Dr. Michael Kaess
Research Scientist [Curriculum Vitae]Computer Science and Artificial Intelligence Lab (CSAIL)
Massachusetts Institute of Technology (MIT)
32 Vassar St., Room 230
Cambridge, MA 02139
Phone: +1 617 452 2852
email: kaess@mit.edu
My research focuses on probabilistic methods in mobile robotics and computer vision. I am working on inference with large sparse matrices and graphical models, in particular exploring their connections. Solutions to this problem are of interest for robot localization and large-scale mapping. One particular application that I have been working on is the navigation for closed loop control of an autonomous underwater vehicle to inspect the hulls of large ships.
I am a member of the Marine Robotics Lab of John Leonard. I finished my PhD with Frank Dellaert at Georgia Tech in 2008. My dissertation introduced an efficient algorithm called iSAM for simultaneous localization and mapping (SLAM), which is the problem of mapping a previously unknown environment while at the same time using this map for localization.
Mapping has many important applications, including inspection of underwater structures, indoor navigation for service robots and search-and-rescue scenarios, navigation for autonomous cars and space applications. Here are some recent highlights from my projects:
Kintinuous: Live dense 3D modeling
Building-scale mapping and long-term mapping
Ship hull inspection: Closed loop control of underwater robot
Highlights [All News...]
- Feb 2013: I appeared on Discovery Channel Canada: Here is the 3 minute Daily Planet segment (starts at 22s after the commercial) where I show the host our latest work: Kintinuous.
- Jan 2013: Our three ICRA submissions all got accepted: Tom's robust Kintinuous, Hordur's long-term mapping and David's non-Gaussian optimization.
- Dec 2012: I will be co-teaching 2.017 (Design of electromechanical robotic systems) this spring.
- Oct 2012: I am Associate Editor for ICRA 2013 and on the program committee for RSS 2013.
- Sep 2012: Two journal papers got accepted: Multi-session visual SLAM (RAS) and In-water ship hull inspection (IJRR).
- Aug 2012: I was a runner up for the Dick Volz Best 2008 U.S. PhD thesis award in robotics and automation.
- May 2012: Check out Kintinuous: Our submission to the RGB-D workshop at RSS 2012.
- May 2012: Hordur Johannsson managed to map a large part of the Stata Center (10 floors!) using his reduced pose graph formulation, see our technical report.
- Apr 2012: We have two papers in FUSION 2012: concurrent smoothing and filtering and inertial navigation with factor graphs.
- Mar 2012: A new, much improved version (1.6) of my iSAM library has been released, featuring Powell's dog leg, stereo support, Quaternions, Eigen...
- Feb 2012: I am Associate Editor for IROS 2012, and on the programm committee of AAAI 2012 and RSS 2012.
- Jan 2012: Check out our paper on robust incremental least-squares in ICRA 2012 using an incrementalized version of Powell's dog leg.
- Oct 2011: Our iSAM2/Bayestree journal paper will appear in IJRR in February 2012.
- Sep 2011: John McDonald presented our work on "6-DOF Multi-session Visual SLAM using Anchor Nodes" at ECMR 2011.
- Jul 2011: I organized the Automated SLAM Evaluation Workshop at RSS 2011 together with Giorgio Grisetti and Kai Ni. Many thanks to all the speakers for their very interesting talks and to the audience for their active participation!
- Jan 2011: Check out our new iSAM2 paper that will appear in ICRA 2011 in Shanghai. I also co-authored a paper with Maurice Fallon that uses iSAM to fuse sonar data for AUV navigation (nominated for best automation paper).
- Dec 2010: I presented our Bayes tree work at WAFR 2010, and also gave a related talk at the National University of Singapore.
- Aug 2010: Check out my open source code for iSAM, now available here under the LGPL license. Includes full documentation, a few datasets and a 3D viewer.
Selected Publications [All Publications...]
“An Incremental Trust-Region Method for Robust Online Sparse Least-Squares Estimation” by D.M. Rosen, M. Kaess, and J.J. Leonard. In IEEE Intl. Conf. on Robotics and Automation, ICRA, (St. Paul, MN), May 2012. Details. Download: PDF.“iSAM2: Incremental Smoothing and Mapping Using the Bayes Tree” by M. Kaess, H. Johannsson, R. Roberts, V. Ila, J.J. Leonard, and F. Dellaert. Intl. J. of Robotics Research, IJRR, vol. 31, Feb. 2012, pp. 217-236. Details. Download: PDF.
“The Bayes Tree: An Algorithmic Foundation for Probabilistic Robot Mapping” by M. Kaess, V. Ila, R. Roberts, and F. Dellaert. In Intl. Workshop on the Algorithmic Foundations of Robotics, WAFR, (Singapore), Dec. 2010. Details. Download: PDF.
“Multiple Relative Pose Graphs for Robust Cooperative Mapping” by B. Kim, M. Kaess, L. Fletcher, J.J. Leonard, A. Bachrach, N. Roy, and S. Teller. In IEEE Intl. Conf. on Robotics and Automation, ICRA, (Anchorage, Alaska), May 2010, pp. 3185-3192. Details. Download: PDF.
“Covariance Recovery from a Square Root Information Matrix for Data Association” by M. Kaess and F. Dellaert. Journal of Robotics and Autonomous Systems, vol. 57, Dec. 2009, pp. 1198-1210. Details. Download: PDF.
“iSAM: Incremental Smoothing and Mapping” by M. Kaess, A. Ranganathan, and F. Dellaert. IEEE Trans. on Robotics, vol. 24, no. 6, Dec. 2008, pp. 1365-1378. Details. Download: PDF.
“Square Root SAM: Simultaneous Localization and Mapping via Square Root Information Smoothing” by F. Dellaert and M. Kaess. Intl. J. of Robotics Research, vol. 25, no. 12, Dec. 2006, pp. 1181-1204. Details. Download: PDF.
“MCMC-based Multiview Reconstruction of Piecewise Smooth Subdivision Curves with a Variable Number of Control Points” by M. Kaess, R. Zboinski, and F. Dellaert. In Eur. Conf. on Computer Vision, ECCV, (Prague, Czech Republic), May 2004, pp. 329-341. Acceptance ratio 34.2% (190 of 555). Details. Download: PDF.