Exploration of a data-driven walking controller for a large-scale bipedal robot with structural deformation
Darwish, Omar
This item is only available for download by members of the University of Illinois community. Students, faculty, and staff at the U of I may log in with your NetID and password to view the item. If you are trying to access an Illinois-restricted dissertation or thesis, you can request a copy through your library's Inter-Library Loan office or purchase a copy directly from ProQuest.
Permalink
https://hdl.handle.net/2142/120463
Description
Title
Exploration of a data-driven walking controller for a large-scale bipedal robot with structural deformation
Author(s)
Darwish, Omar
Issue Date
2023-05-04
Director of Research (if dissertation) or Advisor (if thesis)
Kim, Joohyung
Department of Study
Mechanical Sci & Engineering
Discipline
Mechanical Engineering
Degree Granting Institution
University of Illinois at Urbana-Champaign
Degree Name
M.S.
Degree Level
Thesis
Keyword(s)
Robotics, data-driven controller, bipedal robots
Abstract
Biped robots hold a great potential in advancing multiple industries, from healthcare to disaster relief to entertainment and personal assistance. These robots have the ability to navigate complex environments, interact more efficiently with people, and perform tasks that require balance and dexterity. However, accomplishing those tasks depends on heavily on being able to represent the kinematics of the robot, which can be heavily altered due to structural deformation and motors backlash. This thesis develops a data- driven approach to address the kinematics issues for a large-scale biped robot using Hybird-Leg as well as develop a simple controller for its locomotion.
In order to evaluate the proposed approach, two sets of tests are proposed. First, a test is performed to compare between the performance produced when only using the existing kinematics model versus when using the kine- matics model along with an Artificial Neural Network (ANN) model, which showed that the developed approach improved the tracking. The next set of tests develops and validates a simple walking controller using simulation. Then, the generated walking trajectory is inputted into the ANN model in order to be implemented on the hardware. The outputted trajectory by the ANN model was not feasible or safe for hardware implementation, therefore hardware implementation of the walking trajectory was not completed.
The contribution of this work is highlighted in providing a skeleton for developing a data-driven approach in order to limit the effects of structural deformation in the development of successfully walking bipeds. This work provides a detailed approach to a possible method in order to limit the effects of inaccurate kinematic modeling.
Use this login method if you
don't
have an
@illinois.edu
email address.
(Oops, I do have one)
IDEALS migrated to a new platform on June 23, 2022. If you created
your account prior to this date, you will have to reset your password
using the forgot-password link below.
