Bio-inspired motion control of the musculoskeletal BioBiped1 robot based on a learned inverse dynamics model

Based on the central hypothesis that a humanoid robot with human-like walking and running performance requires a bio-inspired embodiment of the musculoskeletal functions of the human leg as well as of its control structure, a bio-inspired approach for joint position control of the BioBiped1 robot is presented in this paper. This approach combines feed-forward and feedback control running at 1 kHz and 40 Hz, respectively. The feed-forward control is based on an inverse dynamics model which is learned using Gaussian process regression to account for the robot's body dynamics and external influences. For evaluation the learned model is used to control the robot purely feed-forward as well as in combination with a slow feedback controller. Both approaches are compared to a basic feedback PD-controller with respect to their tracking ability in experiments. It is shown, that the combined approach yields good results and outperforms the basic feedback controller when applied to the same set-point trajectories for the leg joints.

[1]  Oskar von Stryk,et al.  Concept and Design of the BioBiped1 Robot for Human-like Walking and Running , 2011, Int. J. Humanoid Robotics.

[2]  Koh Hosoda,et al.  Pneumatic-driven jumping robot with anthropomorphic muscular skeleton structure , 2010, Auton. Robots.

[3]  B. Webb.  Neural mechanisms for prediction: do insects have forward models? , 2004, Trends in Neurosciences.

[4]  Christopher K. I. Williams,et al.  Gaussian Processes for Machine Learning (Adaptive Computation and Machine Learning) , 2005 .

[5]  E. Zehr,et al.  What functions do reflexes serve during human locomotion? , 1999, Progress in Neurobiology.

[6]  Duy Nguyen-Tuong,et al.  Local Gaussian Process Regression for Real Time Online Model Learning , 2008, NIPS.

[7]  Byung Kook Yoo,et al.  Adaptive control of robot manipulator using fuzzy compensator , 2000, IEEE Trans. Fuzzy Syst..

[8]  M. Bobbert,et al.  Mechanical output from individual muscles during explosive leg extensions: the role of biarticular muscles. , 1996, Journal of biomechanics.

[9]  Bernhard Schölkopf,et al.  Learning Inverse Dynamics: a Comparison , 2008, ESANN.

[10]  Sebastian Klug,et al.  Biologically Inspired Robot Manipulator for New Applications in Automation Engineering , 2008 .

[11]  Christopher G. Atkeson,et al.  Compliant control of a hydraulic humanoid joint , 2007, 2007 7th IEEE-RAS International Conference on Humanoid Robots.

[12]  Koh Hosoda,et al.  3D bipedal robot with tunable leg compliance mechanism for multi-modal locomotion , 2008, 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[13]  D M Wolpert,et al.  Multiple paired forward and inverse models for motor control , 1998, Neural Networks.

[14]  I. K. Wood,et al.  Neuroscience: Exploring the brain , 1996 .

[15]  M. Bobbert,et al.  The unique action of bi-articular muscles in complex movements. , 1987, Journal of anatomy.

[16]  Reinhard Blickhan,et al.  Compliant leg behaviour explains basic dynamics of walking and running , 2006, Proceedings of the Royal Society B: Biological Sciences.

[17]  E. Taub,et al.  Deafferentation in monkeys: Pointing at a target without visual feedback , 1975, Experimental Neurology.

[18]  A. V. Voronov,et al.  The Roles of Monoarticular and Biarticular Muscles of the Lower Limbs in Terrestrial Locomotion , 2004, Human Physiology.

[19]  Matthew M. Williamson,et al.  Series elastic actuators , 1995, Proceedings 1995 IEEE/RSJ International Conference on Intelligent Robots and Systems. Human Robot Interaction and Cooperative Robots.