of space in an arbitrary size, and grid points G are defined at the center positions of the divided pieces of grid. Each grid point G has position information indicating the position of the grid point. The position of the translational shift of the target object 200 is specified by specifying the grid point G. In fig. 6, the target set 201 is divided into N1 (=25) grids in a matrix form, and has 25 grid points G. In fig. 7, N2 (=3) modes in which the rotation deviation angles of the in-plane rotation of the target object 200 are different are shown. In fig. 8, different N3 (=3) hand approach paths HA, HB, HC are shown.

Regarding each grid point G, since the change of N2 patterns with different rotation deviation angles of the in-plane rotation of the target object 200 and the change of N3 patterns with different hand approach paths are considered, there are maximum n2×n3 wave patterns for one grid point G. Regarding N2, N3, the number greater than or equal to 1 is set in advance. The scale value of the rotation deviation angle can be arbitrarily determined. Accordingly, the fluctuation pattern of the total n1×n2×n3 target objects 200 is set in the target set 201, and the trajectory generation unit 310 calculates branching points for each of the n1×n2×n3 fluctuation patterns.

Next, the track generation unit 310 shown in fig. 4 will be described. The track generation unit 310 is composed of a reference track generation unit 304 and a branch point specification unit 305.

Fig. 9 is a flowchart showing the operation of the track generation unit 310 that adds a reference track and a branching point from the target set 201 in embodiment 1 to the individual branching point storage unit 306. Fig. 10 is a conceptual diagram showing an example of selection of the position of the target object 200 used for generating the reference trajectory in embodiment 1. Fig. 11 is a conceptual diagram showing another alternative example of the position of the target object 200 used for generating the reference trajectory in embodiment 1. Fig. 12 is a conceptual diagram showing another alternative example of the position of the target object 200 used for generating the reference trajectory in embodiment 1.

First, the reference trajectory generation unit 304 reads the target set 201 calculated by the target set setting unit 303 and all the fluctuation patterns of the target objects 200 included in the target set 201 (S101).

Next, the reference orbit generation unit 304 selects a fluctuation mode having a part of grid points G among the plurality of fluctuation modes, and generates a reference orbit having an evaluation value within a first range using the selected fluctuation mode (S102). That is, the reference orbit generation unit 304 generates a reference orbit having an evaluation value within a first range by using an arbitrary orbit generation method with a predetermined start point Ps as an operation start point and at least one specific fluctuation pattern within the target set 201 as an end point Pe.

In the present embodiment, the operation time is selected as the evaluation value. As the first range, the operation time is set to be within T1 s. Therefore, a reference track having an operation time of T1 or less is generated. As the evaluation value, a predicted lifetime of the robot 101 when the operation of the robot 101 with respect to the target object 200 is repeatedly performed, power consumed by the robot 101 when the operation of the robot 101 with respect to the target object 200 is performed, and the like may be used.

The fluctuation mode selected by S102 may be arbitrarily selected, but as shown in fig. 10, for example, the fluctuation mode of the grid point G of the grid having the center position of the target set 201 as the position information is selected in consideration. In addition, when the target set 201 is wide, as shown in fig. 11, the fluctuation pattern of the grid points G of the grid at the center of the target set 201 and the fluctuation pattern of the grid points G having the four corner portions of the target set 201 as the position information may be selected. In the method of fig. 11, a plurality of grid points G are selected, and thus a plurality of reference trajectories are generated. Regarding the rotation deviation angle, for example, one of the changes of the N2 patterns is selected in advance.

As shown in fig. 12, the fluctuation pattern selected in S102 may be selected so that the fluctuation pattern having, as the position information, the grid points included in the grid G1 having the highest frequency of occurrence at the time of actual operation is included in the target set 201. Grid G1 is shown by a thick line. If this selection method is selected, a track in which the evaluation value falls within the first range for the mode having the highest probability of occurrence at the time of actual operation can be generated, and therefore productivity is improved.

As a track generation method for generating the reference track, a random sampling method such as RRT (Rapidly Exploring Random Trees), a track parameter optimization method such as CHOMP (Covariant Hamiltonian Optimization for Motion Planning), a graph solution, a reinforcement learning method, and other known techniques can be used. In addition, these methods may be used in combination. On this basis, the system user can also use the track taught by the manual.

In the present embodiment, as restrictions related to the generation of the orbit of the reference orbit, there are considered angular restrictions, angular velocity restrictions, angular acceleration restrictions, and no collision between the robot 101 and the obstacle included in the robot system 100, for each axis motor of the robot 101. Therefore, the generated reference orbit is an orbit capable of avoiding interference with an obstacle located in the work space.

Next, it is checked whether or not a branch point is recorded for each fluctuation pattern calculated by the target set setting unit 303 (S103). When the branch points are recorded for all the fluctuation modes (S103: yes), the process is ended. If not (S103: no), the process proceeds to S104.

The branch point specification section 305 selects one fluctuation mode from among all the fluctuation modes in which no branch point has been recorded yet (S104).

Next, the branch point specification unit 305 selects one reference track (S105). In the case where the reference track is one, the track is selected. In the present embodiment, as shown in fig. 11, when there are a plurality of reference tracks, a reference track having the closest end point Pe, that is, grid point G, with respect to the position of the fluctuation pattern selected as the target is selected. For example, in fig. 11, when the leftmost grid 2 from the top is selected as the fluctuation mode, the reference track of the grid having the upper left corner as the end point Pe is selected. When there are a plurality of reference tracks of the same approximation degree, a rule for selecting one reference track from the plurality of reference tracks is set in advance.

Fig. 13 is a conceptual diagram showing a relationship between the reference track 202 and the branch point 204 in embodiment 1. The relationship between the fulcrum 204 and the reference rail 202 is described. The operation orbit of the robot 101 within the target set 201 for reaching the end point Pe other than the reference orbit 202 is referred to as an individual orbit 203. The individual rail 203 shares a portion of the rail with the reference rail 202. The start position of the track section where the individual track 203 is different from the reference track 202 is referred to as a branching point 204. That is, the branching point 204 is a point for one individual track 203, and is selected from points in the reference track 202. An obstruction 205 is illustrated in fig. 13.

Fig. 14 is a flowchart showing a procedure of calculating the branch point 204 in embodiment 1, and details of the processing performed in S106 in fig. 9 are shown. In S106, candidates of the branch point whose evaluation value is within the second range are searched for. Fig. 15 is a conceptual diagram showing a relationship between the reference rail 202 and the obstacle 205 in embodiment 1. Fig. 16 is a conceptual diagram relating to the selection range of the branch point 204 in embodiment 1.

First, the specification process of the branch point 204 performed by the branch point specification unit 305 will be described. The branch point specification unit 305 extracts a plurality of operation points to be candidates for the branch point 204 from the reference trajectory 202 (S201). The candidate operation points to be the branch points 204 are represented by a group of state amounts that uniquely determine the position and posture of the end effector 102 provided in the robot 101. In the present embodiment, the set of angles of the motors of each axis of the robot 101 indicates a candidate operation point to be the branch point 204. At this time, the reference orbit 202 is configured by a set of data that gives a time tag when the start point Ps is set at time 0 to the group of angles of the motors of the respective axes of the robot 101.

Next, a section of the reference track 202 including the operation points that become candidates for the branch point 204 in the present embodiment will be described. As described above, the reference rail 202 is a rail capable of avoiding interference with the obstacle 205 located in the work space. The fulcrum 204 is designated at an early stage of the reference trajectory 202, and if an attempt is made to generate the remaining portion up to the end point Pe of the individual trajectory 203, the number of obstacles 205 to be considered increases, and the generation of the individual trajectory 203 takes time. Therefore, it is preferable to branch the track from the branching point 204 to the individual track 203 at a stage where the obstacle 205 is avoided as much as possible.

For example, as shown in fig. 15, a temporary obstacle 206 is set, which takes a margin at a predetermined distance L from the outer periphery of an obstacle 205 existing in the work space. The distance L may be arbitrarily specified. When the reference trajectory 202 is traced back from the end point Pe to the start point Ps, the section that becomes a candidate for the branch point 204 is set to a section C1 up to a point Px1 where the reference trajectory collides with the temporary obstacle 206. That is, the fulcrum 204 is selected from the section of the reference orbit 202 where interference between the end effector 102 and the robot body does not occur with respect to the temporary obstacle 206. The robot body refers to a portion of the robot 101 other than the end effector 102.

If the branching point 204 is specified near the end point Pe of the reference track 202 when the position of the selected fluctuation pattern is closer to the start point Ps than the end point Pe of the reference track 202, the individual track 203 may return from the branching point 204 to the position of the selected fluctuation pattern, and the value of the evaluation value may deteriorate. Therefore, as shown in fig. 16, the operation point included in the section C2 on the reference track 202 corresponding to the range d between the end point Pe of the reference track 202 and the position Px2 of the selected fluctuation pattern is excluded from the candidates of the branch point 204.

Next, the branch point specification unit 305 confirms whether or not the evaluation value of each individual track 203 is calculated with respect to all candidates of the branch points 204 (S202). When the evaluation value of each individual track 203 is not calculated (S202: no) for all candidates of the branch point 204, the branch point specification unit 305 selects one point from the plurality of candidates of the branch point 204 extracted in S201 (S203). Then, the branch point specification unit 305 generates the individual track 203, which is a track from one candidate of the selected branch point 204 to the position of one fluctuation pattern selected in S104 (S204). Here, the branch point specification unit 305 uses the same method as that of the individual track 203 generated by the individual track generation unit 308 described later in the generation of the individual track 203. The branch point specification unit 305 calculates the evaluation value of the individual track 203 generated in S204 (S205).

Fig. 17 is a conceptual diagram showing a method of generating a track by the individual track generating unit 308 in embodiment 1. The branch point specification unit 305 generates the individual track 203 by processing similar to fig. 17. Next, a method of generating the individual tracks 203 will be described. In the present embodiment, the trajectory after the branch point 204 is generated using the joint interpolation method. The joint interpolation method is one of the 1 st operations for generating the individual tracks 203.

Fig. 17 is a conceptual diagram showing a case where the number of axes of the robot 101 is 2, but the same method can be applied to a case where the number of axes is greater than or equal to 2. Fig. 17 shows a timing chart of the commanded angular velocity of the shaft 1 and a timing chart of the commanded angular velocity of the shaft 2. In fig. 17, kt represents acceleration time, gt represents deceleration time, and tt represents constant speed time.

First, the individual trajectory generation unit 308 selects an axis having the largest angle change amount from the branching point 204 to the end point Pe, that is, the position of the selected one fluctuation mode, for the highest speed that can be output by each axis motor of the robot 101. The axis at which this angle change amount becomes maximum is referred to as a representative axis j' in the present embodiment. According to fig. 17, since the angle change amount of the shaft 1 is larger than the angle change amount of the shaft 2, the shaft 1 becomes a representative shaft j'. The representative axis j' is selected from a plurality of axes j (j=1, 2, …) to satisfy the expression (1). Here, θ _sj Is the angle, θ, of the axis j at the branching point 204 _Gj Is the angle of the axis j at the end point Pe, i.e. the position of the selected one of the modes of fluctuation, v_max _j Is the highest speed that can be achieved by the motor of shaft j. The right side of formula (1) shows the angular change of axis jThe quantity is relative to the highest speed v_max that can be achieved by the motor of the shaft j _j To what extent.

[ 1 ]

The highest speed vj 'on the track representing the axis, axis j' (j '=1)' _max Set to vj' _max ＝v_max _j′ . At this time, the highest speed vj in the operation of the motor representing the shaft j other than the shaft j' will be represented _max Temporarily set according to the formula (2). In formula (2), (θ) _Gj －θ _sj ) Represents the angle change of the axis j except the representative axis j' (theta) _Gj′ －θ _sj′ ) The amount of angular change representing the axis j' is shown.

[ 2 ]

According to the highest speed vj in the action of the temporarily set shafts _max Angular velocity vj at branching point 204 of each axis _s Angular velocity vj at end point Pe of each shaft _e The acceleration time kt and the deceleration time gt are determined by the speed limit and the acceleration limit of the robot 101. If the acceleration time kt and the deceleration time gt are determined, the following equation (3) relating to the representative axis j' is solved for the constant speed time tt, whereby the constant speed time tt is determined. In formula (3), the reference numerals in formula (3) are omitted for convenience.

[ 3 ] of the following

By taking reference to vj _max Solving equation (3) to determine the highest speed vj of the shaft j other than the temporarily set representative shaft j _max . Through the above calculation, the robot 101 is fully involvedThe angular velocity command value after the branching point 204 is obtained. Therefore, by integrating the angular velocity command value at the time, the group of the operation points and the time tags after the fulcrum 204 can be calculated.

As described above, the branch point specification unit 305 generates the individual track 203, which is a track from one branch point 204 selected from the plurality of candidates of the branch point 204 to the position of one fluctuation pattern selected in S104 (S204), and then calculates the evaluation value of the generated individual track (S205). Next, the branch point specification section 305 confirms whether or not the calculated evaluation value is within the second range (S206). In the present embodiment, the second range is set to be within the operation time T2 s. Set to T2 > T1. The second range is set to a range larger than the first range.

When the evaluation value is within the second range (S206: yes), the branch point specification unit 305 specifies the selected branch point as the formal branch point (S108). If the evaluation value does not fall within the second range (S206: no), the sequence returns to S202. If the sequence returns to S202, another candidate among the plurality of candidates of the branch point 204 extracted in S201 is selected, and then the processing of S202 to S206 is performed, whereby it is determined whether or not the evaluation value of the individual track 203 of the selected another candidate is within the second range. As described above, by repeating the processing of S202 to S206, the branch point 204 at which the evaluation value of the individual track 203 is within the second range is obtained.

Further, after the processing of S202 to S206 is repeated, if the determination of S202 is Yes, the condition that all the plurality of candidates of the branch point 204 extracted in S201 do not deviate from the second range is not satisfied. In the case as described above, the sequence proceeds to S107. In S107, a new reference track 202 other than the reference track 202 generated in S102 is added to the reference track 202. Specifically, a fluctuation pattern different from the fluctuation pattern selected in S104 is selected, and a reference track 202 having an evaluation value within a first range is generated using the newly selected fluctuation pattern (S107). The processing of S202 to S206 is repeated using the new reference trajectory 202, and thereby the branch point 204 whose evaluation value is within the second range is determined.

If the branch point 204 is specified in S108, the branch point specification unit 305 sets the fluctuation pattern selected in S104, the reference track 202 selected in S105, or the reference track 202 added in S107, and the branch point 204 specified in S108 as a set, and records the set to the individual branch point storage unit 306 (S109). That is, the branch point specification unit 305 records track set data, which is a set of the wobble pattern, the reference track 202, and the branch point 204, in the individual branch point storage unit 306.

If the recording in the individual branch point storage unit 306 is completed, the sequence proceeds to S103. In S103, it is determined whether or not the branch point 204 is recorded with respect to all the fluctuation patterns calculated by the target set setting unit 303. When no fluctuation pattern of the recording branch points 204 remains, the track generation unit 310 repeatedly executes the processing of S103 to S109 and the processing of S201 to S206, and records the branch points 204 related to all the fluctuation patterns in the individual branch point storage unit 306.

Fig. 18 is a flowchart showing the processing procedure of the individual trajectory generation unit 308 and the robot control unit 309 in embodiment 1. The robot control unit 309 generates an operation command of the robot 101 using the recorded information of the individual branch point storage unit 306 at the time of actual work in the robot 101.

First, the sensor information acquisition unit 307 measures the actual position, posture, and shape of the target object 200 using the external sensor 104 (S301). The individual track generation unit 308 acquires the type of the end effector 102 and the type of the target object 200 from the task information acquisition unit 302, and acquires the moving speed V of the belt conveyor 103 from the peripheral device information acquisition unit 301. The individual trajectory generation unit 308 selects the approach path of the hand based on the shape of the target object 200 and the task information.

Next, the individual track generation unit 308 reads out track set data having a fluctuation pattern closest to the actual position and posture of the target object 200 from the individual branch point storage unit 306. The individual track generation unit 308 then extracts the reference track 202 and the branch point 204 included in the read track set data (S302).

Next, the individual trajectory generation unit 308 generates an individual trajectory from the extracted branch point 204 to the end point Pe 'using the actual position and posture of the target object 200 acquired in S301 as the end point Pe' (S303). The method described with reference to fig. 17 is used for generating individual tracks. The individual track generation unit 308 directly uses the reference track 202 taken out in S302 for the track from the start point Ps to the branch point 204 taken out in S302. The trajectory from the start point Ps to the branch point 204 corresponds to the 1 st action trajectory of the claims, and the trajectory from the branch point 204 to the end point Pe' corresponds to the 2 nd action trajectory of the claims.

Finally, the individual track generation unit 308 transmits an operation command including the individual track from the generated branching point 204 to the ending point Pe' and the reference track 202 of the section from the starting point Ps to the branching point 204 to the robot control unit 309 (S304). The robot control unit 309 drives the robot 101 in accordance with the received operation command. Thus, the robot 101 moves the end effector 102 from the start point Ps to the end point Pe, and performs a predetermined task on the target object 200.

As described above, according to embodiment 1, the reference trajectory 202, which is an operation trajectory that is not interfered with the obstacle 205 until the branching point 204 is used as it is and satisfies the first range of the evaluation value, is generated from the starting point Ps to the target object 200, and the operation trajectory is generated in accordance with the actual position and posture of the target object 200, so that even when the position and posture of the target object 200 fluctuate, the time required for calculating the trajectory whose evaluation value falls within a certain range can be shortened, and the productivity can be improved.

In addition, according to embodiment 1, the split point 204 is selected so that the evaluation value falls within a certain range when the individual track 203 is generated. Therefore, the generated track can ensure a certain productivity. Further, according to embodiment 1, the candidate of the fulcrum 204 is selected from among the operation points included in the reference orbit 202, which do not interfere with the temporary obstacle 206 that gets the margin at a certain distance L from the outer periphery of the obstacle 205 around the orbit. Therefore, the number of obstacles 205 to be considered in the generation of the individual tracks 203 decreases, and thus the time taken for the generation of the individual tracks 203 can be shortened.

Further, according to embodiment 1, since the fluctuation in the target object 200 includes the difference in the hand approach path, the task can be efficiently achieved by including the operation of the end effector 102 in accordance with the fluctuation in the target object 200. In addition, according to embodiment 1, when there is no branch point 204 in the reference trajectory 202 at which the evaluation value of the individual trajectory 203 falls within the predetermined range, the reference trajectory 202 is newly generated and added, so that it is ensured that the operation trajectory is generated for all the fluctuation patterns in the target set 201.

Embodiment 2.

In embodiment 1, the reference trajectory 202 and the branch point 204 are set for all the fluctuation modes of the target object 200 included in the target set 201. In contrast, in embodiment 2, 1 reference track 202 and branch point 204 are set for all the fluctuation modes included in the target set 201. That is, in embodiment 2, 1 reference track 202 and 1 branch point 204 are shared by all the fluctuation modes.

Fig. 19 is a flowchart showing an operation procedure of generating the individual track 203 in embodiment 2. Fig. 20 is a conceptual diagram showing the reference track 202 and the individual track 203 in embodiment 2. In embodiment 2, the same robot system as in embodiment 1 is used as the robot system 100. The target set setting unit 303 and the reference trajectory generation unit 304 also execute the same processing as in embodiment 1.

As shown in fig. 20, in embodiment 2, 1 reference trajectory 202 and 1 branch point 204 are provided for all the fluctuation modes of the target object 200 included in the target set 201. Therefore, in embodiment 2, the reference trajectory 202 is used directly in the section from the start point Ps to the branch point 204, and the individual trajectory from the branch point 204 to the actual end point Pe is generated in accordance with the position and posture of the actual end point Pe in the section from the branch point 204 to the actual end point Pe. For example, if the actual end point Pe is the upper left corner, the individual track 203-C is generated, if the actual end point Pe is the lower left corner, the individual track 203-D is generated, if the actual end point Pe is the upper right corner, the individual track 203-a is generated, and if the actual end point Pe is the lower right corner, the individual track 203-B is generated.

Next, a procedure for generating 1 reference track 202 and 1 branch point 204 will be described with reference to fig. 19. First, the target set setting unit 303 reads all the fluctuation patterns of the target set 201 calculated by the target set setting unit 303 and the target object 200 included in the target set 201, as in embodiment 1 (S401).

Next, the reference orbit generation unit 304 selects a fluctuation mode having a part of grid points G among the plurality of fluctuation modes, and generates a reference orbit 202 having an evaluation value within a first range by using an arbitrary orbit generation method in the same manner as in embodiment 1, with the selected fluctuation mode being the end point Pe and the predetermined start point Ps being the operation start point (S402). The evaluation value of the track is set to the operation time as in embodiment 1.

As the track generation method, known techniques such as a random sampling method such as RRT, a track parameter optimization method such as CHOMP, a schematic drawing method, and a reinforcement learning method can be used. In addition, these methods may be used in combination. On this basis, the system user can also use the track taught by the manual.

Next, the branch point specification unit 305 specifies a point that is a candidate for the branch point 204 from among the operation points of the reference track 202 (S403). As described with reference to fig. 14 to 16, the section to be selected as the fulcrum 204 employs the same method as that of embodiment 1.

Next, the branch point specification unit 305 confirms whether or not the evaluation value of the individual track 203 in the case of using the branch point 204 specified in S403 is calculated for all the fluctuation modes of the target set 201 (S404). If there is an uncomputed fluctuation pattern (S404: no), one of the uncomputed fluctuation patterns is selected, and the evaluation value of the individual track 203 is calculated (S405). The method of calculating the evaluation value of each individual track 203 is the same as that of embodiment 1. When the calculation of the evaluation value is completed for all the fluctuation modes (Yes in S404), the process proceeds to S406.

Next, the branch point specification unit 305 calculates expected values of the evaluation values of all the fluctuation modes for the candidates of the branch point selected in S403 (S406).

As a method for calculating the expected value of the evaluation value, for example, an equivalent average of the evaluation values of all the fluctuation modes is adopted. The weight average is performed so that the weight of the evaluation value of the individual track 203 corresponding to the fluctuation pattern in the central portion of the target set 201 is higher than the weight of the evaluation value of the individual track 203 corresponding to the fluctuation pattern in the outer periphery of the target set 201. The expected value may be calculated based on the number of times each fluctuation pattern occurs in the actual production line.

The branch point specification unit 305 confirms whether or not the expected value of the calculated evaluation value falls within the second range (S407). As the second range, the operation time is set to be within T2 s (T2 > T1). When the expected value of the evaluation value falls within the second range (S407: yes), the routine proceeds to S409. If the expected value of the evaluation value does not fall within the second range (S407: no), it is checked whether or not another point serving as a candidate for the branch point 204 remains in the reference track 202 (S408). If candidates remain (Yes in S408), the process returns to S403 to set candidates of another branch point 204. If there is No reserve candidate (S408: no), the process proceeds to S409. In embodiment 2, in the determination of S407, a third range larger than the second range may be set.

In S409, the branch point specification unit 305 selects one branch point 204, and adds the selected branch point 204 and the reference track 202 to the individual branch point storage unit 306 as a set. In the present embodiment, the branch point 204 at which the expected value of the evaluation value, that is, the expected value of the operation time becomes minimum is selected. A set of branch points 204 and reference rails 202 are stored in an individual branch point storage unit 306.

In actual operation of the robot 101, the individual branch point storage unit 306 generates an operation track of the robot 101 using the group of branch points 204 and the reference track 202 stored in the individual branch point storage unit 306, as described in fig. 18. First, the actual position, posture, and shape of the target object 200 are measured using the external sensor 104. Next, a set of the fulcrums 204 and the reference rail 202 are taken out from the individual branch point storage section 306. Next, using the method described in fig. 17, an individual trajectory from the extracted branch point 204 to the end point Pe 'is generated with the actual position and posture of the target object 200 as the end point Pe'. The reference track 202 is directly used as the track from the start point Ps to the branch point 204. Then, the robot 101 is driven based on an operation command including the individual trajectory from the generated branching point 204 to the ending point Pe' and the reference trajectory 202 of the section from the starting point Ps to the branching point 204.

As described above, according to embodiment 2, since only one set of the reference track 202 and the set of the branch points 204 is set for a plurality of fluctuation modes, the amount of data stored in the individual branch point storage unit 306 can be reduced. In addition, according to embodiment 2, since the point at which the candidate evaluation value is smallest is selected from among the points within the reference trajectory 202 as the branching point 204, the generated trajectory ensures a constant productivity as in embodiment 1. However, if compared with embodiment 1 in which the split point 204 is specified for each fluctuation mode of the target object 200, there is a possibility that the productivity of individual tracks for the respective fluctuation modes is lowered.

Here, a hardware configuration of the robot control device shown in fig. 4 will be described. Fig. 21 is a block diagram showing a hardware configuration of the robot control device according to embodiment 1 and embodiment 2.

The robot control device can be realized by a hardware configuration 406 including the arithmetic device 404 and the storage device 405 shown in fig. 21. Examples of computing device 404 are a CPU (also referred to as Central Processing Unit, central processing device, computing device, microprocessor, microcomputer, processor, DSP (Digital Signal Processor)) or system LSI (Large Scale Integration). Examples of storage devices 405 are RAM (Random Access Memory) or ROM (Read Only Memory).

The robot control device is realized by the arithmetic device 404 reading and executing a program for executing the operation of the robot control device stored in the storage device 405. The program may be a sequence or a method for causing a computer to execute the robot control device.

The storage device 405 stores the fulcrum 204 and the reference rail 202. The storage device 405 is also used as a temporary memory when various processes are performed by the arithmetic device 404.

The program executed by the computing device 404 may be provided as a computer program product by being stored in a computer-readable storage medium in an installable form or a file in an executable form. The program executed by the arithmetic unit 404 may be supplied to the robot control device via a network such as the internet.

The robot control device may be realized by dedicated hardware. The functions of the robot control device may be partly implemented by dedicated hardware, and partly implemented by software or firmware. For example, the computer-executable target set setting unit 303, the reference trajectory generation unit 304, the branch point specification unit 305, and the individual branch point storage unit 306 may be configured to cause the robot controller to execute the sensor information acquisition unit 307, the individual trajectory generation unit 308, and the robot control unit 309.

The configuration shown in the above embodiment represents a part of the content of the present invention, and may be combined with other known techniques, and may be appropriately combined, or a part of the configuration may be omitted or changed without departing from the scope of the present invention.

Description of the reference numerals

100 robot system, 101 robot, 102 end effector, 103 belt conveyor, 104 external sensor, 200 object, 201 target set, 202 reference orbit, 203 individual orbit, 204 branch point, 205 obstacle, 206 temporary obstacle, 301 peripheral device information acquisition unit, 302 task information acquisition unit, 303 target set setting unit, 304 reference orbit generation unit, 305 branch point designation unit, 306 individual branch point storage unit, 307 sensor information acquisition unit, 308 individual orbit generation unit, 309 robot control unit, 310 orbit generation unit, 404 arithmetic device, 405 storage unit, 406 hardware configuration, pe end point, ps start point.

Claims

1. A robot control device for moving an end effector of a robot from a start point to an end point, performing a predetermined task with respect to a target object whose position and posture are not fixed,

the robot control device is characterized by comprising:

A storage unit that stores a reference rail and a branching point in association with each of a plurality of sets of positions and postures that can be obtained by the target object;

a measuring unit that measures a position and an attitude of the target object when the task is executed;

an individual trajectory generation unit that acquires the reference trajectory and the branch point corresponding to the measured position and posture of the target object from the storage unit, sets the measured position and posture of the target object as the end point, generates a 1 st operation trajectory, which is an operation trajectory of the robot from the start point to the branch point, using the acquired reference trajectory, and generates a 2 nd operation trajectory, which is an operation trajectory of the robot from the acquired branch point to the end point; and

a robot control unit that controls driving of the robot in accordance with an operation command including the 1 st operation track and the 2 nd operation track,

the reference trajectory is an operation trajectory of the robot from the start point to a 1 st point which is one of a plurality of positions and postures which can be obtained by the target object, is an operation trajectory which can avoid interference with an obstacle and satisfies a condition that an evaluation value enters a first range,

The branching point is a point on the reference rail, and is an operation rail in which the evaluation value of the operation rail from the branching point to the 1 st point is within a second range larger than the first range.

2. The robot control device of claim 1, wherein the control device comprises a plurality of control units,

the device comprises:

a target set setting unit that sets a set of a plurality of sets of positions and postures that can be obtained by the target object as a target set;

a reference trajectory generation unit that selects a group of at least one position and posture of the target object included in the target set, and generates the reference trajectory based on the selected group of the position and posture of the target object; and

and a branch point specification unit that extracts a plurality of candidate points of the branch point from the generated reference trajectory, selects an operation trajectory having the evaluation value within the second range from among a plurality of operation trajectories from the plurality of extracted candidate points to the 1 st point, performs 1 st processing of determining a candidate point corresponding to the selected operation trajectory as a branch point corresponding to the 1 st point with respect to all of a plurality of groups of positions and orientations that can be obtained by the target object, calculates the branch point for each of the groups of positions and orientations that can be obtained by the target object, and stores the calculated branch point and the generated reference trajectory in the storage unit.

3. The robot control device according to claim 2, wherein,

the reference trajectory generation unit generates one reference trajectory common to a plurality of groups of positions and orientations that can be obtained by the target object.

4. The robot control device according to claim 3, wherein,

the reference track generating unit sets the one reference track at a center position of the target set.

5. The robot control device according to claim 3, wherein,

the reference orbit generation unit sets the one reference orbit at a position where the frequency of occurrence is highest at the time of actual operation of the robot.

6. The robot control device according to claim 2, wherein,

the reference trajectory generation unit sets the reference trajectory at the 1 st point different from the set of positions and attitudes that can be obtained by the target object.

7. The robot control device according to claim 2, wherein,

the reference trajectory generation unit generates a new reference trajectory when the branching point satisfying the condition that the evaluation value enters the second range does not exist.

8. The robot control device of claim 1, wherein the control device comprises a plurality of control units,

The storage unit stores a set of the reference orbit, the branching point, and an orbit of a hand approach path to the target object in association with a set of a plurality of positions and postures that can be obtained by the target object.

9. A robot control device for moving an end effector of a robot from a start point to an end point, performing a predetermined task with respect to a target object whose position and posture are not fixed,

the robot control device is characterized by comprising:

a storage unit that stores a common reference rail and a branching point for a plurality of sets of positions and postures that can be obtained by the target object;

an individual trajectory generation unit that acquires the reference trajectory and the branching point from the storage unit, sets the measured position and posture of the target object as the end point, generates a 1 st operation trajectory, which is an operation trajectory of the robot from the start point to the branching point, using the acquired reference trajectory, and generates a 2 nd operation trajectory, which is an operation trajectory of the robot from the branching point to the end point; and

the branching point is a point on the reference trajectory, and is an operation trajectory that satisfies a second range, which is larger than the first range, from the branching point to an average of the evaluation values of a plurality of operation trajectories up to a plurality of positions and postures that can be obtained by the target object.

10. The robot control device of claim 9, wherein the control device comprises a plurality of sensors,

the device comprises:

a target set setting unit that sets a set of a plurality of positions and postures that can be obtained by the target object as a target set;

a reference trajectory generation unit that selects a set of one position and one posture of the target object included in the target set, and generates the reference trajectory based on the selected set of the position and the posture of the target object; and

And a branch point specification unit that extracts a plurality of candidate points of the branch point from the generated reference trajectory, performs a 2 nd process of obtaining an average of the evaluation values of a plurality of operation trajectories from 1 candidate point among the plurality of extracted candidate points to a group of a plurality of positions and orientations that can be obtained by the target object, regarding all the candidate points, determines a candidate point in which the average of the evaluation values among the plurality of candidate points falls within the second range as the branch point, and stores the determined branch point and the reference trajectory in the storage unit.

11. The robot control device of claim 10, wherein the control device comprises a plurality of sensors,

the average of the evaluation values is a weight of the evaluation value of the central portion of the target set, and the weight average is increased as compared with a weight of the evaluation value of the outer periphery of the target set.

12. The robot control device according to claim 2 or 10, wherein,

the branch point specification unit selects the branch point from among the reference rails, the branch point being selected from among sections in which interference between the end effector and the robot main body does not occur with respect to a temporary obstacle that takes a margin at a predetermined distance from the outer periphery of the obstacle.

13. The robot control device according to any one of claims 1 to 12, characterized in that,

the evaluation value is an operation time of the robot.

14. A robot control program for moving an end effector of a robot from a start point to an end point, performing a predetermined task with respect to a target object whose position and posture are not fixed,

the robot control program causes a computer to execute the steps of:

storing a reference track and a branching point in a storage unit in correspondence with each of a plurality of sets of positions and postures that can be obtained by the target object;

measuring the position and the posture of the target object when the task is executed;

the method includes obtaining the reference orbit and the branch point corresponding to the measured position and posture of the target object from the storage unit, setting the measured position and posture of the target object as the end point, generating a 1 st operation orbit, which is an operation orbit of the robot from the start point to the branch point, using the obtained reference orbit, and performing a 1 st operation to generate a 2 nd operation orbit, which is an operation orbit of the robot from the obtained branch point to the end point; and

The robot is driven and controlled according to the action instructions including the 1 st action track and the 2 nd action track,

15. A robot control method for moving an end effector of a robot from a start point to an end point, performing a predetermined task for a target object whose position and posture are not fixed,

the robot control method is characterized by comprising the following steps: