JPS62279409A - Robot movement simulator - Google Patents

Abstract

PURPOSE:To make parameter adjustment that optimizes the action without moving the robot actually, by simulating an action close to an actual robot by a robot action simulator with a driving system simulator and a mechanism system simulator. CONSTITUTION:In a simulation mode, a program controlling section 106 sends a target value of operation a1 to a locus forming section 107 according to a program. The locus forming section 107 calculates interpolation points of the locus of a robot and converts them to driving command values b1. A driving system simulator 108 calculates response of internal model to the values b1 based on the value of a parameter 112, and a mechanism system simulator 9 calculates response d1 of a mechanism system in the case of an actual robot from the driving output value c1 using a parameter 113. The resultant display data e1 is displayed in a display section 105. In a data control mode, control of inputting and outputting of conversion formula, mechanism system data file, controlling system of a driving system, constants of an actuator and driving system and driving system data file.

Description

[Detailed Description of the Invention] 3. Detailed Description of the Invention [Field of Industrial Application] The present invention is a robot motion simulator, particularly a motion simulation including a control system, drive system, and +A system of a physical robot. This invention relates to a robot VTTH production simulator that can be realized.

[Technological environment]

In recent years, robot control devices have increasingly digitalized control systems, including servo systems, and the control loop itself is executed as software on a microprocessor, and control parameters are now held as numerical data. The background to the practical application of these software controls is the development of digital signal processing technology, the tendency for microprocessors to become faster and more multifunctional9, and the need to reduce manufacturing costs by reducing the number of parts, especially analog parts, and adjustment man-hours. , it is possible to achieve low temperature and light construction through high-density packaging, so it is likely to become the mainstream of future construction equipment.

On the other hand, the control force formula of the digital servo system involves wasted time such as the time for calculating the 1'B-l vector during sampling, which is not found in the conventional analog servo system, making adjustment difficult.

[Conventional technology]

In the conventional adjustment method, the analog interface part of the digital servo system, such as the output of the current amplifier or the speed feedback output, is viewed with a synchroscope and the parameters of the control system, such as speed feedback gain and current amplification coefficient, are numerically assigned. You can determine the optimal value based on the waveform, or you can actually move the robot and check its behavior, whether there is overshoot or vibration, etc.
Parameters were determined based on sound, time, etc.

FIG. 8 is a block diagram showing the actual configuration of the robot control unit, in which the operation program input from the input unit 203 in the program control unit 204 is stored in the program storage 205! 7
The trajectory command value a2 is read out and passed to the trajectory generation unit 206. The trajectory generation section generates a drive command value b2 with reference to the mechanism parameter 209, and passes it to the drive control section 207 t'L.

The drive control section refers to the control parameters 210 and calculates data C2 to be applied to the drive section 208 in accordance with a control law designed in advance. The drive unit drives the manipulator mechanism system 202 with a dynamic response characteristic determined by constants of the actuator and load, and as a result, draws a motion trajectory d2 in the work space.

In conventional adjustment methods, the operator 200
1 to move the robot 202 and draw a certain trajectory d2 to observe its behavior.Based on the results, the control parameters j1210 are adjusted to appropriate values. In this case, for example, adjustments are made repeatedly in the direction of the trajectory. In some cases, the validity of the parameters is determined by observing signals inside the drive unit 208 or the drive control unit 207.

[Problem that the invention seeks to solve]

In the conventional method described above, it is necessary to determine whether the optimal parameters have been obtained based on a certain signal, and
This method had the disadvantage that the results were not uniformly determined because it was judged based on the experience and subjectivity of the coordinator. In addition, a physical robot had to be used for adjustment, resulting in low work efficiency.

On the other hand, there is a method of using a servo system simulator for each drive axis, but it is not possible to evaluate the dynamic characteristics when the entire robot moves, and as a result, the actual robot behavior is not necessarily optimized, and The disadvantage is that it does not follow the trajectory of

[Means for solving problems]

The robot yt' motion simulator of the present invention includes a program control unit that sequentially executes a program that describes the robot's motion, and converts numerical data described by the program into a trajectory of the robot's movement, which is then converted into drive data for each drive axis. a drive system simulator that simulates the movement of each drive axis of the physical robot that is discretely driven and controlled by the drive data, and a drive system simulator that simulates the movement of each drive axis of the physical robot that is output from the simulator. The system includes a mechanical simulator that simulates the locus of movement of the robot, and a display control section and a display section that display these loci and the robot body.

〔Example〕

Next, embodiments of the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing one embodiment of the present invention.
01: Input section, 102: Simulation control section, 10
3: data management section, 104: display control section, ios: display section. This simulator has two states: a simulation mode in which simulation is executed, and a data management mode in which data necessary for simulation execution is managed.

In the simulation mode, the program control unit 106
sequentially sends the operation target value a1 to the trajectory generating section 107 according to the program read out from the program storage 110. The trajectory generating section calculates an interpolation point of the robot's trajectory from the motion target value and the speed profile, and converts this into a drive command value b1 that is set to the drive axis by a conversion formula determined by the mechanical parameters 111. The drive system simulator 108 calculates the response of the internal model to the drive command value input based on the value of the control parameter 112, and from this drive output value C1, the mechanical system simulator 109 calculates the response of the mechanical system in the case of a physical robot. d1 as mechanical parameter 113
Calculate using. This result is geometrically converted by the display control section, and display data e1 is displayed on the display section.

In data management mode, you can create mechanical data files that include the mechanical structure, dimensions, and drive axis configuration of the robot to be controlled, as well as conversion formulas and their parameters, as well as drive system control methods, actuator constants, and drive system data files. Manages the input and output of drive system data files, including constants and drive system models that can be configured.

In the second figure, the input trajectory 301 written in the program is
This shows the process of conversion within the control unit until it is output as a trajectory 306 in the working space of the physical robot.In 301, the robot is programmed to move in a straight line from the starting point A to the ending point B according to the velocity profile 302. In this case, the trajectory generation section first generates the interpolation point sequence P (1
JP in). Next, the trajectory generation section generates the corresponding displacement θi of the drive shaft for each interpolation point P(il).The drive control section receives the discrete interpolation points of the drive shaft, and then follows this. The drive is controlled as follows.

The control/drive system shows a response according to the dynamic characteristics determined by its configuration and constants, and outputs, for example, a response 305 with dead time or -order delay. As a result, the actually driven mechanical system does not always coincide with the desired straight line AB. 306 shows the situation.

FIG. 3 is a block diagram showing one example of the configuration of a drive system simulator, which constitutes a position follow-up control system that feeds back an output g4 with respect to a target value a4. Position deviation a4
A deviation input b4 obtained by multiplying -g4 by a gain Kp401 is applied to the speed loop 411. b4 and the speed output f4 are multiplied by a speed feedback gain GY409, and the deviation from the feedback signal i4 is multiplied by a speed loop gain Kv402, and the deviation human power C4 is applied to the torque loop 410.

The flow of calculation when calculating this "7'-□-ta ζ hole L/-'・71 will be explained. Position target value a4 and output g4
From b+=Ko(a4-g4) ++++++++-・
... becomes ■. Next, the calculation of the velocity loop is c4=Kv (b 4 - i 4 ) ==-... from b4 and feedback i4.
...■i 4=Gv f 4 ・
・・・・・・・・・・・・■. Also, the calculation of the torque loop is d4= from 04 and feedback h4, (C4-h4)...
・・・■h4=GL C4・・・・・・・・・・・・■
becomes. In an actual robot control system, Le and Lm correspond to the electrical system and mechanical system of the driving section of the physical robot, respectively, and the output e4 is determined according to the input d4, and the output f4 is determined in response to the input e4.

This relationship will be explained in more detail in FIG.

Figure 4 (al indicates a physical model. When the armature of the motor is modeled by an inductance La and a resistance R, the armature voltage ea is the difference between the input voltage e1 and the back electromotive force eb, e a = e i e b...
・・・・・・・・・■. If the armature current is 1, i e =L −+Ri ・・・・・・・・・・・・
・■a adt relationship. Also, conversion to torque is τ=Kti ・・・・・・・・・・・・
■It is given by. Next, in the mechanical system, for the inertia J1 friction w τ=τ−+Dw ・・・・・・・・・・・・■
The equation of motion for t is established, and in this case, J and D take values determined by the position of the robot. This method will be explained using FIG. 6.

Also, regarding the back electromotive force eb, eb=Kew ・・・・・・・・・・・・
It is given by

Here, the calculations from ■ to ■ are performed discretely from the observed values read every sampling period Ts, but on the other hand, the parts from ■ to [phase] are actually continuous. system, and it is necessary to discretize this model so that it can be handled uniformly. The block diagram in Figure 4 (bl is first modeled as a continuous system for ■~(seconds).The block diagram in Figure 4 is a block diagram that is easier to discretize.
Figure (cl). If we take the armature current i and the angular velocity output W from the state variation of the drive unit a, we get the state space representation of this system.If we discretize this, we get the following [7...・・・－・・・(bag ■2 unit matrix n: indicates the nth state.In this way, the drive unit simulator can be uniformly treated as a discrete control system by the sampling period T of the control system of the physical robot. 9 can be handled.

FIG. 5 shows an example of a physical robot. Three drive shafts 601
603 and three link mechanisms Go4 to 606, J of each drive shaft is given as follows.

f, g, h; Function Ii; Moment of inertia of the first link m1 Mass of the ni-th link θi; Angle γi of the i-th drive shaft; Reduction ratio of the i-th drive shaft 11. The value of element J of the coefficient matrix F of Equation 13 is J=JRi + Ji...
...[Yes] However, JRi: Inertia J of the 'th drive section; Determined by the inertia of the link applied to the first drive shaft. In the actual calculation, equation 16 is substituted into equation 13.14, and equation 12 is solved from this.

Although this is uniquely determined by the specifications of the Equation 12 model, it is effective to make some changes and check the response, taking into account the inaccuracy of the model or the manufacturing errors and nonlinearity of the actual product. As parameters, there are coefficients that appear in equations 1 to 5, and the actual adjustment work consists of determining these values.

FIG. 6 shows a combination of the speed command waveform a8 and the follow-up waveform b8 when the input locus AB drawn in the SPD diagram explaining the flow of this calculation is moved linearly.

〔Effect of the invention〕

The robot motion simulator of the present invention includes a drive system simulator that simulates the movement of each drive axis of a physical robot that is discretely and directly driven and controlled, and a drive system simulator that simulates the movement of each drive axis of a physical robot that is directly driven and controlled. A mechanical simulator that simulates the trajectory can simulate behavior that is closer to that of a real robot, so parameters can be adjusted to optimize the robot's behavior without actually operating the robot, and it is possible to design a robot control device using digital servos.・This has the effect of making adjustment easier.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing one embodiment of the present invention, and FIG.
The figure is a graph showing how input signals are processed and output inside the controller, Figure 3 is a block diagram showing an example of a control system, and Figures 4 (al to (C) are drive 5 is a conceptual diagram showing an example of a mechanical system model of the robot, FIG. 6 is an SPD diagram showing simulation calculations, and FIG. 7 is a graph showing an example of simulation output. FIG. 8 is a block diagram showing an example of the conventional system. 101: Input section, 102: Simulation control section, 1
03: Data management section, 104: Display control section, 105: Display section, 106: Program control section, 107: Trajectory generation section, 108: Drive system simulator, 109: Mechanism system simulator, 110 Nibragram storage, 111: Mechanism parameter, 112: Control parameter, 113: Mechanism parameter, 1
14: Mechanism system data, 115: Drive system data, al: Operation target value, bl: Drive command value, C1: Drive output value, dl
: trajectory, el: display data 200: operator, 201: control device, 202: robot, 203: input section, 204 nibogram control section, 2
05 Niprogram memory, 206: Trajectory generation unit, 207:
Drive control unit, 208: Drive unit, 209: Mechanism parameter,
210: Control parameter, C2: Operation target value, b2: Drive command value, C2: Drive signal, d2: Trajectory 301: Trajectory target value, 302: Speed profile, 30
3: Interpolation point sequence, 304: Drive axis interpolation point sequence, 305: Drive axis output, 306: Trajectory 401: Position loop gain, 4o2: Speed loop gain, 403: ) Loop gain, 404: Drive system, 40
5:ii Electrical system 4o6: Mechanical system, 407: Integral element, 4
08: Current feedback coefficient, 409: Speed feedback coefficient, 41°: Torque loop, 411: Speed loop, C4: Position command value, b4: Speed command value, C4: Torque command value, d4: [Pressure, C4: Current, f4: speed output,
g4: position output, h4: current, feedback, i4:
Speed feedback 501: electrical system, 502: mechanical system, 5o3: inductance, 504: resistance, 505: armature, 506: load inertia, 507: load friction, 508: electrical system block, 509: armature (torque conversion), 510: Mechanical system, 5
11: armature (back electromotive force), ei: input voltage, eb: back electromotive force, i: armature current, t: torque, W: angular velocity 601: first drive shaft, 602: second drive shaft, 603: Third drive shaft, 604: first link, 605: second link,
606: Third link, 607: Center of gravity of second link, 60
8: Center of gravity of the third link, C8: Drive command value, b8: Response.柘田 岨田P5 し 0

Claims (1)

[Claims] a program control section that sequentially executes a program that describes the movement of the robot; a trajectory generation section that converts the numerical data described by the program into a locus of movement of the robot and converts it into drive data for each drive axis; a drive system simulator that simulates the movement of each drive axis of a physical robot that is driven and controlled in a discrete manner by data; a mechanical system simulator that simulates the trajectory of the physical robot based on the movement of each drive axis that is output from the simulator; A robot motion simulator comprising a display control section and a display section that display these trajectories and the robot body.