WHAT: Presentations and media availability for Ashutosh Saxena, assistive robotics researcher.
WHEN: Tuesday, May 15 from 8:30 a.m. to 8 p.m. and Wednesday, May 16. Individual interviews by appointment.
WHERE: Crowne Plaza Hotel, St. Paul, Minn.
ITHACA, N.Y. – Assistive robots one day will help an elderly person take proper medications, aid a disabled person by loading dishes into a dishwashing machine, or even help pet owners clean up after their dogs and cats.
Cornell University robotics researcher Ashutosh Saxena, assistant professor of Computer Science, is developing a new generation of assistive robots that would perform a multitude of tasks. Saxena will deliver three presentations during the 2012 IEEE International Conference on Robotics and Automation (ICRA) conference in St. Paul, Minn May 14 through 18:
- Learning to Place New Objects: http://pr.cs.cornell.edu/placingobjects/
- Unstructured Human Activity Detection from RGBD Images: http://pr.cs.cornell.edu/humanactivities/
- Learning Hardware Agnostic Grasps for a Universal Jamming Gripper: http://pr.cs.cornell.edu/grasping/
Saxena works on a new generation of robots that are designed to operate autonomously in human environments. His research is focused on the development of new machine-learning algorithms that enable robots to process massive amounts of sensory input data in real-time and learn how to perform tasks in unstructured environments. His primary application domain is in assistive robotics, where his algorithms have enabled robots to perform tasks such as fetching items on verbal request, performing basic household chores, and identifying and assisting in human activities. He hopes to see such assistive robots appear in our homes, offices and nursing homes soon.
Saxena directs the Cornell Personal Robotics group: http://pr.cs.cornell.edu/ . His learning algorithms have enabled personal robots to perform tasks such as unloading items from a dishwasher, place items in a refrigerator perform household tasks: http://www.youtube.com/watch?v=P5_GtWIMQ_8&feature=youtu.be