Seminar: Patricio A. Vela

“Perceptual Representations for Autonomous Mobile Robot Navigation of Unknown Worlds”
Thursday, Oct. 14 at 1:00pm
Via Zoom
Email for Zoom info

sponsored by the ECE IDEA Committee
Celebrating National Hispanic Heritage Month
Sept. 15–Oct. 15
Thanks to Texas Instruments for Their Support


The general topic is autonomous navigation of unknown or uncertain environments and the role visual information plays in permitting fast, safe, and accurate trajectory generation for goal attainment. Navigating an environment under partial map information requires establishing a free-space map at the same time as safely maneuvering through the environment to arrive at the goal state. In this case, many of the theoretical guarantees of existing path planning methods fail to apply. Furthermore, real-time constraints prevent the use of a single technique. Instead, hierarchical navigation strategies prevail with a slow global approach for hypothesizing the best possible path given all information, and a fast local approach for identifying the best local goal option aligned with the global information. This talk will describe how Marr’s visual hierarchy together with Gibson’s notion of affordances provides a framework for deriving a hierarchical navigation scheme with provable properties augmented by machine learning. It relies on using world representations and data structures more closely tied to perception space, i.e., being more image-like. The scheme also has favorable properties with regards to computational costs, safety, robustness, and task completion. How these are achieved by looking at systems level issues will be discussed. The results to date support a long term goal of facilitating navigation through unknown, GPS-denied environments.


Patricio A. Vela is an associate professor in the School of Electrical and Computer Engineering at the Georgia Institute of Technology. Dr. Vela’s research focuses on geometric perspectives to control theory and computer vision, particularly how concepts from control and dynamical systems theory can serve to improve computer vision algorithms used in the decision-loop. These efforts are part of a broad program to understand research challenges associated with autonomous robotic operation in uncertain environments. Dr. Vela received a B.S. (1998) and a Ph.D. (2003) from the California Institute of Technology. He was a post-doctoral researcher at the Georgia Institute of Technology from 2003 to 2005.