Estimating the multivariable human ankle impedance in dorsi-plantarflexion and inversion-eversion directions using EMG signals and artificial neural networks

  • Houman Dallali
  • Lauren Knop
  • Leslie Castelino
  • Evandro Ficanha
  • Mohammad RastgaarEmail author
Regular Paper


The use of a suitably designed ankle-foot prosthesis is essential for transtibial amputees to regain lost mobility. A desired ankle–foot prosthesis must be able to replicate the function of a healthy human ankle by transferring the ground reaction forces to the body, absorbing shock during contact, and providing propulsion. During the swing phase of walking, the human ankle is soft and relaxed; however, it hardens as it bears the body weight and provides force for push-off. The stiffness is one of the components of the mechanical impedance, and it varies with muscle activation (Stochastic estimation of human ankle mechanical impedance in medial-lateral direction, 2014, Stochastic estimation of the multivariable mechanical impedance of the human ankle with active muscles, 2010). This study defines the relationship between ankle impedance and the lower extremity muscle activations using artificial neural networks (ANN). We used the Anklebot, a highly backdrivable, safe, and therapeutic robot to apply stochastic position perturbations to the human ankle in the sagittal and frontal planes. A previously proposed system identification method was used to estimate the target ankle impedance to train the ANN. The ankle impedance was estimated with relaxed muscles and with lower leg muscle activations at 10 and 20% of the maximum voluntary contraction (MVC) of each individual subject. Given the root mean squared (rms) of the electromyography (EMG) signals, the proposed ANN effectively predicted the ankle impedance with mean accuracy of 89.8 ± 6.1% in DP and mean accuracy of 88.3 ± 5.7% in IE, averaged across three muscle activation levels and all subjects.


Electromyography Ankle impedance Human ankle Artificial neural networks 



This work is supported by the National Science Foundation under CAREER grant no. 1350154. The authors would like to thank Chen Jia for his help with the preparation of this paper.


  1. Barbero, M., Merletti, R., Rainoldi, A.: Atlas of muscle innervation zones: understanding surface electromyography and its applications. Springer, Mailand (2012)Google Scholar
  2. Braun, D.J., Mitchell, J.E., Goldfarb, M.: Actuated dynamic walking in a seven-link biped robot. Mechatron. IEEE/ASME Trans. 17(1), 147–156 (2012)CrossRefGoogle Scholar
  3. Chau, T.: A review of analytical techniques for gait data. Part 2: neural network and wavelet methods. Gait Posture 13(2), 102–120 (2001)MathSciNetCrossRefGoogle Scholar
  4. Davis, R.B., DeLuca, P.A.: Gait characterization via dynamic joint stiffness. Gait Posture 4(3), 224–231 (1996)CrossRefGoogle Scholar
  5. Davy, D., Audu, M.: A dynamic optimization technique for predicting muscle forces in the swing phase of gait. J. Biomech. 20(2), 187–201 (1987)CrossRefGoogle Scholar
  6. Ficanha, E., Rastgaar, M., Kaufman, K.R.: Gait emulator for evaluation of ankle--foot prostheses capable of turning. ASME J. Med. Device 9(3), 0309081–0309082 (2015)CrossRefGoogle Scholar
  7. Ficanha, E. M., Rastgaar, M.: Preliminary design and evaluation of a multi-axis ankle--foot prosthesis. In: 5th IEEE RAS/EMBS International Conference on Biomedical Robotics and Biomechatronics, BioRob (2014)Google Scholar
  8. Ficanha, E. M., Rastgaar, M.: Stochastic estimation of human ankle mechanical impedance in medial-lateral direction. In: ASME Dynamic Systems and Control Conference (DSCC) (2014)Google Scholar
  9. Ficanha, E.M., Rastgaar, M., Kaufman, K.R.: A two-axis cable-driven ankle--foot mechanism. Robot. Biomim. 1, 17 (2014)CrossRefGoogle Scholar
  10. Ficanha, E. M., Rastgaar, M., Kaufman, K. R.: Control of a 2-dof powered ankle-foot mechanism. In: IEEE International Conference on Robotics and Automation (ICRA), pp. 6439–6444 (2015)Google Scholar
  11. Ficanha, E. M., Ribeiro, G., Rastgaar, M.: Instrumented walkway for estimation of the ankle impedance in dorsiflexion–plantarflexion and inversion–eversion during standing and walking. In: ASME 2015 Dynamic Systems and Control Conference (2015), American Society of Mechanical Engineers, pp. V003T42A001–V003T42A001 (2015)Google Scholar
  12. Ficanha, E.M., Ribeiro, G.A., Rastgaar, M.: Mechanical impedance of the non-loaded lower leg with relaxed muscles in the transverse plane. Front. Bioeng. Biotechnol. 3, 198 (2015)CrossRefGoogle Scholar
  13. Ficanha, E.M., Ribeiro, G.A., Rastgaar, M.: Design and evaluation of a 2-dof instrumented platform for estimation of the ankle mechanical impedance in the sagittal and frontal planes. IEEE-ASME Trans. Mechatron. 21(5), 2531–2542 (2016)CrossRefGoogle Scholar
  14. Funahashi, K.I.: On the approximate realization of continuous mappings by neural networks. Neural Netw. 2(3), 183–192 (1989)CrossRefGoogle Scholar
  15. Glaister, B., Bernatz, G., Klute, G., Orendurff, M.: Video task analysis of turning during activities of daily living. Gait Posture 25(2), 289–94 (2007)CrossRefGoogle Scholar
  16. Glaister, B., Schoen, A., Orendurff, M.S., Klute, G.K.: A mechanical model of the human ankle in the transverse plane during straight walking: implications for prosthetic design. J. Biomech. Eng. 131, 3 (2009)Google Scholar
  17. Gumerov, N.A., Duraiswami, R.: Fast radial basis function interpolation via preconditioned Krylov iteration. SIAM J. Sci. Comput. 29(5), 1876–1899 (2007)MathSciNetCrossRefzbMATHGoogle Scholar
  18. Hansen, A.H., Childress, D.S., Miff, S.C., Gard, S.A., Mesplay, K.P.: The human ankle during walking: implications for design of biomimetic ankle prostheses. J. Biomech. 37(10), 1467–1474 (2004)CrossRefGoogle Scholar
  19. Hincapie, J.G., Kirsch, R.F.: Feasibility of EMG-based neural network controller for an upper extremity neuroprosthesis. IEEE Trans. Neural Syst. Rehabil. Eng. 17(1), 80–90 (2009)CrossRefGoogle Scholar
  20. Hinton, G.E.: Connectionist learning procedures. Artif. Intell. 40(1), 185–234 (1989)CrossRefGoogle Scholar
  21. Ho, P., Lee, H., Krebs, H. I., Hogan, N.: Directional variation of active and passive ankle static impedance. In: ASME Dynamic Systems and Control Conference (2009)Google Scholar
  22. Hunter, I., Kearney, R.: Dynamics of human ankle stiffness: variation with mean ankle torque. J. Biomech. 15(10), 747–752 (1982)CrossRefGoogle Scholar
  23. Iqbal, K., Roy, A.: A novel theoretical framework for the dynamic stability analysis, movement control, and trajectory generation in a multisegment biomechanical model. J. Biomech. Eng. 131(1), 011002 (2009)CrossRefGoogle Scholar
  24. Jinoh, L., Pyung Hun, C., Jamisola, R.S.: Relative impedance control for dual-arm robots performing asymmetric bimanual tasks. Ind. Electron. IEEE Trans. 61(7), 3786–3796 (2014)CrossRefGoogle Scholar
  25. Karsoliya, S.: Approximating number of hidden layer neurons in multiple hidden layer BPNN architecture. Int. J. Eng. Trends Technol. 3(6), 713–717 (2012)Google Scholar
  26. Kearney, R., Hunter, I.: Dynamics of human ankle stiffness: variation with displacement amplitude. J. Biomech. 15(10), 753–756 (1982)CrossRefGoogle Scholar
  27. Kearney, R.E., Stein, R.B., Parameswaran, L.: Identification of intrinsic and reflex contributions to human ankle stiffness dynamics. Biomed. Eng. IEEE Trans. 44(6), 493–504 (1997)CrossRefGoogle Scholar
  28. Kim, H.K., Kang, B., Kim, B., Park, S.: Estimation of multijoint stiffness using electromyogram and artificial neural network. IEEE Trans. Syst. Man Cybern. Part A Syst Hum 39(5), 972–980 (2009)CrossRefGoogle Scholar
  29. Kleijnen, J.P.C.: Kriging metamodeling in simulation: a review. Eur. J. Oper. Res. 192(3), 707–716 (2009)MathSciNetCrossRefzbMATHGoogle Scholar
  30. Koike, Y., Kawato, M., van Soest, A., Stroeve, S.: Estimation of movement from surface EMG signals using a neural network model. Springer, New York, pp. 440–457 (2000)Google Scholar
  31. Lamontagne, A., Malouin, F., Richards, C.L.: Viscoelastic behavior of plantar flexor muscle-tendon unit at rest. J. Orthop. Sports Phys. Ther. 26(5), 244–252 (1997)CrossRefGoogle Scholar
  32. Lee, H., Ho, P., Rastgaar, M., Krebs, H. I., Hogan, N.: Quantitative characterization of steady-state ankle impedance with muscle activation. In: ASME Dynamic Systems and Control Conference (2010)Google Scholar
  33. Lee, H., Ho, P., Rastgaar, M., Krebs, H.I., Hogan, N.: Multivariable static ankle mechanical impedance with relaxed muscles. J. Biomech. 44, 1901–1908 (2011)CrossRefGoogle Scholar
  34. Lee, H., Ho, P., Rastgaar, M., Krebs, H.I., Hogan, N.: Multivariable static ankle mechanical impedance with active muscles. IEEE Trans. Neural Syst. Rehabil. Eng. 22(1), 44–52 (2014)CrossRefGoogle Scholar
  35. Lee, H., Hogan, N.: Time-varying ankle mechanical impedance during human locomotion. IEEE Trans. Neural Syst. Rehabil. Eng. 62, 755–764 (2014)Google Scholar
  36. Lee, H., Krebs, H.I., Hogan, N.: Multivariable dynamic ankle mechanical impedance with active muscles. Neural Syst. Rehabil. Eng. IEEE Trans. 22(5), 971–981 (2014)CrossRefGoogle Scholar
  37. Lee, H., Krebs, H.I., Hogan, N.: Multivariable dynamic ankle mechanical impedance with relaxed muscles. IEEE Trans. Neural Syst. Rehabil. Eng. 22(6), 1104–1114 (2014)CrossRefGoogle Scholar
  38. Lee, H., Wang, S., Hogan, N.: Relationship between ankle stiffness structure and muscle activation. In: Engineering in Medicine and Biology Society (EMBC), 2012 Annual International Conference of the IEEE, pp. 4879–4882 (2012)Google Scholar
  39. Ludvig, D., Kearney, R.E.: Real-time estimation of intrinsic and reflex stiffness. IEEE Trans. Biomed. Eng. 54(10), 1875–1884 (2007)CrossRefGoogle Scholar
  40. Ohno-Machado, L., Rowland, T.: Neural network applications in physical medicine and rehabilitation1. Am J. Phys. Med. Rehabil. 78(4), 392–398 (1999)CrossRefGoogle Scholar
  41. Osu, R., Gomi, H.: Multijoint muscle regulation mechanisms examined by measured human arm stiffness and EMG signals. J. Neurophysiol. 81(4), 1458–1468 (1999)Google Scholar
  42. Palmer, M.L.: Sagittal plane characterization of normal human ankle function across a range of walking gait speeds. University of Utah, Thesis (2002)Google Scholar
  43. Rastgaar, M., Ho, P., Lee, H., Krebs, H. I., Hogan, N.: Stochastic estimation of multi-variable human ankle mechanical impedance. In: ASME Dynamic Systems and Control Conference (2009)Google Scholar
  44. Rastgaar, M., Ho, P., Lee, H., Krebs, H. I., Hogan, N.: Stochastic estimation of the multi-variable mechanical impedance of the human ankle with active muscles. In: ASME Dynamic Systems and Control Conference (2010)Google Scholar
  45. Rastgaar, M., Lee, H., Ficanha, E.M., Ho, P., Krebs, H.I., Hogan, N.: Multi-directiOnal Dynamic Mechanical Impedance Of The Human Ankle A Key To Anthropomorphism In Lower Extremity Assistive Robots. Springer, New York (2014). (book section 4)CrossRefGoogle Scholar
  46. Roy, A., Krebs, H.I., Williams, D.J., Bever, C.T., Forrester, L.W., Macko, R.M., Hogan, N.: Robot-aided neurorehabilitation: a novel robot for ankle rehabilitation. IEEE Trans. Robot. 25(3), 569–582 (2009)CrossRefGoogle Scholar
  47. Schöllhorn, W.I.: Applications of artificial neural nets in clinical biomechanics. Clin. Biomech. 19(9), 876–898 (2004)CrossRefGoogle Scholar
  48. Sepulveda, F., Wells, D.M., Vaughan, C.L.: A neural network representation of electromyography and joint dynamics in human gait. J. Biomech. 26(2), 101–109 (1993)CrossRefGoogle Scholar
  49. Sheela, K. G., Deepa, S.: Review on methods to fix number of hidden neurons in neural networks. Math. Probl. Eng. (2013). doi: 10.1155/2013/425740
  50. Sinkjaer, T., Toft, E., Andreassen, S., Hornemann, B.C.: Muscle stiffness in human ankle dorsiflexors: intrinsic and reflex components. J. Neurophysiol. 60(3), 1110–1121 (1988)Google Scholar
  51. Smith, D.G., Fergason, J.R.: Transtibial amputations. Clin. Ortho. Relat. Res. 361, 108–115 (1999)CrossRefGoogle Scholar
  52. Soares, A., Andrade, A., Lamounier, E., Carrijo, R.: The development of a virtual myoelectric prosthesis controlled by an EMG pattern recognition system based on neural networks. J. Intell. Inf. Syst. 21(2), 127–141 (2003)CrossRefGoogle Scholar
  53. Taylor, M.J.D., Dabnichki, P., Strike, S.C.: A three-dimensional biomechanical comparison between turning strategies during the stance phase of walking. Hum. Mov. Sci. 24, 558–573 (2005)CrossRefGoogle Scholar
  54. Todorov, E.: Optimality principles in sensorimotor control. Nat. Neurosci. 7(9), 907–915 (2004)CrossRefGoogle Scholar
  55. Unuma, M., Anjyo, K., Takeuchi, R.: Fourier principles for emotion-based human figure animation. In: Proceedings of the 22nd annual conference on Computer graphics and interactive techniques (1995), ACM, pp. 91–96 (1995)Google Scholar
  56. Wang, L., Buchanan, T.S.: Prediction of joint moments using a neural network model of muscle activations from EMG signals. IEEE Trans. Neural Syst. Rehabil. Eng. 10(1), 30–37 (2002)CrossRefGoogle Scholar
  57. Weiss, P., Kearney, R., Hunter, I.: Position dependence of ankle joint dynamics—I. Passive mechanics. J. Biomech. 19(9), 727–735 (1986)CrossRefGoogle Scholar
  58. Weiss, P.L., Kearney, R., Hunter, I.: Position dependence of ankle joint dynamics—II. Active mechanics. J. Biomech. 19(9), 737–751 (1986)CrossRefGoogle Scholar
  59. Weyand, P.G., Sternlight, D.B., Bellizzi, M.J., Wright, S.: Faster top running speeds are achieved with greater ground forces not more rapid leg movements. J. Appl. Physiol. 89, 1991–1999 (2000)Google Scholar
  60. Ziegler-Graham, K., MacKenzie, E.J., Ephraim, P.L., Travison, T.G., Brookmeyer, R.: Estimating The Prevalence Of Limb Loss In The United States: 2005 To 2050. Arch. Phys. Medi. Rehabil. 89(3), 422–429 (2008)CrossRefGoogle Scholar
  61. Zinder, S.M., Granata, K.P., Padua, D.A., Gansneder, B.M.: Validity and reliability of a new in vivo ankle stiffness measurement device. J. Biomech. 40(2), 463–467 (2007)CrossRefGoogle Scholar

Copyright information

© Springer Singapore 2017

Authors and Affiliations

  • Houman Dallali
    • 1
  • Lauren Knop
    • 1
  • Leslie Castelino
    • 1
  • Evandro Ficanha
    • 1
  • Mohammad Rastgaar
    • 1
    Email author
  1. 1.Department of Mechanical Engineering-Engineering MechanicsMichigan Technological UniversityHoughtonUSA

Personalised recommendations