Vassili Korotkine

Macdonald Engineering Building

McGill University

817 Sherbrooke St W, Montreal

My research lies under the umbrella of Simultaneous Localization and Mapping (SLAM), where the robot seeks to answer the following questions: 1) Where am I? and 2) What does the environment look like?

I am particularly interested in situations where there is inherent ambiguity in the environment, such as the case of unknown data associations, where the robot does not know which landmark a measurement came from.

Outside of work, I am interested in anything outdoors. From 2020-2022 I was on the exec council for McGill Triathlon Club organizing their social events.

news

Sep 15, 2024	Our paper on improving the use of gaussian mixtures in common robotics optimizers was published in Robotics & Automation Letters!
Jun 11, 2024	Was very happy to present at the 2024 edition of the Montreal Robotics Summer School at the Mila - Quebec AI Institute on the topic of Monte Carlo Filtering! Slides.

selected publications

RA-L

A Hessian for Gaussian Mixture Likelihoods in Nonlinear Least Squares

Vassili Korotkine, Mitchell Cohen, and James Richard Forbes

IEEE Robotics and Automation Letters, 2024

DOI Bib PDF Code

@article{hessianKorotkine2024,
  author = {Korotkine, Vassili and Cohen, Mitchell and Forbes, James Richard},
  journal = {IEEE Robotics and Automation Letters},
  title = {A Hessian for Gaussian Mixture Likelihoods in Nonlinear Least Squares},
  year = {2024},
  volume = {9},
  number = {9},
  pages = {7891-7898},
  keywords = {Optimization;Simultaneous localization and mapping;Optimization methods;State estimation;Standards;Newton method;Jacobian matrices;Localization;optimization and optimal control;probabilistic inference;sensor fusion;SLAM},
  doi = {10.1109/LRA.2024.3432350},
  selected = true,
  code = https://github.com/decargroup/hessian_sum_mixtures,
  bibtex_show = true
}

IROS

navlie: A Python Package for State Estimation on Lie Groups

Charles Champagne Cossette, Mitchell Cohen, Vassili Korotkine, and 3 more authors

In 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2023

DOI Bib PDF Code

@inproceedings{10342362,
  author = {Cossette, Charles Champagne and Cohen, Mitchell and Korotkine, Vassili and Del Castillo Bernal, Arturo and Shalaby, Mohammed Ayman and Forbes, James Richard},
  booktitle = {2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)},
  title = {navlie: A Python Package for State Estimation on Lie Groups},
  year = {2023},
  volume = {},
  number = {},
  pages = {5282-5287},
  keywords = {Manifolds;Estimation error;Navigation;Prototypes;Libraries;State estimation;Task analysis;Localization;Sensor Fusion;Software Tools for Robot Programming},
  doi = {10.1109/IROS55552.2023.10342362},
  selected = true,
  code = https://github.com/decargroup/navlie/tree/main/navlie/utils,
  bibtex_show = true
}

RA-L

Koopman Linearization for Data-Driven Batch State Estimation of Control-Affine Systems

Zi Cong Guo, Vassili Korotkine, James R. Forbes, and 1 more author

IEEE Robotics and Automation Letters, 2022

DOI Bib PDF

@article{9645374,
  author = {Guo, Zi Cong and Korotkine, Vassili and Forbes, James R. and Barfoot, Timothy D.},
  journal = {IEEE Robotics and Automation Letters},
  title = {Koopman Linearization for Data-Driven Batch State Estimation of Control-Affine Systems},
  year = {2022},
  volume = {7},
  number = {2},
  pages = {866-873},
  keywords = {Kernel;State estimation;Control systems;Robots;Training;Nonlinear systems;Training data;Localization;probabilistic inference},
  doi = {10.1109/LRA.2021.3133587},
  selected = true,
  bibtex_show = true
}