Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox
Picture for Igor Gilitschenski

Igor Gilitschenski

Neighborhood Mixup Experience Replay: Local Convex Interpolation for Improved Sample Efficiency in Continuous Control Tasks


May 18, 2022
Ryan Sander, Wilko Schwarting, Tim Seyde, Igor Gilitschenski, Sertac Karaman, Daniela Rus

Figure 1 for Neighborhood Mixup Experience Replay: Local Convex Interpolation for Improved Sample Efficiency in Continuous Control Tasks
Figure 2 for Neighborhood Mixup Experience Replay: Local Convex Interpolation for Improved Sample Efficiency in Continuous Control Tasks
Figure 3 for Neighborhood Mixup Experience Replay: Local Convex Interpolation for Improved Sample Efficiency in Continuous Control Tasks
Figure 4 for Neighborhood Mixup Experience Replay: Local Convex Interpolation for Improved Sample Efficiency in Continuous Control Tasks
* Accepted to L4DC 2022 
  Access Paper or Ask Questions

Learning Interactive Driving Policies via Data-driven Simulation


Nov 23, 2021
Tsun-Hsuan Wang, Alexander Amini, Wilko Schwarting, Igor Gilitschenski, Sertac Karaman, Daniela Rus

Figure 1 for Learning Interactive Driving Policies via Data-driven Simulation
Figure 2 for Learning Interactive Driving Policies via Data-driven Simulation
Figure 3 for Learning Interactive Driving Policies via Data-driven Simulation
Figure 4 for Learning Interactive Driving Policies via Data-driven Simulation
* The first two authors contributed equally to this this work. Code is available here: http://vista.csail.mit.edu/ 
  Access Paper or Ask Questions

VISTA 2.0: An Open, Data-driven Simulator for Multimodal Sensing and Policy Learning for Autonomous Vehicles


Nov 23, 2021
Alexander Amini, Tsun-Hsuan Wang, Igor Gilitschenski, Wilko Schwarting, Zhijian Liu, Song Han, Sertac Karaman, Daniela Rus

Figure 1 for VISTA 2.0: An Open, Data-driven Simulator for Multimodal Sensing and Policy Learning for Autonomous Vehicles
Figure 2 for VISTA 2.0: An Open, Data-driven Simulator for Multimodal Sensing and Policy Learning for Autonomous Vehicles
Figure 3 for VISTA 2.0: An Open, Data-driven Simulator for Multimodal Sensing and Policy Learning for Autonomous Vehicles
Figure 4 for VISTA 2.0: An Open, Data-driven Simulator for Multimodal Sensing and Policy Learning for Autonomous Vehicles
* First two authors contributed equally. Code and project website is available here: https://vista.csail.mit.edu 
  Access Paper or Ask Questions

Is Bang-Bang Control All You Need? Solving Continuous Control with Bernoulli Policies


Nov 03, 2021
Tim Seyde, Igor Gilitschenski, Wilko Schwarting, Bartolomeo Stellato, Martin Riedmiller, Markus Wulfmeier, Daniela Rus

Figure 1 for Is Bang-Bang Control All You Need? Solving Continuous Control with Bernoulli Policies
Figure 2 for Is Bang-Bang Control All You Need? Solving Continuous Control with Bernoulli Policies
Figure 3 for Is Bang-Bang Control All You Need? Solving Continuous Control with Bernoulli Policies
Figure 4 for Is Bang-Bang Control All You Need? Solving Continuous Control with Bernoulli Policies

  Access Paper or Ask Questions

HYPER: Learned Hybrid Trajectory Prediction via Factored Inference and Adaptive Sampling


Oct 05, 2021
Xin Huang, Guy Rosman, Igor Gilitschenski, Ashkan Jasour, Stephen G. McGill, John J. Leonard, Brian C. Williams

Figure 1 for HYPER: Learned Hybrid Trajectory Prediction via Factored Inference and Adaptive Sampling
Figure 2 for HYPER: Learned Hybrid Trajectory Prediction via Factored Inference and Adaptive Sampling
Figure 3 for HYPER: Learned Hybrid Trajectory Prediction via Factored Inference and Adaptive Sampling
Figure 4 for HYPER: Learned Hybrid Trajectory Prediction via Factored Inference and Adaptive Sampling
* 12 pages, 10 figures, 4 tables 
  Access Paper or Ask Questions

Deep Latent Competition: Learning to Race Using Visual Control Policies in Latent Space


Feb 19, 2021
Wilko Schwarting, Tim Seyde, Igor Gilitschenski, Lucas Liebenwein, Ryan Sander, Sertac Karaman, Daniela Rus

Figure 1 for Deep Latent Competition: Learning to Race Using Visual Control Policies in Latent Space
Figure 2 for Deep Latent Competition: Learning to Race Using Visual Control Policies in Latent Space
Figure 3 for Deep Latent Competition: Learning to Race Using Visual Control Policies in Latent Space
Figure 4 for Deep Latent Competition: Learning to Race Using Visual Control Policies in Latent Space
* Wilko, Tim, and Igor contributed equally to this work; published in Conference on Robot Learning 2020 
  Access Paper or Ask Questions

Deep Context Map: Agent Trajectory Prediction using Location-specific Latent Maps


Dec 14, 2019
Igor Gilitschenski, Guy Rosman, Arjun Gupta, Sertac Karaman, Daniela Rus

Figure 1 for Deep Context Map: Agent Trajectory Prediction using Location-specific Latent Maps
Figure 2 for Deep Context Map: Agent Trajectory Prediction using Location-specific Latent Maps
Figure 3 for Deep Context Map: Agent Trajectory Prediction using Location-specific Latent Maps
Figure 4 for Deep Context Map: Agent Trajectory Prediction using Location-specific Latent Maps

  Access Paper or Ask Questions

SiPPing Neural Networks: Sensitivity-informed Provable Pruning of Neural Networks


Oct 11, 2019
Cenk Baykal, Lucas Liebenwein, Igor Gilitschenski, Dan Feldman, Daniela Rus

Figure 1 for SiPPing Neural Networks: Sensitivity-informed Provable Pruning of Neural Networks
Figure 2 for SiPPing Neural Networks: Sensitivity-informed Provable Pruning of Neural Networks
Figure 3 for SiPPing Neural Networks: Sensitivity-informed Provable Pruning of Neural Networks
Figure 4 for SiPPing Neural Networks: Sensitivity-informed Provable Pruning of Neural Networks
* First two authors contributed equally 
  Access Paper or Ask Questions

Data-Dependent Coresets for Compressing Neural Networks with Applications to Generalization Bounds


Sep 05, 2018
Cenk Baykal, Lucas Liebenwein, Igor Gilitschenski, Dan Feldman, Daniela Rus

Figure 1 for Data-Dependent Coresets for Compressing Neural Networks with Applications to Generalization Bounds
Figure 2 for Data-Dependent Coresets for Compressing Neural Networks with Applications to Generalization Bounds
Figure 3 for Data-Dependent Coresets for Compressing Neural Networks with Applications to Generalization Bounds
Figure 4 for Data-Dependent Coresets for Compressing Neural Networks with Applications to Generalization Bounds
* First two authors contributed equally 
  Access Paper or Ask Questions

Map Management for Efficient Long-Term Visual Localization in Outdoor Environments


Aug 08, 2018
Mathias Bürki, Marcin Dymczyk, Igor Gilitschenski, Cesar Cadena, Roland Siegwart, Juan Nieto

Figure 1 for Map Management for Efficient Long-Term Visual Localization in Outdoor Environments
Figure 2 for Map Management for Efficient Long-Term Visual Localization in Outdoor Environments
Figure 3 for Map Management for Efficient Long-Term Visual Localization in Outdoor Environments
Figure 4 for Map Management for Efficient Long-Term Visual Localization in Outdoor Environments
* 7p 
  Access Paper or Ask Questions

Appearance-Based Landmark Selection for Efficient Long-Term Visual Localization


Aug 08, 2018
Mathias Bürki, Igor Gilitschenski, Elena Stumm, Roland Siegwart, Juan Nieto

Figure 1 for Appearance-Based Landmark Selection for Efficient Long-Term Visual Localization
Figure 2 for Appearance-Based Landmark Selection for Efficient Long-Term Visual Localization
Figure 3 for Appearance-Based Landmark Selection for Efficient Long-Term Visual Localization
Figure 4 for Appearance-Based Landmark Selection for Efficient Long-Term Visual Localization
* 7 pages 
  Access Paper or Ask Questions

LandmarkBoost: Efficient Visual Context Classifiers for Robust Localization


Jul 13, 2018
Marcin Dymczyk, Igor Gilitschenski, Juan Nieto, Simon Lynen, Bernhard Zeisl, Roland Siegwart

Figure 1 for LandmarkBoost: Efficient Visual Context Classifiers for Robust Localization
Figure 2 for LandmarkBoost: Efficient Visual Context Classifiers for Robust Localization
Figure 3 for LandmarkBoost: Efficient Visual Context Classifiers for Robust Localization
Figure 4 for LandmarkBoost: Efficient Visual Context Classifiers for Robust Localization

  Access Paper or Ask Questions

Learning 3D Segment Descriptors for Place Recognition


Apr 24, 2018
Andrei Cramariuc, Renaud Dubé, Hannes Sommer, Roland Siegwart, Igor Gilitschenski

Figure 1 for Learning 3D Segment Descriptors for Place Recognition
Figure 2 for Learning 3D Segment Descriptors for Place Recognition
Figure 3 for Learning 3D Segment Descriptors for Place Recognition
Figure 4 for Learning 3D Segment Descriptors for Place Recognition
* Presented at IROS 2017 Workshop on Learning for Localization and Mapping 
  Access Paper or Ask Questions

Free LSD: Prior-Free Visual Landing Site Detection for Autonomous Planes


Feb 25, 2018
Timo Hinzmann, Thomas Stastny, Cesar Cadena, Roland Siegwart, Igor Gilitschenski

Figure 1 for Free LSD: Prior-Free Visual Landing Site Detection for Autonomous Planes
Figure 2 for Free LSD: Prior-Free Visual Landing Site Detection for Autonomous Planes
Figure 3 for Free LSD: Prior-Free Visual Landing Site Detection for Autonomous Planes
Figure 4 for Free LSD: Prior-Free Visual Landing Site Detection for Autonomous Planes
* Accepted for publication in IEEE International Conference on Robotics and Automation (ICRA), 2018, Brisbane and IEEE Robotics and Automation Letters (RA-L), 2018 
  Access Paper or Ask Questions

Why and How to Avoid the Flipped Quaternion Multiplication


Feb 05, 2018
Hannes Sommer, Igor Gilitschenski, Michael Bloesch, Stephan Weiss, Roland Siegwart, Juan Nieto

Figure 1 for Why and How to Avoid the Flipped Quaternion Multiplication
Figure 2 for Why and How to Avoid the Flipped Quaternion Multiplication
Figure 3 for Why and How to Avoid the Flipped Quaternion Multiplication
Figure 4 for Why and How to Avoid the Flipped Quaternion Multiplication
* 16 pages, 1 figure, 2 tables (minor improvements and fixes over v1, smaller page margins) 
  Access Paper or Ask Questions

Voliro: An Omnidirectional Hexacopter With Tiltable Rotors


Jan 14, 2018
Mina Kamel, Sebastian Verling, Omar Elkhatib, Christian Sprecher, Paula Wulkop, Zachary Taylor, Roland Siegwart, Igor Gilitschenski

Figure 1 for Voliro: An Omnidirectional Hexacopter With Tiltable Rotors
Figure 2 for Voliro: An Omnidirectional Hexacopter With Tiltable Rotors
Figure 3 for Voliro: An Omnidirectional Hexacopter With Tiltable Rotors
Figure 4 for Voliro: An Omnidirectional Hexacopter With Tiltable Rotors
* Submitted to Robotics and Automation Magazine 
  Access Paper or Ask Questions

Directional Statistics and Filtering Using libDirectional


Dec 28, 2017
Gerhard Kurz, Igor Gilitschenski, Florian Pfaff, Lukas Drude, Uwe D. Hanebeck, Reinhold Haeb-Umbach, Roland Y. Siegwart

Figure 1 for Directional Statistics and Filtering Using libDirectional
Figure 2 for Directional Statistics and Filtering Using libDirectional
Figure 3 for Directional Statistics and Filtering Using libDirectional
Figure 4 for Directional Statistics and Filtering Using libDirectional
* Version accepted for Publication in the Journal of Statistical Software 
  Access Paper or Ask Questions

maplab: An Open Framework for Research in Visual-inertial Mapping and Localization


Nov 28, 2017
Thomas Schneider, Marcin Dymczyk, Marius Fehr, Kevin Egger, Simon Lynen, Igor Gilitschenski, Roland Siegwart

Figure 1 for maplab: An Open Framework for Research in Visual-inertial Mapping and Localization
Figure 2 for maplab: An Open Framework for Research in Visual-inertial Mapping and Localization
Figure 3 for maplab: An Open Framework for Research in Visual-inertial Mapping and Localization
Figure 4 for maplab: An Open Framework for Research in Visual-inertial Mapping and Localization

  Access Paper or Ask Questions

Visual-inertial self-calibration on informative motion segments


Aug 08, 2017
Thomas Schneider, Mingyang Li, Michael Burri, Juan Nieto, Roland Siegwart, Igor Gilitschenski

Figure 1 for Visual-inertial self-calibration on informative motion segments
Figure 2 for Visual-inertial self-calibration on informative motion segments
Figure 3 for Visual-inertial self-calibration on informative motion segments
Figure 4 for Visual-inertial self-calibration on informative motion segments
* Robotics and Automation (ICRA), 2017 IEEE International Conference on 
  Access Paper or Ask Questions

Recursive Bayesian Filtering in Circular State Spaces


Jan 21, 2015
Gerhard Kurz, Igor Gilitschenski, Uwe D. Hanebeck

Figure 1 for Recursive Bayesian Filtering in Circular State Spaces
Figure 2 for Recursive Bayesian Filtering in Circular State Spaces
Figure 3 for Recursive Bayesian Filtering in Circular State Spaces
Figure 4 for Recursive Bayesian Filtering in Circular State Spaces

  Access Paper or Ask Questions

Unscented Orientation Estimation Based on the Bingham Distribution


Nov 22, 2013
Igor Gilitschenski, Gerhard Kurz, Simon J. Julier, Uwe D. Hanebeck

Figure 1 for Unscented Orientation Estimation Based on the Bingham Distribution
Figure 2 for Unscented Orientation Estimation Based on the Bingham Distribution
Figure 3 for Unscented Orientation Estimation Based on the Bingham Distribution

  Access Paper or Ask Questions

Recursive Estimation of Orientation Based on the Bingham Distribution


Apr 30, 2013
Gerhard Kurz, Igor Gilitschenski, Simon Julier, Uwe D. Hanebeck

Figure 1 for Recursive Estimation of Orientation Based on the Bingham Distribution
Figure 2 for Recursive Estimation of Orientation Based on the Bingham Distribution
Figure 3 for Recursive Estimation of Orientation Based on the Bingham Distribution
Figure 4 for Recursive Estimation of Orientation Based on the Bingham Distribution

  Access Paper or Ask Questions