Search

Navigator: RI | People | Parag Batavia

Graphics enhanced version of this site

Parag Batavia
NREC Commercialization Specialist

No longer a member of RI.

Jump to: Research interests | Keywords | Labs & Groups | Projects | Publications

Research interests

I am primarily interested in the applications of robotics technology to areas which can have a high impact on everyday life. My work in field robotics, personal robotics, and driver safety is aimed at this goal.

Field Robotics: My current work in this area involves automated turf management. Managing large areas of "green space," such as golf courses, sports fields, and city parks is a costly, labor intensive task. For instance, mowing a golf course requires great precision, to create the cross hatch patterns which are required. Automating this task requires addressing three main issues: High precision navigation, robust obstacle detection, and user interface development. I am interested in the development of all of these areas, combining them to make a useful, robust, system for turf management.

Personal Robots: It is now possible for the average person to own a robot. Although currently aimed primarily at hobbyists (Mindstorms, Aibo, Cye), their ability to do useful work, such as aid people with disabilities, is increasing. My work in this area has been in path planning, specialized to work in areas with dynamic clutter (such as houses). These planners are computationally efficient, and take into account environmental uncertainty. At the same time, they generate paths which are "human-like," by centering on corridors, taking appropriate shortcuts, and avoiding obstacles.

Driver Assistance: Each year, about 40,000 people are killed in highway accidents. A large number of these fatalities are due to Single Vehicle Roadway Departure accidents. My interest in this area is in developing models of driver behavior which can be used to modulate a lane departure warning system. I.e., we learn if a particular driver normally maintains "tight" lane control. If this is the case, then we can warn when the driver is behaving abnormally, usually much earlier than we can for someone who is normally a "loose" driver.

Research interest keywords

computer vision, field robotics, mobile robots, motion planning, and obstacle avoidance

Past Labs & Groups

Past Projects

Recent publications [View all 11 publications]

The Robotics Institute is part of the School of Computer Science, Carnegie Mellon University.
For updates and comments, please see these instructions.
This page maintained by robotwebmaster@ri.cmu.edu