Learning dynamic arm motions for postural recovery

The biomechanics community has recently made progress toward understanding the role of rapid arm movements in human stability recovery. However, comparatively little work has been done exploring this type of control in humanoid robots. We provide a summary of recent insights into the functional contributions of arm recovery motions in humans and experimentally demonstrate advantages of this behavior on a dynamically stable mobile manipulator. Using Bayesian optimization, the robot efficiently discovers policies that reduce total energy expenditure and recovery footprint, and increase ability to stabilize after large impacts.

[1]  Harold J. Kushner,et al.  A New Method of Locating the Maximum Point of an Arbitrary Multipeak Curve in the Presence of Noise , 1964 .

[2]  B. E. Maki,et al.  Early activation of arm muscles follows external perturbation of upright stance , 1995, Neuroscience Letters.

[3]  O. SIAMJ.,et al.  A CLASS OF GLOBALLY CONVERGENT OPTIMIZATION METHODS BASED ON CONSERVATIVE CONVEX SEPARABLE APPROXIMATIONS∗ , 2002 .

[4]  J. Allum,et al.  Age‐dependent variations in the directional sensitivity of balance corrections and compensatory arm movements in man , 2002, The Journal of physiology.

[5]  Taku Komura,et al.  The dynamic postural adjustment with the quadratic programming method , 2002, IEEE/RSJ International Conference on Intelligent Robots and Systems.

[6]  A. Patla,et al.  Role of the unperturbed limb and arms in the reactive recovery response to an unexpected slip during locomotion. , 2003, Journal of neurophysiology.

[7]  J. Misiaszek Early activation of arm and leg muscles following pulls to the waist during walking , 2003, Experimental Brain Research.

[8]  Robert O. Ambrose,et al.  Mobile manipulation using NASA's Robonaut , 2004, IEEE International Conference on Robotics and Automation, 2004. Proceedings. ICRA '04. 2004.

[9]  J. Misiaszek,et al.  Restricting arm use enhances compensatory reactions of leg muscles during walking , 2005, Experimental Brain Research.

[10]  Sergey V. Drakunov,et al.  Capture Point: A Step toward Humanoid Push Recovery , 2006, 2006 6th IEEE-RAS International Conference on Humanoid Robots.

[11]  Taku Komura,et al.  Stepping motion for a human-like character to maintain balance against large perturbations , 2006, Proceedings 2006 IEEE International Conference on Robotics and Automation, 2006. ICRA 2006..

[12]  Christopher G. Atkeson,et al.  Multiple balance strategies from one optimization criterion , 2007, 2007 7th IEEE-RAS International Conference on Humanoid Robots.

[13]  J. Laumond,et al.  Motion Planning for Humanoid Robots : Highlights with HRP-2 , 2007 .

[14]  Marcus R. Frean,et al.  Using Gaussian Processes to Optimize Expensive Functions , 2008, Australasian Conference on Artificial Intelligence.

[15]  D. Lizotte Practical bayesian optimization , 2008 .

[16]  G. Caldwell,et al.  Predicting dynamic postural instability using center of mass time-to-contact information. , 2008, Journal of biomechanics.

[17]  M. P. Mcguigan,et al.  The role of arm movement in early trip recovery in younger and older adults. , 2008, Gait & posture.

[18]  Victor B. Zordan,et al.  Momentum control for balance , 2009, ACM Trans. Graph..

[19]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[20]  Jessica K. Hodgins,et al.  Simulating balance recovery responses to trips based on biomechanical principles , 2009, SCA '09.

[21]  M. Grabiner,et al.  Theoretical contribution of the upper extremities to reducing trunk extension following a laboratory-induced slip. , 2009, Journal of biomechanics.

[22]  I. Kingma,et al.  Armed against falls: the contribution of arm movements to balance recovery after tripping , 2010, Experimental Brain Research.

[23]  Mike Stilman,et al.  Golem Krang: Dynamically stable humanoid robot for mobile manipulation , 2010, 2010 IEEE International Conference on Robotics and Automation.

[24]  Ying Zheng,et al.  Contact feature extraction on a balancing manipulation platform , 2010, 2010 IEEE International Conference on Robotics and Automation.

[25]  Nando de Freitas,et al.  A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning , 2010, ArXiv.

[26]  Scott Kuindersma Control Model Learning for Whole-Body Mobile Manipulation , 2010, AAAI.

[27]  Christopher G. Atkeson,et al.  Push Recovery by stepping for humanoid robots with force controlled joints , 2010, 2010 10th IEEE-RAS International Conference on Humanoid Robots.