Summary: | Thesis: Ph. D., Massachusetts Institute of Technology, Department of Mechanical Engineering, 2018. === Cataloged from PDF version of thesis. === Includes bibliographical references (pages 169-175). === Despite remarkable recent advances in robotic research, legged machines are still far from robustly executing physical actions with comparable performance to humans. Yet, the potential applications for robots with such unique capabilities range from disaster response all the way to elderly care and further. Hence, an intuitive short-term answer for this issue lies on harnessing human motor control abilities and transferring them to the remote robot via whole-body teleoperation while providing the operator with real-time physical feedback from his/her actions. Motivated by such a promising solution, this Thesis presents an introductory study to achieve human and bipedal robot dynamic synchronization via whole-body teleoperation and bilateral feedback. This work describes how we can utilize powerful simple models to explore the interplay between human Center of Mass motion and the contact forces with the environment in order to transmit to the robot the underlying balancing and stepping strategy. All the necessary fundamental equations for the coupled dynamics in the Frontal Plane are presented along with the human feedback law and motion data mapping derived from the imposition of dynamic similarity. We take a closer look on how the natural frequency of each system influences the resulting motion and analyze how the coupled system responds to various robot scales. We present experiments in which a human operator controls a bipedal robot to show how the feedback from the Human-Machine Interface varies according to the robot's characteristic time response and the perturbations from its surrounding environment. Finally, we describe the implementation of the presented strategy on a small-scale dynamic robot, Little HERMES, to allow it to balance, jump and take steps in place simultaneously with the human operator. We expect that the results presented in this Thesis will eventually allow robots to achieve motor dexterity and coordination that can rival their biological counterparts. === by João Luiz Almeida de Souza Ramos. === Ph. D.
|