Building Robust Systems out of Non-Robust Components

Pradeep K. Khosla

College of Engineering &

School of Computer Science

Carnegie Mellon University





Most of current machine vision systems suffer from a lack of flexibility to account for the high variability of unstructured environments. As the state of the world evolves the knowledge provided by different visual attributes changes, breaking the initial assumptions of the vision system. This results in the vision algorithm/system not being able to accomplish its original goals.  We have developed a new approach for the creation of an adaptive visual system that demonstrates robust behavior even when its components are non-robust.   The system is able to selectively combine information from different visual algorithms depending on its evaluation of which algorithms are making a contribution to its goal.  Using a probabilistic approach and uncertainty metrics, the system is able to take appropriate decisions about the more relevant visual attributes to consider. The system is based on an intelligent agent paradigm. Each visual algorithm is implemented as an agent, which adapts its behavior according to uncertainty considerations. The proposed system aims to achieve robustness and efficiency. By combining the outputs of multiple vision modules the assumptions and constraints of each module are factored out resulting in a more robust system. Efficiency is achieved through the on-line selection and specialization of the agents. An implementation of the system for the task of human tracking has demonstrated excellent results.


This is joint work with Alvaro Soto.



Brief Biography:


Pradeep K. Khosla is the Philip and Marsha Dowd Professor in the College of Engineering and School of Computer Science at Carnegie Mellon.  He is also the Head of the Department of Electrical and Computer Engineering and The Information Networking Institute.  His research interests are in the areas of Distributed Information Systems, Distributed Robotic Systems, and Reconfigurable Hardware and Real-time Software.  From 1994 to 1996 he served as a DARPA Program Manager where he managed several programs in Robotics, AIT, and Real-time Software Systems.  His contributions to research and education have been recognized by many awards including the 1999 George Westinghouse Award for Education and the 2001 Wallace McDowell Award from the IEEE Computer Society.