Robust Systems Laboratory Overview

A UNIFYING THEME, ROBUSTNESS AND UNCERTAINTY MANAGEMENT: Recent technological advances hold the promise of significantly changing the way we live and interact with our environment. Smart environments will enable elderly people to carry on independent lives and can enhance both safety and self--discovery at kindergartens and schools. Computers that interpret facial expressions and human gestures can lead to simpler interfaces. Autonomous systems can free humans from repetitive or dangerous tasks. Finally, intelligent activity surveillance systems can substantially improve our ability to prevent tragedies. Indeed, computer vision and control are already linked through many successful proof--of--concept systems developed at several research institutions, including Penn State. However, while highly optimized for the tasks that they have been designed for, these systems remain fragile to uncertainty. The goal of the ROBUST SYSTEMS LABORATORY is to develop both theoretical tools and specific algorithms leading to robust systems, capable of achieving near optimal performance under a wide range of conditions.

		ROBUSTNESS against disturbances and model uncertainty is at the heart of control practice. Indeed, in the (completely unrealistic) case where both all external disturbances and a model of the system to be controlled are exactly known, there is no need for feedback: optimal performance can be achieved with an open loop controller. Interest in robust control arose in the late 70's where it was shown that many popular control methods led to fragile closed loop systems, and the field has been very active since. Indeed, very recent research has shown that the concept of robustness through feedback is not limited just to control, appearing in fields as dissimilar as physics, network management and biology. At the Robust Systems Lab we are developing both theory and tractable algorithms to address various aspects of the problem ranging from the transformation of experimental signals from the physical plant to a set of models (robust identification), to the synthesis of a controller for that set of models (robust control).
"Robot Visions" by Ralph McQuarrie.		COMPUTER VISION SYSTEMS bring together imaging devices, computers, and sophisticated algorithms to solve problems in areas such as industrial inspection, autonomous navigation, human-computer interfaces, medicine, image retrieval from databases, realistic computer graphics rendering, document analysis, and remote sensing. The goal of computer vision is to make useful decisions about real physical objects and scenes based on sensed images. Achieving this goal requires obtaining and using descriptions (models) of the sensors and the world. At the Robust Systems Laboratory we study how to build these models and how to use them while being robust against disturbances such as noise, clutter, and model uncertainty. Computer vision is an exciting but disorganized field that builds on very diverse disciplines such as image processing, statistics, pattern recognition, control theory and system identification, physics, geometry, computer graphics, and learning theory.