|Title||:||Towards An Architecture for Human-Robot Collaboration|
|Speaker||:||Mohan Sridharan (The University of Auckland, NZ)|
|Details||:||Sun, 21 Dec, 2014 2:00 PM @ BSB 361|
|Abstract:||:||Robots deployed to collaborate with humans in homes, offices, and other domains, have to represent knowledge and reason at both the sensorimotor level and the cognitive level. One fundamental challenge
posed by this objective is that of representing, learning from, and reasoning with qualitative and quantitative descriptions of uncertainty and incomplete domain knowledge obtained from different sources. This talk will describe an architecture that exploits the interplay between learning, representation and reasoning to address this challenge. I shall talk about the integration of the commonsense reasoning capabilities of declarative programming with the uncertainty modeling capabilities of probabilistic graphical models. I shall also briefly describe probabilistic algorithms that use the representation and reasoning capabilities to guide the learning of object models based on visual context and appearance cues. If time permits, I shall also illustrate the use of these algorithms for estimation tasks in non-robotics application domains.
Speaker bio: Mohan Sridharan is a Senior Lecturer in the Department of Electrical and Computer Engineering at The University of Auckland. Prior to his current appointment, he was a faculty member at Texas Tech University (USA), where he is currently an Adjunct Associate Professor in the Department of Mathematics and Statistics. Dr. Sridharan received his Ph.D. in Electrical and Computer Engineering from The University of Texas at Austin (USA) in 2007, and spent one year as a Research Fellow in the School of Computer Science at The University of Birmingham (UK). His current research interests include machine learning, knowledge representation and reasoning, computational vision, and cognitive science, as applied to autonomous robots and intelligent agents. Furthermore, he is interested in designing stochastic algorithms for estimation and inference tasks in domain characterized by a significant amount of uncertainty.