2D3D Interactive Feedback Control for Autonomous Systems
Abstract
Over the past twenty years, we have witnessed a dramatic increase in the availability of time-varying imageinformation acquired through varying systems. In turn, the need to exploit such dynamical systems hasresulted in numerous advances in applying control theoretics to problems including, but not limited to,target tracking, shape reconstruction, and machine learning. However, there remains a “gap” to developfully autonomous systems and at best, such control active vision frameworks require user-experts to combatthe risk complexities of real-world scenarios. That is, there still exists no “universal” framework capableof handling real-world scenarios and more often than not, simply “throwing” feedback control at suchproblems in order to “plug” this information gap often results in restricted solutions driven by particularapplication or system.As such, this research program aims to devise interactive control theoretic schemes directly on the 3Dphysical system using 2D image information as opposed to exploiting artificial image dynamics and thenlifting such dynamics in a heuristic decoupled manner. In particular, the thematic vision focuses on thetheoretical understanding of how geometry can affect control and how this interplay will enable one to“close the loop” of common imaging tasks by incorporating 3D human user (computer) interaction as formof feedback control.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Apr 09, 2018
- Source ID
- FA95501810130
Entities
People
- Romeil Sandhu
Organizations
- Air Force Office of Scientific Research
- Research Foundation for the State University of New York
- United States Air Force