NIPS 2006 Workshop

On

Learning Applied to Ground Robots: Sensing and Locomotion

Organizers:

Greg Grudic, University of Colorado, Boulder, CO, USA

Larry Jackel, Program Manager, DARPA

Jane Mulligan, University of Colorado, Boulder, CO, USA

Link to Updated Schedule

Workshop Description:

Autonomous robot navigation in unstructured outdoor environments remains a critical challenge for tasks such as reconnaissance, search and rescue and automated driving. Completion of the DARPA Grand Challenge (www.darpa.mil/grandchallenge ), was an exciting step toward this goal, but competitors still required extensive use of well chosen GPS way points, sometimes only a few meters apart. Indeed, successful navigation (between way points a few hundred meters apart) in unfamiliar dynamic outdoor environments is a key research goal of the DARPA Learning Applied to Ground Robots (LAGR) program. In addition, legged robots hold promise for traversal of large, irregular obstacles by unmanned vehicles, but hard coded motion algorithms typically cannot account for variable and unexpected terrain. Addressing these problems is the goal of the DARPA Learning Locomotion program.

Machine Learning algorithms are a principled approach for dealing with uncertainty in sensing, computation and actuation. As such, machine learning theory offers a unique opportunity for addressing the open research problems in robot locomotion and outdoor navigation.

The goal of this workshop is to bring together Robotics and Machine learning researchers to discuss how they are currently applying Machine Learning techniques to open Robotics problems in navigation and locomotion. Furthermore, we hope to look to the future to see how Machine Learning and Robotics can leveraged to the benefit of both communities.