Graphics enhanced version of this site

Daniel Nikovski
PhD Student

No longer a member of RI.

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

My principal research interest is in developing algorithms for learning probabilistic models of robotic domains and using these models for reasoning and planning. Decision-theoretic planning and control deal naturally with the uncertainty and unobservability present in robotic tasks such as mobile robot navigation, but providing the appropriate probabilitic models of the problem domain requires significant human effort. My research aims to produce algorithms for the autonomous acquisition of such models from training data.

machine learning, mobile robots, and planning

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

• Amelia - An improved Xavier
• Mercator - Cooperative Multi-Robot Mapping and Planning
• Xavier - Perceptual, reasoning and learning abilities in autonomous mobile robots

• State-Aggregation Algorithms for Learning Probabilistic Models for Robot Control
  D. Nikovski
  doctoral dissertation, tech. report CMU-RI-TR-02-04, Robotics Institute, Carnegie Mellon University, February, 2002. [Abstract]
  Download: pdf [984 KB], ps.gz [702 KB] copyrighted

• Learning State Grounding for Optimal Visual Servocontrol of Dynamic Manipulation
  D. Nikovski and I. Nourbakhsh
  Proceedings of the AAAI Spring Symposium on State Grounding and Abstraction, March, 2001.
  Download: pdf [118 KB], ps.gz [97 KB] copyrighted

• Constructing Bayesian networks for medical diagnosis from incomplete and partially correct statistics
  D. Nikovski
  EEE Transactions on Knowledge and Data Engineering, Vol. 12, No. 4, July, 2000, pp. 509 - 516. [Abstract]
  Download: pdf [166 KB], ps.gz [57 KB], ps.z [77 KB] copyrighted

• Learning Probabilistic Models for Decision-Theoretic Navigation of Mobile Robots
  D. Nikovski and I. Nourbakhsh
  Proceedings of the International Conference on Machine Learning 2000, 2000.
  Download: pdf [154 KB], ps.gz [52 KB] copyrighted

• Learning discrete Bayesian models for autonomous agent navigation
  D. Nikovski and I. Nourbakhsh
  Proceedings of IEEE Conference on Computational Intelligence in Robotics and Automation (CIRA '99), November, 1999, pp. 137 - 143. [Abstract]
  Download: pdf [155 KB], ps.gz [49 KB] copyrighted

• Learning stationary temporal probabilistic networks
  D. Nikovski
  Conference on Automated Learning and Discovery, 1998.
  Download: pdf [117 KB], ps.gz [41 KB] copyrighted

• Artificial Intelligence
  D.P. Dimitrov and D. Nikovski
  Technical University Sofia Press, Sofia, 1997.

• Visual memory-based learning for mobile robot navigation
  D. Nikovski
  Proceedings of the Second International Conference on Computational Intelligence and Neurosciences, Vol. 2, 1997, pp. 1-4.
  Download: pdf [109 KB], ps.gz [32 KB] copyrighted

• Amelia
  R. Simmons, S. Thrun, G. Armstrong, R. Goodwin, K. Haigh, S. Koenig, S. Mahamud, D. Nikovski, and J. O'Sullivan
  Proceedings of the AAAI Thirteenth National Conference on Artificial Intelligence, 1996.

• Comparison of two learning networks for time series prediction
  D. Nikovski and M. Zargham
  Proceedings of the Ninth International Conference IEA/AIE 96 (Industrial and Engineering Applications of Artificial Intelligence and Expert Systems) June 4-7, 1996, pp. 531-535.
  Download: pdf [122 KB], ps.gz [35 KB], ps.z [49 KB] copyrighted