Research

Vehicle Safeguarding :: Previous Research :: Publications

Autonomous Tractor
JPL Mars Yard

Previous Research

I spent the summer of 2001 working with Larry Matthies at the NASA Jet Propulsion Laboratory evaluating and improving Visual Odometry. This technique uses stereo cameras to track features across frames to find the motion of the rover. This is important for environments such as Mars, where there is no GPS, and wheel odometry may be incorrect due to wheel slip. The Visual Odometry system is currently running on the Mars rovers.

It was a great experience, and as shown to the right, I got to hang out with (test) rovers on (simulated) Mars.

Publications

C. Wellington, A. Courville, and A. Stentz, A Generative Model of Terrain for Autonomous Navigation in Vegetation, International Journal of Robotics Research (IJRR), 2006.
Abstract | PDF Preprint | DOI | Bibtex

C. Wellington, Learning a Terrain Model for Autonomous Navigation in Rough Terrain, Doctoral Dissertation, Technical Report CMU-RI-TR-05-59, Robotics Institute, Carnegie Mellon University, 2005.
Abstract | PDF | PDF (book margins) | Bibtex

C. Wellington, A. Courville, and A. Stentz, Interacting Markov Random Fields for Simultaneous Terrain Modeling and Obstacle Detection, Robotics Science and Systems Conference, 2005.
Abstract | PDF | Bibtex

C. Wellington and A. Stentz, Online Adaptive Rough-Terrain Navigation in Vegetation, IEEE International Conference on Robotics and Automation (ICRA), 2004.
Abstract | PDF | Bibtex

C. Wellington and A. Stentz, Learning Predictions of the Load-Bearing Surface for Autonomous Rough-Terrain Navigation in Vegetation, Field and Service Robotics Conference (FSR), 2003.
Abstract | PDF | Bibtex

A. Stentz, C. Dima, C. Wellington, H. Herman, and D. Stager, A System for Semi-Autonomous Tractor Operations, Autonomous Robots, 2002.
Abstract | Bibtex

C. Wellington, R. Bock, and B. Maxwell, Combining Color and Edge Information to Find Door Locations in an Image, SPIE Intelligent Robots and Computer Vision, 1999.
Abstract | Bibtex