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. Successful navigation (between way points a few hundred meters apart) in unfamiliar dynamic outdoor environments remains a key open research problem. Such Robotic tasks are characterized by a high dimensional input space that represents the world mediated by robot sensors (vision, sonar data, etc). The robot experiences millions of sensor readings at many frames per second, which must be processed and acted upon in real time. The key open questions are: 1) What information must be extracted from sensors? and, 2) How can the robot use this information to act appropriately in the world? Machine Learning techniques offer powerful tools to model complex real world situations and produce coherent behavior. Indeed many of the fundamental goals of Machine Learning are also those of Robotics, which establishes a synergy between the two fields that can serve as a catalyst for advancing theory and practice in both.

This talk will describe the Machine Learning algorithms we are currently applying to autonomous navigation in unstructured outdoor environments. These represent work done under the DARPA Learning Applied to Ground Robots (LAGR) program, and NSF funded Human-to-Robot Skill transfer research. As an autonomous robot passes through varied environments, it should learn through experience and interaction, it should keep and reuse models it learns over time, and it should know when it doesn't know how to act and new models are required. Our near-to-far Learning approach constructs models of traversable terrain based on near field Stereo and uses them to classify far field regions of the image for long range path planning. Models are built in real-time as the robot navigates, and are intended to be used over the robot's lifetime. This Long Term Ongoing Learning, where an agent learns and maintains knowledge of the environment to improve performance over time, is a key goal of Learning and Robotics. It poses many interesting questions regarding how to maintain large model sets, and when and how to apply, refine or discard models. Experimental evidence from actual robot trials shows that this approach significantly outperforms purely range based navigation.

BIO
Greg Grudic received his Ph.D. in Electrical and Computer Engineering at the University of British Columbia in 1997. He was a Post Doctoral Fellow at the GRASP Lab at the University of Pennsylvania between 1998 and 2001. In 2001 he joined the Department of Computer Science at the University of Colorado, where he is currently an Assistant Professor. This work is partially funded by DOD AFRL FA8650-07-C-7702, NSF IIS 0535269, and NSF CNS 043059.