event
PhD Defense by Daniel Pickem
Primary tabs
Ph.D. Thesis Proposal Announcement
Title: Self-Reconfigurable Multi-Robot Systems
Daniel PickemRobotics Ph.D. StudentSchool of Electrical and Computer EngineeringCollege of EngineeringGeorgia Institute of Technologyhttp://www.danielpickem.com
Date: July 29, 2015 (Wednesday)Time: 10am - 12pm ESTLocation: TSRB 530
Committee:Dr. Justin Romberg, School of Electrical and Computer Engineering, Georgia TechDr. Martha A. Grover, School of Chemical and Biomolecular Engineering, Georgia TechDr. Jun Ueda, School of Mechanical Engineering, Georgia TechDr. Jeff S. Shamma, School of Electrical and Computer Engineering, Georgia TechDr. Magnus Egerstedt, School of Electrical and Computer Engineering, Georgia Tech
Abstract:Self-reconfigurable robot systems are comprised of individual modules which are able to connect to and disconnect from one another to form larger structures. These systems therefore have the ability to change their morphology and functionality and thus adapt to changing tasks and environments. Optimally reconfiguring such systems is NP-hard, which makes centralized solutions unsuitable for solving the self-reconfiguration problem for large systems. As a result, self-reconfiguration approaches aim at approximating optimal solutions. Nonetheless, even for approximate solutions, centralized methods scale poorly in the number of modules. Therefore, the objective of this proposal is the development of decentralized methods for the self-reconfiguration of modular robotic systems. Building on completeness results of the centralized algorithms in this proposal, decentralized solution methods are developed that guarantee convergence to global optima, i.e. reconfigure initial into target configurations. A game theoretic approach lays the theoretical foundation of a novel potential game-based formulation of the self-reconfiguration problem. Stochastic convergence guarantees are provided for a large class of utility functions used by purely self-interested agents. This flexibility in the choice of utility functions makes the presented approach and the underlying theory suitable for a wide array of problems that rely on decentralized local control to guarantee global, emerging properties.
Title: Self-Reconfigurable Multi-Robot Systems
Daniel PickemRobotics Ph.D. StudentSchool of Electrical and Computer EngineeringCollege of EngineeringGeorgia Institute of Technologyhttp://www.danielpickem.com
Date: July 29, 2015 (Wednesday)Time: 10am - 12pm ESTLocation: TSRB 530
Committee:Dr. Justin Romberg, School of Electrical and Computer Engineering, Georgia TechDr. Martha A. Grover, School of Chemical and Biomolecular Engineering, Georgia TechDr. Jun Ueda, School of Mechanical Engineering, Georgia TechDr. Jeff S. Shamma, School of Electrical and Computer Engineering, Georgia TechDr. Magnus Egerstedt, School of Electrical and Computer Engineering, Georgia Tech
Abstract:Self-reconfigurable robot systems are comprised of individual modules which are able to connect to and disconnect from one another to form larger structures. These systems therefore have the ability to change their morphology and functionality and thus adapt to changing tasks and environments. Optimally reconfiguring such systems is NP-hard, which makes centralized solutions unsuitable for solving the self-reconfiguration problem for large systems. As a result, self-reconfiguration approaches aim at approximating optimal solutions. Nonetheless, even for approximate solutions, centralized methods scale poorly in the number of modules. Therefore, the objective of this proposal is the development of decentralized methods for the self-reconfiguration of modular robotic systems. Building on completeness results of the centralized algorithms in this proposal, decentralized solution methods are developed that guarantee convergence to global optima, i.e. reconfigure initial into target configurations. A game theoretic approach lays the theoretical foundation of a novel potential game-based formulation of the self-reconfiguration problem. Stochastic convergence guarantees are provided for a large class of utility functions used by purely self-interested agents. This flexibility in the choice of utility functions makes the presented approach and the underlying theory suitable for a wide array of problems that rely on decentralized local control to guarantee global, emerging properties.
Groups
Status
- Workflow Status:Published
- Created By:Tatianna Richardson
- Created:07/13/2015
- Modified By:Fletcher Moore
- Modified:10/07/2016
Categories
Keywords
Target Audience