Dr. Maria Gini gini@cs.umn.edu (612) 625-5582 http://www.cs.umn.edu/~gini

My area of research is artificial intelligence and robotics. I am mainly interested in intelligent agents, in particular self-interested software agents that achieve their tasks by engaging in negotiations with other agents, and robotic agents that operate in unstructured and only partially known environments.

Even though robots and software agents appear quite different on the surface, they require answering similar research questions such as: how to design software architectures that distribute intelligence among autonomous entities, how to obtain intelligent behavior from systems with limited computing power and limited memory, how to allow agents to make informed decisions and learn despite their limited sensing capabilities, how to accomplish complex tasks by negotiation and/or cooperation.

See Dr. Maria Gini's web page for further information on her research and publications.

Dr. Nikolaos Papanikolopoulos npapas@cs.umn.edu (612) 625-0163 http://www.cs.umn.edu/~npapas

My work focuses on robotics, computer vision, and sensors for transportation applications. My group studies vision-based robot control and eye-in-hand robotic systems. I am particularly interested in problems such as vision-based robotic grasping, the active derivation of depth maps from controlled motion of a robotic system, the active calibration of the robot-camera system, the problem of automatically detecting moving objects of interest, the use of deformable models for tracking and manipulation of rigid and non-rigid objects, and the computation of the relative position of the target from the camera. Also emphasized is experimental verification of the theory done in the Artificial Intelligence, Robotics and Vision Laboratory which has three Datacube systems and the Minnesota Robotic Visual Tracker (a flexible eye-in-hand robotic system). In computer vision, we have developed a method called "shape metamorphosis" for object recognition.

In transportation, I am interested in the use of computer vision techniques for pedestrian detection and tracking, vision-based vehicle following, bicycle counting, monitoring driver fatigue, and monitoring safety in work zones.

See Dr. Nikolaos Papanikolopoulos's web page for further information on his research and publications.

Dr. Richard Voyles voyles@cs.umn.edu (612) 624-8306 http://www.cs.umn.edu/~voyles

My research interests revolve around the field of "rapidly deployable systems." These are real-time, mechatronic systems and support services that can be quickly developed and quickly re-tasked for new applications. Major research efforts are in the area of programming by human demonstration, skill learning, reconfigurable multi-agent architectures, and tactile sensing and actuation for HCI.

Programming by human demonstration is a new paradigm for programming intended to replace the current approach of generating programs in text form. It is difficult enough to try to describe a complex task in natural language (imagine teaching a child to tie shoes over the phone), let alone in a programming language. In my view, the ultimate goal for programming is to show the robot what to do and have the robot create its own program. The emphasis would then shift from programming experts to task experts. But because each human is different, this sort of system requires the ability to interpret the human's intention to create task abstractions, rather than to just mimic what is observed. Gesture interpretation is a strong component of this aspect of my work as well as other interests in HCI.

See Dr. Richard Voyles' web page for further information on his research and publications.