The Robotics Institute
 Search the site
RI | People | Sebastian Thrun

Text only version of this site

Sebastian Thrun
Adjunct Associate Professor, MLD/CSD/RI (Adjunct)

No longer a member of RI.

Email address: thrun@stanford.edu

For more information, see my personal homepage.

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

Research interests

I seek effective means for engineering software for complex embedded systems, equipped with sensors and actuators. My group and I have developed statistical algorithms for perception, learning and decision making in robotics (see introductory report). We have applied these algorithms to various problems in mobile robotics, such as exploration, map learning, human robot interaction, and multi-robot team coordination (see 3D maps). Two of our mobile robots were deployed as interactive museum tour-guides. At present, I am involved in a multi-university initiative to develop robots that assist elderly people.

I am equally interested in machine learning. I believe learning and teaching should be seamlessly integrated into mainstream software development. My current research focuses on programming languages that support learning from experience and probabilistic computation (see preliminary report). In the past, I also worked on lifelong learning algorithms, which enable agents to transfer knowledge among families of related learning tasks (see book and thesis).

For complete current information on my research, please see my SCS homepage.

Research interest keywords

artificial intelligence, computer vision, gesture recognition, human-computer interaction, machine learning, mobile robots, multi-agent systems, sensor fusion, and statistics

Past Labs & Groups

Robot Learning Lab - Pursuing research on robot learning, mobile robotics, multi-robotics, human robot interaction, and basic machine learning.
 

Past Projects

3D Mine Mapping - We are developing robotic systems for automated 3D mine mapping.
Amelia - An improved Xavier
BORG - Framework for the development of distributed, secure, reliable, robust and scalable systems
Mercator - Cooperative Multi-Robot Mapping and Planning
Minerva - Minerva is an autonomous robot that daily moves through crowds at the Smithsonian's National Museum of American History.
Personal Robotic Assistants For The Elderly - Pearl will help researchers explore the ways autonomous mobile robots can help the elderly live independently in their own homes.
Statistical Methods For Learning Maps with Mobile Robots
Super-Resolved Texture Tracking - Video-stream object motion estimation and tracking.
The Lifelong Learning Project
Xavier - Perceptual, reasoning and learning abilities in autonomous mobile robots

Recent publications [View all 135 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