LOGIN TO YOUR ACCOUNT

Username
Password
Remember Me
Or use your Academic/Social account:

CREATE AN ACCOUNT

Or use your Academic/Social account:

Congratulations!

You have just completed your registration at OpenAire.

Before you can login to the site, you will need to activate your account. An e-mail will be sent to you with the proper instructions.

Important!

Please note that this site is currently undergoing Beta testing.
Any new content you create is not guaranteed to be present to the final version of the site upon release.

Thank you for your patience,
OpenAire Dev Team.

Close This Message

CREATE AN ACCOUNT

Name:
Username:
Password:
Verify Password:
E-mail:
Verify E-mail:
*All Fields Are Required.
Please Verify You Are Human:
fbtwitterlinkedinvimeoflicker grey 14rssslideshare1
Diamond, Alan (2013)
Languages: English
Types: Doctoral thesis
Subjects: TJ0210.2
Introducing robots into human environments requires them to handle settings designed specifically for human size and morphology, however, large, conventional humanoid robots with stiff, high powered joint actuators pose a significant danger to humans. By contrast, “anthropomimetic” robots mimic both human morphology and internal structure; skeleton, muscles, compliance and high redundancy. Although far safer, their resultant compliant structure presents a formidable challenge to conventional control. Here we review, and seek to address, characteristic control issues of this class of robot, whilst exploiting their biomimetic nature by drawing upon biological motor control research. We derive a novel learning controller for discovering effective reaching actions created through sustained activation of one or more muscle synergies, an approach which draws upon strong, recent evidence from animal and humans studies, but is almost unexplored to date in musculoskeletal robot literature. Since the best synergies for a given robot will be unknown, we derive a deliberately simple reinforcement learning approach intended to allow their emergence, in particular those patterns which aid linearization of control. We also draw upon optimal control theories to encourage the emergence of smoother movement by incorporating signal dependent noise and trial repetition.\ud \ud In addition, we argue the utility of developing a detailed dynamic model of a complete robot and present a stable, physics-­‐‑based model, of the anthropomimetic ECCERobot,\ud running in real time with 55 muscles and 88 degrees of freedom. \ud \ud Using the model, we find that effective reaching actions can be learned which employ only two sequential motor co-­‐‑activation patterns, each controlled by just a single common driving signal. Factor analysis shows the emergent muscle co-­‐‑activations can be reconstructed to significant accuracy using weighted combinations of only 13 common fragments, labelled “candidate synergies”. Using these synergies as drivable units the same controller learns the same task both faster and better, however, other reaching tasks perform less well, proportional to dissimilarity; we therefore propose that modifications enabling emergence of a more generic set of synergies are required.\ud \ud Finally, we propose a continuous controller for the robot, based on model predictive control, incorporating our model as a predictive component for state estimation, delay-­‐‑\ud compensation and planning, including merging of the robot and sensed environment into a single model. We test the delay compensation mechanism by controlling a second copy of the model acting as a proxy for the real robot, finding that performance is significantly improved if a precise degree of compensation is applied and show how rapidly an un-­‐‑compensated controller fails as the model accuracy degrades.
  • The results below are discovered through our pilot algorithms. Let us know how we are doing!

    • 6.6.4   Predicting  intended  motor  signals  with  an  updateable    “buffer”  ............... 1 45  
    • 6.6.5   Delay  compensating  design  .............................................................................. 1 46  
    • 6.7   Experiment  exploring  delay  compensation  ........................................................... 1 48  
    • 6.7.3   Characterising  Effects  of  Model  Divergence  .................................................. 1 54  
    • 6.8   Discussion  and  Conclusion  ...................................................................................... 1 56  
    • 7.2   Original  Contributions  of  Thesis  ............................................................................. 1 60  
    • 7.2.3   A   low   dimensional   reaching   controller   for   biomimetic   musculoskeletal  
    • modelled  robot    based  on  extracted  emergent  synergies . ......................................... 1 62  
    • 7.2.4   Optimal   control   principles   can   be   exploited   through   RL   trial   repetition   to  
    • refine  movements . ........................................................................................................... 1 62  
    • 7.2.6   An  MPC-­‐‑based  design  for  continuous  control  of  an  anthropomimetic  robot  
    • incorporating  delay  compensation . .............................................................................. 1 63  
    • Collewijn,  H.,  Erkelens,  C.  &  Steinman,  R.,  1988.  Binocular  co-­‐‑ordination  of  human   horizontal  saccadic  eye  movements.  The  Journal  of  Physiology,  404(1),  pp.157-182.  
    • Cueli,  J.  &  Bordons,  C.,  2008.  Iterative  nonlinear  model  predictive  control.  Stability,   robustness  and  applications.  Control  Engineering  Practice,  16(9),  pp.1023-1034.  
    • D'Avella,  A.  et  al.,  2006.  Control  of  fast-­‐‑reaching  movements  by  muscle  synergy   combinations.  The  Journal  of  neuroscience  :  the  official  journal  of  the  Society  for   Neuroscience,  26(30),  pp.7791-810.  
    • D'Avella,  A.  et  al.,  2008.  Modulation  of  phasic  and  tonic  muscle  synergies  with   reaching  direction  and  speed.  Journal  of  neurophysiology,  100(3),  pp.1433-54.  
    • D'Avella,  A.  &  Bizzi,  E.,  2005a.  Shared  and  specific  muscle  synergies  in  natural  motor   behaviors.  Proceedings  of  the  National  Academy  of  Sciences  of  the  United  States  of   America,  102(8),  pp.3076-3081.  
    • D'Avella,  A.  &  Bizzi,  E.,  2005b.  Shared  and  specific  muscle  synergies  in  natural  motor   behaviors.  Proceedings  of  the  National  Academy  of  Sciences  of  the  United  States  of   America,  102(8),  pp.3076-81.  
    • D'Avella,  A.,  Portone,  A.  &  Lacquaniti,  F.,  2011.  Superposition  and  modulation  of   muscle  synergies  for  reaching  in  response  to  a  change  in  target  location.  Journal  of   neurophysiology,  106(6),  pp.2796-812.  
    • D'Avella,  A.,  Saltiel,  P.  &  Bizzi,  E.,  2003.  Combinations  of  muscle  synergies  in  the   construction  of  a  natural  motor  behavior.  Nature  neuroscience,  6(3),  pp.300-8.  
    • Ganor,  I.  &  Golani,  I.,  1980.  Coordination  and  integration  in  the  hindleg  step  cycle  of   the  rat:  kinematic  synergies.  Brain  research,  195(1),  pp.57-67.  
    • García,  C.,  Prett,  D.  &  Morari,  M.,  1989.  Model  predictive  control:  Theory  and   practice-A  survey.  Automatica,  25(3),  pp.335-348.  
    • Giszter,  S.,  Mussa-­‐‑Ivaldi,  F.  &  Bizzi,  E.,  1993.  Convergent  force  fields  organized  in  the   frog's  spinal  cord.  Journal  of  Neuroscience,  13(2),  pp.467-91.  
    • Jones,  K.,  Hamilton,  A.  &  Wolpert,  D.,  2002.  Sources  of  signal-­‐‑dependent  noise  during   isometric  force  production.  Journal  of  Neurophysiology,  88(3),  pp.1533-1544.  
    • Kagami,  S.  et  al.,  2003.  Humanoid  Arm  Motion  Planning  using  Stereo  Vision  and  RRT   Search.  In  Proceedings  of  the  2003  IEEE/RSJ  International  Conference  on  lntelligent   Robots  and  Systems.  pp.  2167-2172.  
    • Kargo,  W.  et  al.,  2010.  A  Simple  Experimentally  Based  Model  Using  Proprioceptive   Regulation  of  Motor  Primitives  Captures  Adjusted  Trajectory  Formation  in  Spinal   Frogs.  Journal  of  Neurophysiology,  103(1),  pp.573-590.  
    • Kargo,  W.  &  Giszter,  S.,  2000.  Rapid  correction  of  aimed  movements  by  summation  of   force-­‐‑field  primitives.  Journal  of  Neuroscience,  20(1),  pp.409-26.  
    • Microsoft,  2013.  The  Microsoft  Kinect  Sensor.  Available  at:   http://www.microsoft.com/en-­‐‑us/kinectforwindows/  [Accessed  June  10,  2013].  
    • Miyamoto,  H.  et  al.,  2004.  TOPS  (Task  Optimization  in  the  Presence  of  Signal-­‐‑ Dependent  Noise)  model.  Systems  and  Computers  in  Japan,  35(11),  pp.48-58.  
    • Mizuuchi,  I.  et  al.,  2007.  An  advanced  musculoskeletal  humanoid  Kojiro.  In  2007  7th   IEEE-­‐‑RAS  International  Conference  on  Humanoid  Robots.  IEEE,  pp.  294-299.  
    • Neptune,  R.,  Clark,  D.  &  Kautz,  S.,  2009.  Modular  control  of  human  walking:  a   simulation  study.  Journal  of  Biomechanics,  42(9),  pp.1282-1287.  
    • Wan,  E.  &  Van  Der  Merwe,  R.,  2001.  The  Unscented  Kalman  Filter  S.  Haykin,  ed.   Kalman  Filtering  and  Neural  Networks,  5(1),  pp.221-280.  
    • Webb,  B.,  2004.  Neural  mechanisms  for  prediction:  do  insects  have  forward  models?   Trends  in  neurosciences,  27(5),  pp.278-82.  
    • Weiss,  E.J.  &  Flanders,  M.,  2004.  Muscular  and  postural  synergies  of  the  human  hand.   Journal  of  Neurophysiology,  92(1),  pp.523-35.  
  • No related research data.
  • No similar publications.

Share - Bookmark

Download from

Cite this article