Machine Learning Simulation: Torso Dynamics of Robotic Biped

Simple item page
dc.contributor.authorRenner, Michael Roberten
dc.contributor.committeechairGranata, Kevin P.en
dc.contributor.committeememberHong, Dennis W.en
dc.contributor.committeememberKasarda, Mary E.en
dc.contributor.committeememberReinholtz, Charles F.en
dc.contributor.committeememberSandu, Corinaen
dc.contributor.departmentMechanical Engineeringen
dc.date.accessioned2014-03-14T20:43:40Zen
dc.date.adate2007-08-22en
dc.date.available2014-03-14T20:43:40Zen
dc.date.issued2007-08-10en
dc.date.rdate2007-08-22en
dc.date.sdate2007-08-17en
dc.description.abstractMilitary, Medical, Exploratory, and Commercial robots have much to gain from exchanging wheels for legs. However, the equations of motion of dynamic bipedal walker models are highly coupled and non-linear, making the selection of an appropriate control scheme difficult. A temporal difference reinforcement learning method known as Q-learning develops complex control policies through environmental exploration and exploitation. As a proof of concept, Q-learning was applied through simulation to a benchmark single pendulum swing-up/balance task; the value function was first approximated with a look-up table, and then an artificial neural network. We then applied Evolutionary Function Approximation for Reinforcement Learning to effectively control the swing-leg and torso of a 3 degree of freedom active dynamic bipedal walker in simulation. The model began each episode in a stationary vertical configuration. At each time-step the learning agent was rewarded for horizontal hip displacement scaled by torso altitude--which promoted faster walking while maintaining an upright posture--and one of six coupled torque activations were applied through two first-order filters. Over the course of 23 generations, an approximation of the value function was evolved which enabled walking at an average speed of 0.36 m/s. The agent oscillated the torso forward then backward at each step, driving the walker forward for forty-two steps in thirty seconds without falling over. This work represents the foundation for improvements in anthropomorphic bipedal robots, exoskeleton mechanisms to assist in walking, and smart prosthetics.en
dc.description.degreeMaster of Scienceen
dc.identifier.otheretd-08172007-151826en
dc.identifier.sourceurlhttp://scholar.lib.vt.edu/theses/available/etd-08172007-151826/en
dc.identifier.urihttp://hdl.handle.net/10919/34602en
dc.publisherVirginia Techen
dc.relation.haspartetd_part3.pdfen
dc.relation.haspartetd_part2.pdfen
dc.relation.haspartetd_part1.pdfen
dc.rightsIn Copyrighten
dc.rights.urihttp://rightsstatements.org/vocab/InC/1.0/en
dc.subjectDynamic Bipedal Walkingen
dc.subjectReinforcement Learningen
dc.subjectQ-Learningen
dc.subjectTorsoen
dc.subjectNEAT+Qen
dc.titleMachine Learning Simulation: Torso Dynamics of Robotic Bipeden
dc.typeThesisen
thesis.degree.disciplineMechanical Engineeringen
thesis.degree.grantorVirginia Polytechnic Institute and State Universityen
thesis.degree.levelmastersen
thesis.degree.nameMaster of Scienceen

Files

Original bundle
Now showing 1 - 3 of 3
Thumbnail Image
Name:
etd_part1.pdf
Size:
204.6 KB
Format:
Adobe Portable Document Format
Download
Thumbnail Image
Name:
etd_part2.pdf
Size:
312.95 KB
Format:
Adobe Portable Document Format
Download
Thumbnail Image
Name:
etd_part3.pdf
Size:
1.56 MB
Format:
Adobe Portable Document Format
Download

Collections

Masters Theses