0
Research Papers

Modeling and Parameter Estimation of Robot Manipulators Using Extended Flexible Joint Models

[+] Author and Article Information
Stig Moberg

Division of Automatic Control,
Department of Electrical Engineering,
Linköping University,
Linköping SE-581 83, Sweden;
ABB AB—Robotics
Västerås SE-721 68, Sweden
e-mail: stig.moberg@se.abb.com

Erik Wernholt

Division of Automatic Control,
Department of Electrical Engineering,
Linköping University,
Linköping SE-581 83, Sweden
e-mail: erik@wernholt.se

Sven Hanssen

ABB AB—Robotics
Västerås SE-721 68, Sweden;
Department of Solid Mechanics,
Royal Institute of Technology,
Stockholm SE-100 44, Sweden
e-mail: sven.hanssen@se.abb.com

Torgny Brogårdh

ABB AB—Robotics,
Västerås SE-721 68, Sweden
e-mail: torgny.brogardh@se.abb.com

The same modeling principle can of course be applied to parallel kinematic structures and serial structures with parallelogram linkages.

Due to the symmetrical inertia tensor, only six components of Ji need to be defined.

The electrical drive system is not included in this model and is assumed to be ideal for the frequency range considered here. The drive system can be taken into account as described in Sec. 3.4.

The friction of the gearbox is actually split into two components and distributed between fm and fg.

In this paper, the number of DOF refers to the number of independent coordinates necessary to specify the (internal) configuration of a system.

Bearing elasticity is called joint flexibility in Ref. [18], while the transmission elasticity is called drive flexibility.

These nonlinearities can also be regarded as deterministic disturbances.

The path contains straight lines, circular segments, corner zones, and has a geometry that is expected to excite the manipulator over a wide frequency range.

Contributed by the Dynamic Systems Division of ASME for publication in the JOURNAL OF DYNAMIC SYSTEMS, MEASUREMENT, AND CONTROL. Manuscript received November 24, 2010; final manuscript received December 17, 2013; published online February 10, 2014. Assoc. Editor: Rama K. Yedavalli.

J. Dyn. Sys., Meas., Control 136(3), 031005 (Feb 10, 2014) (13 pages) Paper No: DS-10-1346; doi: 10.1115/1.4026300 History: Received November 24, 2010; Revised December 17, 2013

This paper considers the problem of dynamic modeling and identification of robot manipulators with respect to their elasticities. The so-called flexible joint model, modeling only the torsional gearbox elasticity, is shown to be insufficient for modeling a modern industrial manipulator accurately. Another lumped parameter model, called the extended flexible joint model, is therefore used to improve the model accuracy. In this model, nonactuated joints are added to model the elasticity of the links and bearings. The unknown elasticity parameters are estimated using a frequency-domain gray-box identification method. The conclusion is that the obtained model describes the movements of the motors and the tool mounted on the robot with significantly higher accuracy than the flexible joint model. Similar elasticity model parameters are obtained when using two different output variables for the identification, the motor position and the tool acceleration.

Copyright © 2014 by ASME
Your Session has timed out. Please sign back in to continue.

References

Sweet, L. M., and Good, M. C., 1985, “Redefinition of the Robot Motion-Control Problem,” IEEE Control Syst. Mag., 5(3), pp. 18–25. [CrossRef]
Spong, M. W., 1987, “Modeling and Control of Elastic Joint Robots,” ASME J. Dyn. Syst., Meas., Control, 109(4), pp. 310–319. [CrossRef]
Tomei, P., 1991, “A Simple PD Controller for Robots With Elastic Joints,” IEEE Trans. Autom. Control, 36(10), pp. 1208–1213. [CrossRef]
De Luca, A., 2000, “Feedforward/Feedback Laws for the Control of Flexible Robots,” IEEE International Conference on Robotics and Automation (ICRA ’00), San Francisco, CA, April 24–28, pp. 233–240. [CrossRef]
De Luca, A., and Siciliano, B., 1991, “Closed-Form Dynamic Model of Planar Multilink Lightweight Robots,” IEEE Trans. Syst., Man, Cybernet., 21(4), pp. 826–839. [CrossRef]
Book, W., and Obergfell, K., 2000, “Practical Models for Practical Flexible Arms,” IEEE International Conference on Robotics and Automation (ICRA ’00), San Francisco, CA, April 24–28, pp. 835–842. [CrossRef]
Yoshikawa, T., Murakami, H., and Hosoda, K., 1990, “Modeling and Control of a Three Degree of Freedom Manipulator With Two Flexible Links,” 29nd IEEE Conference on Decision and Control, Honolulu, HI, December 5–7, pp. 2532–2537. [CrossRef]
Ueno, H., Xu, Y., and Yoshida, T., 1991, “Modeling and Control Strategy of a 3-D Flexible Space Robot,” IEEE/RSJ International Workshop on Intelligent Robots and Systems (IROS ’91), Osaka, Japan, November 3–5, pp. 978–983. [CrossRef]
Siciliano, B., and Khatib, O., eds., 2008, Springer Handbook of Robotics, Springer-Verlag, Berlin.
Subudhi, B., and Morris, A. S., 2002, “Dynamic Modelling, Simulation and Control of a Manipulator With Flexible Links and Joints,” Rob. Auton. Syst., 41(4), pp. 257–270. [CrossRef]
Brogårdh, T., 2007, “Present and Future Robot Control Development—An Industrial Perspective,” Ann. Rev. Control, 31(1), pp. 69–79. [CrossRef]
Brogårdh, T., 2009, “Robot Control Overview: An Industrial Perspective,” Model. Identif. Control, 30(3), pp. 167–180. [CrossRef]
Öhr, J., Moberg, S., Wernholt, E., Hanssen, S., Pettersson, J., Persson, S., and Sander-Tavallaey, S., 2006, “Identification of Flexibility Parameters of 6-Axis Industrial Manipulator Models,” ISMA International Conference on Noise and Vibration Engineering, Leuven, Belgium, September 18–20, pp. 3305–3314.
Moberg, S., and Hanssen, S., 2007, “A DAE Approach to Feedforward Control of Flexible Manipulators,” IEEE International Conference on Robotics and Automation, Rome, April 10–14, pp. 3439–3444. [CrossRef]
Tuttle, T., and Seering, W., 1996, “A Nonlinear Model of a Harmonic Drive Gear Transmission,” IEEE Trans. Rob. Autom., 12(3), pp. 368–374. [CrossRef]
Kane, T. R., and Levinson, D. A., 1985. Dynamics: Theory and Applications, McGraw-Hill Publishing Company, New York.
Shabana, A., 1998, Dynamics of Multibody Systems, Cambridge University Press, Cambridge, UK.
Hardeman, T., 2008, “Modelling and Identification of Industrial Robots Including Drive and Joint Flexibilities,” Ph.D. thesis, University of Twente, Enschede, The Netherlands.
Huang, Y., and Lee, C. S. G., 1987, “Generalization of Newton-Euler Formulation of Dynamic Equations to Nonrigid Manipulators,” 1987 American Control Conference, Minneapolis, MN, June 10–12, pp. 72–77.
Yoshikawa, T., Ohta, A., and Kanaoka, K., 2001, “State Estimation and Parameter Identification of Flexible Manipulators Based on Visual Sensor and Virtual Joint Model,” IEEE International Conference on Robotics and Automation (2001 ICRA), Seoul, Korea, May 21–26, pp. 2840–2845. [CrossRef]
Pham, M. T., Gautier, M., and Poignet, P., 2001, “Identification of Joint Stiffness With Bandpass Filtering,” IEEE International Conference on Robotics and Automation (2001 ICRA), Seoul, Korea, May 21–26, pp. 2867–2871. [CrossRef]
Pham, M. T., Gautier, M., and Poignet, P., 2002, “Accelerometer Based Identification of Mechanical Systems,” IEEE International Conference on Robotics and Automation (ICRA ’02), Washington, DC, May 11–15, pp. 4293–4298. [CrossRef]
Johansson, R., Robertsson, A., Nilsson, K., and Verhaegen, M., 2000, “State-Space System Identification of Robot Manipulator Dynamics,” Mechatronics, 10(3), pp. 403–418. [CrossRef]
Östring, M., Gunnarsson, S., and Norrlöf, M., 2003, “Closed-Loop Identification of an Industrial Robot Containing Flexibilities,” Control Eng. Practice, 11(3), pp. 291–300. [CrossRef]
Berglund, E., and Hovland, G. E., 2000, “Automatic Elasticity Tuning of Industrial Robot Manipulators,” 39th IEEE Conference on Decision and Control, Sydney, Australia, December 12–15, pp. 5091–5096. [CrossRef]
Hovland, G. E., Berglund, E., and Hanssen, S., 2001, “Identification of Coupled Elastic Dynamics Using Inverse Eigenvalue Theory,” 32nd International Symposium on Robotics (ISR), Seoul, Korea, April 19–21, pp. 1392–1397.
Behi, F., and Tesar, D., 1991, “Parametric Identification for Industrial Manipulators Using Experimental Modal Analysis,” IEEE Trans. Rob. Autom., 7(5), pp. 642–52. [CrossRef]
Hovland, G., Berglund, E., and Sørdalen, O., 1999, “Identification of Joint Elasticity of Industrial Robots,” 6th International Symposium on Experimental Robotics, Sydney, Australia, March 26–28, pp. 455–464.
Wernholt, E., and Gunnarsson, S., 2006, “Nonlinear Identification of a Physically Parameterized Robot Model,” 14th IFAC Symposium on System Identification, Newcastle, Australia, March 29–31, pp. 143–148.
Albu-Schäffer, A., and Hirzinger, G., 2001, “Parameter Identification and Passivity Based Joint Control for a 7 DOF Torque Controlled Light Weight Robot,” IEEE International Conference on Robotics and Automation (2001 ICRA), Seoul, Korea, May 21–26, pp. 2852–2858. [CrossRef]
Ljung, L., 1999, System Identification: Theory for the User, 2nd ed., Prentice-Hall, Upper Saddle River, NJ.
Canudas de Wit, C., Olsson, H., Åström, K., and Lischinsky, P., 1995, “A New Model for Control of Systems With Friction,” IEEE Trans. Autom. Control, 40(3), pp. 419–425. [CrossRef]
Al-Bender, F., Lampaert, V., and Swevers, J., 2005, “The Generalized Maxwell-Slip Model: A Novel Model for Friction Simulation and Compensation,” IEEE Trans. Autom. Control, 50(11), pp. 1883–1887. [CrossRef]
Wernholt, E., and Moberg, S., 2008, “Frequency-Domain Gray-Box Identification of Industrial Robots,” 17th IFAC World Congress, Seoul, Korea, July 6–11, pp. 15372–15380.
Wernholt, E., and Moberg, S., 2011, “Nonlinear Gray-Box Identification Using Local Models Applied to Industrial Robots,” Automatica, 47(4), pp. 650–660. [CrossRef]
Pintelon, R., and Schoukens, J., 2001, System Identification: A Frequency Domain Approach, IEEE Press, New York.
Wernholt, E., and Löfberg, J., 2007, “Experiment Design for Identification of Nonlinear Gray-Box Models With Application to Industrial Robots,” 46th IEEE Conference on Decision and Control, New Orleans, LA, December 12–14, pp. 5110–5116. [CrossRef]
Guillaume, P., 1998, “Frequency Response Measurements of Multivariable Systems Using Nonlinear Averaging Techniques,” IEEE Trans. Instrum. Meas., 47(3), pp. 796–800. [CrossRef]
Wernholt, E., and Moberg, S., 2008, “Experimental Comparison of Methods for Multivariable Frequency Response Function Estimation,” 17th IFAC World Congress, Seoul, Korea, July 6–11, pp. 15359–15366.
Dobrowiecki, T., and Schoukens, J., 2007, “Measuring a Linear Approximation to Weakly Nonlinear MIMO Systems,” Automatica, 43(10), pp. 1737–1751. [CrossRef]
Wernholt, E., and Gunnarsson, S., 2007, “Analysis of Methods for Multivariable Frequency Response Function Estimation in Closed Loop,” 46th IEEE Conference on Decision and Control, New Orleans, LA, December 12–14, pp. 4881–4888. [CrossRef]
Gutt, H.-J., Scholl, F. D., and Blattner, J., 1996, “High Precision Servo Drives With DSP-Based Torque Ripple Reduction,” IEEE Africon, Stellenbosch, South Africa, September 24–27, Vol. 2, pp. 632–637. [CrossRef]
Hanselman, D., 1990, “Resolver Signal Requirements for High Accuracy Resolver-to-Digital Conversion,” IEEE Trans. Ind. Electron., 37(6), pp. 556–561. [CrossRef]
Crandall, S., 2004, “On Using Non-Gaussian Distributions to Perform Statistical Linearization,” Int. J. Non-Linear Mech., 39(9), pp. 1395–1406. [CrossRef]
Feeny, B., and Moon, F., 1994, “Chaos in a Forced Dry-Friction Oscillator: Experiments and Numerical Modelling,” J. Sound Vib., 170(3), pp. 303–323. [CrossRef]
Moberg, S., and Hanssen, S., 2009, “Inverse Dynamics of Flexible Manipulators,” ECCOMAS Thematic Conference on Multibody Dynamics, Warsaw, Poland, June 29–July 2, .
Moberg, S., Öhr, J., and Gunnarsson, S., 2008, “A Benchmark Problem for Robust Control of a Multivariable Nonlinear Flexible Manipulator,” 17th IFAC World Congress, Seoul, Korea, July 6–11, pp. 1206–1211.
Moberg, S., Öhr, J., and Gunnarsson, S., 2009, “A Benchmark Problem for Robust Feedback Control of a Flexible Manipulator,” IEEE Trans. Control Syst. Technol., 17(6), pp. 1398–1405. [CrossRef]

Figures

Grahic Jump Location
Fig. 1

The IRB4600 robot from ABB

Grahic Jump Location
Fig. 2

Examples of elastic elements modeling torsional deflection in three directions: three nonactuated elastic joints (left), two nonactuated elastic joints together with one elastic joint actuated by a motor (right). Dampers are not shown.

Grahic Jump Location
Fig. 3

Three rigid bodies with fixed coordinate systems A, B, and C (solid). The joint vectors li connect the dashed coordinate systems, starting in N. The transformation (rotation) from the one dashed to the next solid coordinate system is given by the kinematic joint angles.

Grahic Jump Location
Fig. 4

A nine DOF extended flexible joint model with two links, two motors (M), three elastic elements (EE), and three rigid bodies (RB)

Grahic Jump Location
Fig. 5

Closed-loop measurement setup

Grahic Jump Location
Fig. 6

Partly linearized system

Grahic Jump Location
Fig. 7

System description including nonlinearities and unmodeled dynamics

Grahic Jump Location
Fig. 8

Stiffening spring characteristics

Grahic Jump Location
Fig. 9

Nonlinear friction characteristics

Grahic Jump Location
Fig. 11

Normalized stiffness for all 12 springs in case 7. True parameter values are 1. The uncertainty is indicated by two standard deviations.

Grahic Jump Location
Fig. 12

One diagonal FRF element in configuration 1 of the simulation study. The magnitude of FRF from motor torque to motor acceleration is shown.

Grahic Jump Location
Fig. 13

An example of measured signals for one axis in one experiment. The steady-state response from 4 s to 14 s is used for FRF estimation.

Grahic Jump Location
Fig. 14

Illustration of all configurations used for identification (maximum payload) and validation (maximum payload and without payload). The robot joints 2, 3, and 5 are illustrated with a dot, the payload with a circle. Note that validation configurations 4–7 are not identical due to different positions of joint 1, 4, and 6.

Grahic Jump Location
Fig. 15

Cost of all models evaluated for the identification FRF set and in the first validation FRF set, both with maximum payload. Same cost function as used in the identification.

Grahic Jump Location
Fig. 16

Cost of all models evaluated for the identification FRF set and in the first validation FRF set, both with maximum payload, and for the second validation FRF set with no payload. Same cost function as used in the identification, but with all elements including motors 5 and 6 removed (unexcited when no payload).

Grahic Jump Location
Fig. 17

All FRF elements in configuration 14 of identification set, showing the measured FRF (thick) and the 18 DOF model FRF (thin). The shaded area indicates the measured FRF uncertainty (1 std). The magnitude of the 6 × 6 FRF is shown, with input motor torque and output motor acceleration.

Grahic Jump Location
Fig. 18

One diagonal FRF element in configuration 6 of identification set

Grahic Jump Location
Fig. 19

The same FRF element as in Fig. 18 but in configuration 9 of validation set 1

Grahic Jump Location
Fig. 20

One nondiagonal FRF element in configuration 3 of identification set

Grahic Jump Location
Fig. 21

Normalized stiffness for all 12 springs in the 18 DOF model. The uncertainty is indicated by two standard deviations.

Grahic Jump Location
Fig. 22

One diagonal FRF element in configuration 12 of validation set 2. The model with stiffening transmission springs is denoted NLS.

Grahic Jump Location
Fig. 23

Illustration of all configurations used for identification using motor sensor and arm accelerometer (maximum payload)

Grahic Jump Location
Fig. 24

The cost for models obtained by using tool acceleration or motor position as output variables. Cost for both output variables is shown.

Grahic Jump Location
Fig. 25

Stiffness parameters for 15 DOF model. The parameters obtained by using motor position are normalized to one and indicated by a circle. The parameters obtained by tool acceleration are indicated by a square. Two standard deviations are indicated.

Grahic Jump Location
Fig. 26

One FRF element (motor torque to motor acceleration)

Grahic Jump Location
Fig. 27

One FRF element (motor torque to tool acceleration)

Grahic Jump Location
Fig. 28

Real (solid) and simulated (dotted) tool speed in all directions. Eighteen DOF extended flexible joint model used in simulation.

Grahic Jump Location
Fig. 29

Real (solid) and simulated (dotted) motor torque for axes 1, 2, and 3. Eighteen DOF extended flexible joint model used in simulation.

Grahic Jump Location
Fig. 30

Real (solid) and simulated path. Eighteen DOF extended flexible joint model (dotted) and flexible joint model (dashed) used in simulation.

Grahic Jump Location
Fig. 31

Error of simulated model when compared with the real robot path. Eighteen DOF extended flexible joint model (thick solid) and flexible joint model (thin solid).

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In