IEEE Robotics & Automation Magazine - December 2017 - 57

During the Cybathlon, we did not face any problems
with or malfunctions of the device. Our pilot was able to
achieve his personal record right at the small final with a
time of 6 min and 44 s for the race distance of 750 m.
During all races, he showed a very smooth cycling movement and did not need to push his legs manually during
dead zones (Figure 11).
Discussion and Conclusions
With our work, we showed that inertial-sensor-based jointangle estimation can be used for realizing FES cycling. Due
to the cycling device's independence, the same setting and
stimulation pattern can be used either for regular or ergometer-based cycling. As shown in Figure 5, the same stimulation setting for the CP area can be used for very different
cycling positions, and it is still possible to achieve a cycling
movement. For a crank-angle-based pattern, the stimulation areas would probably need to be adjusted to achieve a
cycling motion for each cycling position, as the angle ranges vary by up to 35° with respect to the crank angle.
Regarding the bike architecture, our main focus was on
using as many standard components as possible and
enabling an easy and autonomous transfer. Among the 12
Cybathlon FES bike race pilots, only our pilot and the pilots
of the two teams that used the BerkelBike (BerkelBike BV,
Sint-Michielsgestel, The Netherlands) could get onto the tricycle without assistance by a third person. For reaching the
podium the next time, the focus should be changed to a
weight-optimized and streamlined bike. The presented inertial-sensor-based stimulation pattern ensures positive crank
torques that support cycling but might be further optimized
inside the given stimulation ranges, depending on the rider-
tricycle geometry, to improve mechanic and metabolic efficiency. For future developments, we plan to reduce the
number of inertial sensors used for ergometer cycling to two
IMU sensors. Furthermore, we are going to reinvestigate the
influence of the electrode placement to avoid the usage of
the gluteus maximus for cycling. Although the joint-angle
estimation framework used reduces the tuning effort, some
other manual parameter changes for the speed compensation and the maximum current for each muscle group had
to be made according to the daily performance. To reduce
this effort, a learning algorithm that attempts to automatically optimize the stimulation pattern after each crank revolution will be implemented.
Acknowledgments
We greatly appreciate the commitment of our pilot, Hanno
Voigt, who followed such an intensive and ambitious training
plan. We very much appreciated his recommendations on
how to improve the cycling device and the algorithms. We
would also like to thank Dr. Andreas Niedeggen (Unfallkrankenhaus Berlin, Germany) for his medical supervision of the
pilot and the company Hempel Gesundheitspartner for manufacturing and customizing the sensor sleeves and the tricycle
cushion. Likewise, we are very grateful that Margit Gföhler

kindly provided us with the ankle-joint orthoses (produced
by the Technische Universität Wien). Finally, we would like to
thank the company Hasomed GmbH for setting up Team
Hasomed and equipping us with the tricycle, the stimulator,
and electrodes as well providing the necessary funds.
References
[1] W. A. Bauman, A. M. Spungen, M. Raza, J. Rothstein, R. L. Zhang, Y.
G. Zhong, M. Tsuruta, R. Shahidi, R. N. Pierson, and J. Wang, "Coronary artery disease: Metabolic risk factors and latent disease in individuals with paraplegia," Mount Sinai J. Medicine, vol. 59, no. 2, pp.
163-168, Mar. 1992.
[2] S. A. Bloomfield, W. J. Mysiw, and R. D. Jackson, "Bone mass and
endocrine adaptations to training in spinal cord injured individuals,"
Bone, vol. 19, no. 1, pp. 61-68, July 1996.
[3] D. J. Newham, N. De, and N. de N. Donaldson, "FES cycling," Acta
Neurochirurgica Suppl., vol. 97, no. 1, pp. 395-402, 2007.
[4] A. S. Gorgey, D. R. Dolbow, J. D. Dolbow, R. K. Khalil, C. Castillo,
and D. R. Gater, "Effects of spinal cord injury on body composition and
metabolic profile-Part I," J. Spinal Cord Medicine, vol. 37, no. 6, pp.
693-702, 2014.
[5] A. S. Gorgey, D. R. Dolbow, J. D. Dolbow, R. K. Khalil, and D. R.
Gater, "The effects of electrical stimulation on body composition and
metabolic profile after spinal cord injury-Part II," J. Spinal Cord Medicine, vol. 38, no. 1, pp. 23-37, 2015.
[6] K. J. Hunt, J. Fang, J. Saengsuwan, M. Grob, and M. Laubacher.
(2012). On the efficiency of FES cycling: A framework and systematic
review. Technol. Health Care. [Online]. 20(5), pp. 395-422. Available:
http://www.ncbi.nlm.nih.gov/pubmed/23079945
[7] S. Ruppin, C. Wiesener, and T. Schauer. (2016). Inertial sensor-based
control of functional electrical stimulation in paraplegic cycling. 20th
Annu. Conf. Int. Functional Elect. Stimulation Soc. (IFESS). [Online]. Available: https://ifess2016.inria.fr/files/2016/02/IFESS_2016_paper_20.pdf
[8] C. Wiesener, S. Ruppin, and T. Schauer, "Robust discrimination of
flexion and extension phases for mobile functional electrical stimulation (FES) induced cycling in paraplegics," IFAC-PapersOnLine, vol. 49,
no. 32, pp. 210-215, 2016.
[9] R. Riener and T. Fuhr, "Patient-driven control of FES-supported
standing up: A simulation study," IEEE Trans. Rehabil. Eng., vol. 6, no.
2, pp. 113-124, June 1998.
[10] T. Schauer, "Feedback control of cycling in spinal cord injury using
functional electrical stimulation," Ph.D. dissertation, College of Science
and Engineering/School of Engineering, Univ. of Glasgow, Scotland, 2005.
[11] M. Sojka and P. Píša, "Usable Simulink embedded coder target for
Linux," in Proc. 16th Real Time Linux Workshop, 2014, pp. 1-6.
[12] Concerning Medical Devices, European Union Council Directive
93/42/EEC, 1993.
[13] Robots and Robotic Devices-Safety Requirements for Personal
Care Robots, ISO Standard 13482:2014.

Constantin Wiesener, Technische Universität Berlin, Germany.
E-mail: wiesener@control.tu-berlin.de.
Thomas Schauer, Technische Universität Berlin, Germany.
E-mail: schauer@control.tu-berlin.de.

DECEMBER 2017

IEEE ROBOTICS & AUTOMATION MAGAZINE

http://www.ncbi.nlm.nih.gov/pubmed/23079945 https://ifess2016.inria.fr/files/2016/02/IFESS_2016_paper_20.pdf

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - December 2017