Abstract
This study proposes an adaptive control architecture based on an accurate regression method called Locally Weighted Projection Regression (LWPR) and on a bio-inspired module, such as a cerebellar-like engine. This hybrid architecture takes full advantage of the machine learning module (LWPR kernel) to abstract an optimized representation of the sensorimotor space while the cerebellar component integrates this to generate corrective terms in the framework of a control task. Furthermore, we illustrate how the use of a simple adaptive error feedback term allows to use the proposed architecture even in the absence of an accurate analytic reference model. The presented approach achieves an accurate control with low gain corrective terms (for compliant control schemes). We evaluate the contribution of the different components of the proposed scheme comparing the obtained performance with alternative approaches. Then, we show that the presented architecture can be used for accurate manipulation of different objects when their physical properties are not directly known by the controller. We evaluate how the scheme scales for simulated plants of high Degrees of Freedom (7-DOFs).
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Albus JS: A theory of cerebellar function. Math Biosci 10(1–2), 25–61 (1971)
Albus JS: A new approach to manipulator control: the cerebellar model articulation controller (CMAC). J Dyn Syst Meas Control 97(3), 220–227 (1975)
Atkeson CG, Hale JG, Pollick F, Riley M, Kotosaka S, Schaul S, Shibata T, Tevatia G, Ude A, Vijayakumar S, Kawato E, Kawato M: Using humanoid robots to study human behavior. IEEE Intell Syst Appl 15(4), 46–56 (2000)
Attwell P, Cooke S, Yeo C: Cerebellar function in consolidation of a motor memory. Neuron 34(6), 1011–1020 (2002)
Boyden E, Katoh A, Raymond J: Cerebellum-dependent learning: the role of multiple plasticity mechanisms. Neuroscience 27(1), 581–609 (2004)
Carrillo R, Ros E, Boucheny C, Coenen O: A real-time spiking cerebellum model for learning robot control. Biosystems 94(1–2), 18–27 (2008)
Corke PI: A robotics toolbox for matlab. IEEE Robotics Autom Mag 3(1), 24–32 (1996)
Craig JJ: Introduction to robotics: mechanics and control, 3rd edn. Pearson/Prentice Hall, Upper Saddle River (2005)
Dean P, Porrill J, Ekerot C, Jörntell H: The cerebellar microcircuit as an adaptive filter: experimental and computational evidence. Nat Rev Neurosci 11(1), 30–43 (2010)
Fujita M: Adaptive filter model of the cerebellum. Biol Cybern 206(3), 195–206 (1982)
German Aerospace Center (2011) DLR Light-Weight Robot (LWR). http://www.dlr.de/rm/en/desktopdefault.aspx/tabid-3803/6175_read-8963/. Accessed 11 August 2011
Gomi H, Kawato M: Adaptive feedback control models of the vestibulocerebellum and spinocerebellum. Biol Cybern 68(2), 105–114 (1992)
Haddadin S, Albu-SchSffer A, Hirzinger G (2007) Safe physical human-robot interaction: Measurements, analysis and new insights. In: Kaneko M, Nakamura Y (eds) ISRR, Springer, Springer Tracts in Advanced Robotics, vol 66, pp 395–407
Haith A, Vijayakumar S: Implications of different classes of sensorimotor disturbance for cerebellar-based motor learning models. Biol Cybern 100(1), 81–95 (2009)
Hirzinger G, Butterfaß J, Fischer M, Grebenstein M, Hähnle M, Liu H, Schäfer I, Sporer N (2000) A mechatronics approach to the design of light-weight arms and multifingered hands. In: ICRA, pp 46–54
Honda T, Yamazaki T, Tanaka S, Nishino T (2010) A possible mechanism for controlling timing representation in the cerebellar cortex. In: International Symposium on Neural Networks, pp 67–76
Ito M: Mechanisms of motor learning in the cerebellum. Brain Res 886(1-2), 237–245 (2000)
Ito M: Control of mental activities by internal models in the cerebellum. Nat Rev Neurosci 9(4), 304–313 (2008)
Jordan MI (1996) Computational aspects of motor control and motor learning. In: Heuer, H and Keele, S (ed) Handbook of perception and action: Motor Skills, Academic Press, New York, vol 2, pp 71–120
Jordan MI, Rumelhart DE: Forward models: supervised learning with a distal teacher. Cognit Sci 16, 307–354 (1992)
Kawato M (1990) Feedback-error-learning neural network for supervised motor learning. In: R Eckmiller (ed) Advanced neural computers, Elsevier, North-Holland, pp 365–372
Kawato M: Internal models for motor control and trajectory planning. Curr Opin Neurobiol 9(6), 718–727 (1999)
Lonini L, Dipietro L, Zollo L, Guglielmelli E, Krebs HI: An internal model for acquisition and retention of motor learning during arm reaching. Neural Comput 21(7), 2009–2027 (2009)
Luque N, Garrido J, Carrillo R, Coenen O, Ros E: Cerebellarlike corrective model inference engine for manipulation tasks. IEEE Trans Syst Man Cybern B 41(5), 1299–1312 (2011a)
Luque N, Garrido J, Carrillo RR, Tolu S, Ros E: Adaptive cerebellar spiking model embedded in the control loop: context switching and robustness against noise. Int J Neural Syst 21(5), 385–401 (2011b)
Luque NR, Garrido JA, Carrillo RR, Coenen OJMD, Ros E: Cerebellar input configuration toward object model abstraction in manipulation tasks. IEEE Trans Neural Netw 22(8), 1321–1328 (2011c)
Marr D: A theory of cerebellar cortex. J Physiol 202, 437–470 (1969)
Masuda N, Amari S: A computational study of synaptic mechanisms of partial memory transfer in cerebellar vestibulo-ocular-reflex learning. J Comput Neurosci 24(2), 137–156 (2008)
Miyamura A, Kimura H: Stability of feedback error learning scheme. Syst Control Lett 45(4), 303–316 (2002)
Nakanishi J, Schaal S: Feedback error learning and nonlinear adaptive control. Neural Netw 17(10), 1453–1465 (2004)
Nguyen-Tuong D, Peters J (2008) Learning robot dynamics for computed torque control using local gaussian processes regression. In: Proceedings of the 2008 ECSIS Symposium on Learning and Adaptive Behaviors for Robotic Systems, IEEE Computer Society, Washington, DC, USA, pp 59–64
Philipona D, Coenen OJMD: Model of granular layer encoding of the cerebellum. Neurocomputing 58(60), 575–580 (2004)
Porrill J, Dean P: Recurrent cerebellar loops simplify adaptive control of redundant and nonlinear motor systems. Neural Comput 19(1), 170–193 (2007)
Porrill J, Dean P, Stone J: Recurrent cerebellar architecture solves the motor-error problem. Proc Biol Sci 271(1541), 789–796 (2004)
Schaal S, Atkeson CG, Vijayakumar S: Scalable techniques from nonparametric statistics for real time robot learning. Appl Intell 17(1), 49–60 (2002)
Schweighofer N, Spoelstra J, Arbib MA, Kawato M: Role of the cerebellum in reaching movements in humans. II. A neural model of the intermediate cerebellum. Eur J Neurosci 10(1), 95–105 (1998)
Schweighofer N, Doya K, Lay F: Unsupervised learning of granule cell sparse codes enhances cerebellar adaptive control. Neuroscience 103, 35–50 (2001)
Sejnowski TJ: Storing covariance with nonlinearly interacting neurons. J Math Biol 4, 303–321 (1977)
Shibata T, Schaal S: Biomimetic gaze stabilization based on feedback-error-learning with nonparametric regression networks. Neural Netw 14(2), 201–216 (2001)
Smola AJ, Schölkopf B: A tutorial on support vector regression. Stat Comput 14(3), 199–222 (2004)
Van der Smagt P: Cerebellar control of robot arms. Connect Sci 10, 301–320 (1998)
Van der Smagt P, Groen F, Schulten K: Analysis and control of a rubbertuator arm. Biol Cybern 75(5), 433–440 (1996)
Vijayakumar S, Schaal S (2000) Locally weighted projection regression: Incremental real time learning in high dimensional space. In: ICML ’00: Proceedings of the Seventeenth International Conference on Machine Learning, Morgan Kaufmann Publishers Inc., San Francisco, CA, USA, pp 1079–1086
Vijayakumar S, D’Souza A, Schaal S: Incremental online learning in high dimensions. Neural Comput 17(12), 2602–2634 (2005)
Williams CKI, Rasmussen CE (1996) Gaussian processes for regression. In: Advances in neural information processing systems 8. MIT press, pp 514–520
Wolpert DM: Computational approaches to motor control. Trends Cogn Sci 1(6), 209–216 (1997)
Wolpert DM, Miall RC, Kawato M: Internal models in the cerebellum. Trends Cogn Sci 2(9), 338–347 (1998)
Wulff P, Schonewille M, Renzi M, Viltono L, Sassoè-Pognetto M, Badura A, Gao Z, Hoebeek FE, van Dorp S, Wisden W, Farrant M, De Zeeuw CI: Synaptic inhibition of purkinje cells mediates consolidation of vestibulo-cerebellar motor learning. Nat Neurosci 12(8), 1042–1049 (2009)
Yamazaki T, Tanaka S: The cerebellum as a liquid state machine. Neural Netw 20(3), 290–297 (2007)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Tolu, S., Vanegas, M., Luque, N.R. et al. Bio-inspired adaptive feedback error learning architecture for motor control. Biol Cybern 106, 507–522 (2012). https://doi.org/10.1007/s00422-012-0515-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00422-012-0515-5