social linkedin box blue 32
social facebook box blue 32
social twitter box blue 32
social facebook box blue 32

iit-logo-v3-mainsite

People ■ Petar Kormushev
Petar Kormushev's foto

Petar Kormushev

Team Leader

ADVR / Learning and Interaction

Researcher
Phone +39 010 717 81 243
Mobile Phone +39 334 382 42 62
IM Address pkormushev (Skype)
Web Site http://kormushev.com/
Social network profiles
linkedin icon youtube icon

Followers

Profile image

Bio

I am a Team Leader (equivalent to Assistant Professor) at the  Advanced Robotics department of the Italian Institute of Technology, doing research in robotics and machine learning under the supervision of Prof. Darwin Caldwell. 

I am also a Visiting Senior Research Fellow at King’s College London, UK, where I collaborate with Prof. Kaspar Althoefer and Prof. Maria Fox at the Department of Informatics.

My research is focused on reinforcement learning algorithms and their application to robot learning. You can visit the  Research section for more information about my current and past research, as well as the  Publications section to see my published papers. You can watch videos from my research experiments with robots in the  Videos section.

At IIT, I am the Head of the  Learning and Interaction Lab of the Advanced Robotics department. We develop machine learning algorithms and apply them to robots like the compliant humanoid robot  COMAN, the iCub humanoid robot, the Barrett WAM arm manipulator robot, and the Fujitsu HOAP-2 small humanoid robot.

In 2009 I obtained my PhD degree from Tokyo Institute of Technology, where I was doing research in computational intelligence under the supervision of Prof. Kaoru Hirota from  Hirota Laboratory. My PhD thesis was dedicated to methods for speeding up the reinforcement learning process. I was also supervised by Visiting Prof. Kohei Nomoto from Mitsubishi Electric Corp. and Visiting Prof. Shigeaki Sakurai from Toshiba Corp.

I have had the chance to collaborate with many excellent researchers in robotics and machine learning, including Prof. Dragomir N. Nenchev from Tokyo City University, Assoc. Prof. Gennady Agre from Bulgarian Academy of Sciences, and Dr. Barkan Ugurlu from Toyota Technological Institute. I am deeply grateful to my university lecturer Assoc. Prof. Maria Nisheva, for inspiring me to pursue studies in artificial intelligence.

In addition to my scientific research, I have more than 7 years of working experience. In 2008, I worked at Google Japan for 3 months as a software engineer in the Search Quality team. I created a prototype of a new search query categorization system (which I called Google Genus) using machine learning algorithms.

In 2005, I received my MSc degree in Artificial Intelligence from Sofia University, at the Faculty of Mathematics and Informatics. In 2006, shortly before going to Japan, I successfully defended my second MSc degree in Bio- and Medical Informatics. My long-term goal is to combine knowledge from different scientific fields in order to achieve synergetic effect and be able to tackle very complex problems.

Videos of my research experiments:
http://kormushev.com/research/videos/

Click below in Publications or Projects section for more info:

Projects

Pandora_transp_bg_trimmed
PANDORA
 (Persistent Autonomy through learNing, aDaptation, Observation and Re-plAnning)
FP7-ICT-288273, STREP, affiliated with ADVR-IIT
2012-2014
SF_button_transp_bg STIFF-FLOP (STIFFness controllable Flexible and Learn-able Manipulator for surgical OPerations)
FP7-ICT-287728, IP, affiliated with ADVR-IIT
2012-2015
AMARSi_logo_transp_bg AMARSI (Adaptive Modular Architectures for Rich Motor Skills)
FP7-ICT-248311, IP, affiliated with ADVR-IIT
2010-2014
infrawebs_logo_smaller INFRAWEBS (Intelligent Framework for Generating Open Adaptable Development Platforms for Web-Service Enabled Applications Using Semantic Web Technologies)
FP6-IST-511723, affiliated with BAS-BG
2004-2006



 

Selected Publications


My complete up-to-date Publications list is here. Below is a selected subset.

Petar Kormushev, Yiannis Demiris, Darwin G. Caldwell, "Encoderless Position Control of a Two-Link Robot Manipulator"In Proc. IEEE Intl Conf. on Robotics and Automation (ICRA 2015), Seattle, USA, 2015.

Nawid Jamali, Petar Kormushev, Arnau Carrera, Marc Carreras, Darwin G. Caldwell, "Underwater Robot-Object Contact Perception using Machine Learning on Force/Torque Sensor Feedback"In Proc. IEEE Intl Conf. on Robotics and Automation (ICRA 2015), Seattle, USA, 2015.

Seyed Reza Ahmadzadeh, Ali Paikan, Fulvio Mastrogiovanni, Lorenzo Natale, Petar Kormushev, Darwin G. Caldwell, "Learning Symbolic Representations of Actions from Human Demonstrations"In Proc. IEEE Intl Conf. on Robotics and Automation (ICRA 2015), Seattle, USA, 2015.

Joao Bimbo, Petar Kormushev, Kaspar Althoefer, Hongbin Liu, "Global Estimation of an Object's Pose Using Tactile Sensing"In Advanced Robotics, The Robotics Society of Japan, Tokyo, Japan, 2015.

Rodrigo S. Jamisola, Petar Kormushev, Antonio Bicchi, and Darwin G. Caldwell, "Haptic Exploration of Unknown Surfaces with Discontinuities", In Proc. IEEE/RSJ Intl Conf. on Intelligent Robots and Systems (IROS 2014), 2014. [details] 


Robot-Object Contact Perception using Symbolic Temporal Pattern Learning. Nawid Jamali, Petar Kormushev and Darwin G. Caldwell, IEEE Intl. Conf. on Robotics and Automation (ICRA 2014), Hong Kong, China, 2014.
[pdf] [bibtex]

Online Discovery of AUV Control Policies to Overcome Thruster Failures. Seyed Reza Ahmadzadeh, Arnau Carrera, Matteo Leonetti, Petar Kormushev and Darwin G. Caldwell, IEEE Intl. Conf. on Robotics and Automation (ICRA 2014), Hong Kong, China, 2014.
[pdf] [bibtex]

Covariance Analysis as a Measure of Policy Robustness in Reinforcement Learning. Nawid Jamali, Petar Kormushev, Seyed Reza Ahmadzadeh, and Darwin G. Caldwell, OCEANS 2014 MTS/IEEE, Taipei, Taiwan, 2014.
[pdf] [bibtex]

Towards dynamically consistent real-time gait pattern generation for full-size humanoid robots. Przemyslaw Kryczka, Yukitoshi Minami Shiguematsu, Petar Kormushev, Kenji Hashimoto, Hun-ok Lim and Atsuo Takanishi. Proc. ROBIO 2013.
[pdf] [bibtex]

Reinforcement Learning in Robotics: Applications and Real-World Challenges. Petar Kormushev, Sylvain Calinon and Darwin G. Caldwell. MDPI Journal of Robotics (ISSN 2218-6581), Special Issue on Intelligent Robots, vol.2, pp.122-148, 2013.
[pdf] [bibtex]

Improving the Energy Efficiency of Autonomous Underwater Vehicles by Learning to Model Disturbances. Petar Kormushev and Darwin G. Caldwell. Proc. IEEE/RSJ Intl Conf. on Intelligent Robots and Systems (IROS 2013), Tokyo, Japan, 2013.
[pdf] [bibtex]

Visuospatial Skill Learning for Object Reconfiguration Tasks. Seyed Reza Ahmadzadeh, Petar Kormushev and Darwin G. Caldwell. Proc. IEEE/RSJ Intl Conf. on Intelligent Robots and Systems (IROS 2013), Tokyo, Japan, 2013.
[pdf] [bibtex]

On-line Identification of Autonomous Underwater Vehicles Through Global Derivative-free Optimization. George Karras, Charalampos Bechlioulis, Matteo Leonetti, Narcis Palomeras, Petar Kormushev, Kostas Kyriakopoulos and Darwin G. Caldwell. Proc. IEEE/RSJ Intl Conf. on Intelligent Robots and Systems (IROS 2013), Tokyo, Japan, 2013.
[pdf] [bibtex]

Hybrid gait pattern generator capable of rapid and dynamically consistent pattern regeneration. Przemyslaw Kryczka, Petar Kormushev, Kenji Hashimoto, Hun-ok Lim, Nikolaos Tsagarakis, Darwin G. Caldwell and Atsuo Takanishi. Proc. URAI 2013.
[pdf] [bibtex]

Interactive Robot Learning of Visuospatial Skills. Seyed Reza Ahmadzadeh, Petar Kormushev and Darwin G. Caldwell. Proc. IEEE Intl Conf. on Advanced Robotics (ICAR 2013), Montevideo, Uruguay, 2013.
[pdf] [bibtex]

Reinforcement Learning with Heterogeneous Policy Representations. Petar Kormushev and Darwin G. Caldwell. The 11th European Workshop on Reinforcement Learning (EWRL 2013) held as a Dagstuhl Seminar, Dagstuhl, Germany, 2013.
[pdf] [bibtex]

Comparative Evaluation of Reinforcement Learning with Scalar Rewards and Linear Regression with Multidimensional Feedback. Petar Kormushev and Darwin G. Caldwell. The 2013 ECML/PKDD Workshop on Reinforcement Learning from Generalized Feedback: Beyond numeric rewards, Prague, Czech Republic, 2013.
[pdf] [bibtex]

On-line Learning to Recover from Thruster Failures on Autonomous Underwater Vehicles. Matteo Leonetti, Reza Ahmadzadeh and Petar Kormushev, OCEANS 2013 MTS/IEEE, San Diego, USA, 2013.
[pdf] [bibtex]

Towards Improved AUV Control Through Learning of Periodic Signals. Petar Kormushev and Darwin G. Caldwell, OCEANS 2013 MTS/IEEE, San Diego, USA, 2013.
[pdf] [bibtex]

Contact State Estimation using Machine Learning. Nawid Jamali, Petar Kormushev and Darwin G. Caldwell, OCEANS 2013 MTS/IEEE, San Diego, USA, 2013.
[pdf] [bibtex]

Online Direct Policy Search for Thruster Failure Recovery in Autonomous Underwater Vehicles. Seyed Reza Ahmadzadeh, Matteo Leonetti and Petar Kormushev. Evolutionary and Reinforcement Learning for Autonomous Robot Systems (ERLARS 2013), Taormina, Italy, 2013.
[pdf] [bibtex]

Autonomous Robotic Valve Turning: A Hierarchical Learning Approach. Seyed Reza Ahmadzadeh, Petar Kormushev and Darwin G. Caldwell, IEEE Intl. Conf. on Robotics and Automation (ICRA 2013), Karlsruhe, Germany, 2013.
[pdf] [bibtex]

Towards valve turning with an AUV using learning by demonstration. Carrera, A., M. Carreras, P. Kormushev, N. Palomeras, and S. Nagappa. OCEANS 2013 MTS/IEEE, Bergen, Norway, 2013.
[pdf] [bibtex]

Walking despite the passive compliance: Techniques for using conventional pattern generators to control intrinsically compliant humanoid robots. Przemyslaw Kryczka, Petar Kormushev, H. O. Lim, K. Hashimoto, A. Takanishi, N. G. Tsagarakis, D. G. Caldwell, CLAWAR 2013.
[pdf] [bibtex]

Development of a Dynamic Simulator for a Compliant Humanoid Robot Based on a Symbolic Multibody Approach. H. Dallali, M. Mosadeghzad, G. Medrano-Cerda, N. Docquier, P. Kormushev, N. Tsagarakis, Z. Li, D. Caldwell, International Conference on Mechatronics (ICM 2013), Vicenza, Italy, February 27 – March 1, 2013.
[pdf] [bibtex]

Compliant skills acquisition and multi-optima policy search with EM-based reinforcement learning. Sylvain Calinon, Petar Kormushev, Darwin G. Caldwell. Robotics and Autonomous Systems, Volume 61, Issue 4, pp. 369-379, 2013.
[pdf] [bibtex]

The Anatomy of a Fall: Automated Real-time Analysis of Raw Force Sensor Data from Bipedal Walking Robots and Humans. Kormushev, P., Ugurlu, B., Colasanto, L., Tsagarakis, N., and Caldwell, D.G., Proc. IEEE/RSJ Intl Conf. on Intelligent Robots and Systems (IROS 2012), Portugal, 2012.
[pdf] [bibtex]

Simultaneous Discovery of Multiple Alternative Optimal Policies by Reinforcement Learning. Petar Kormushev and Darwin G. Caldwell. Proc. of the IEEE Intelligent Systems Conference (IS 2012), Sofia, Bulgaria, pp. 202-207, 2012.
[pdf] [bibtex] IS 2012 BEST PAPERS

Towards autonomous robotic valve turning. Arnau Carrera Vinas, Seyed Reza Ahmadzadeh, Arash Ajoudani, Petar Kormushev, Marc Carreras, Darwin G. Caldwell. International Journal of Cybernetics and Information Technologies, Vol. 12, No. 3, pp. 17-26, 2012.
[pdf] [bibtex]

Persistent autonomy: the challenges of the PANDORA project. David M. Lane, Francesco Maurelli, Petar Kormushev, Marc Carreras, Maria Fox, and Konstantinos Kyriakopoulos. In Proceedings of IFAC MCMC 2012 – Manoeuvring and Control of Marine Craft, 2012.
[pdf] [bibtex]

On Global Optimization of Walking Gaits for the Compliant Humanoid Robot COMAN Using Reinforcement Learning. Houman Dallali, Petar Kormushev, Zhibin Li, Darwin G. Caldwell. International Journal of Cybernetics and Information Technologies, Vol. 12, No. 3, pp. 39-52, 2012.
[pdf] [bibtex]

Combining Local and Global Direct Derivative-free Optimization for Reinforcement Learning. Matteo Leonetti, Petar Kormushev, Simone Sagratella. International Journal of Cybernetics and Information Technologies, Vol. 12, No. 3, pp. 53-65, 2012.
[pdf] [bibtex]

Learning Fast Quadruped Robot Gaits with the RL PoWER Spline Parameterization. Haocheng Shen, Jason Yosinski, Petar Kormushev, Darwin G. Caldwell and Hod Lipson. International Journal of Cybernetics and Information Technologies, Vol. 12, No. 3, pp. 66-75, 2012.
[pdf] [bibtex]

Optimization of a compact model for the compliant humanoid robot COMAN using reinforcement learning. Luca Colasanto, Petar Kormushev, Nikolaos Tsagarakis, Darwin G. Caldwell. International Journal of Cybernetics and Information Technologies, Vol. 12, No. 3, pp. 76-85, 2012.
[pdf] [bibtex]

Direct Policy Search Reinforcement Learning based on Particle Filtering. Kormushev, P. and Caldwell, D.G., European Workshop on Reinforcement Learning (EWRL 2012), part of the International Conference on Machine Learning (ICML 2012), Edinburgh, 2012.
[pdf] [bibtex]

Challenges for the Policy Representation when Applying Reinforcement Learning in Robotics. Kormushev, P., Calinon, S., Ugurlu, B., and Caldwell, D.G., International Joint Conference on Neural Networks (IJCNN 2012), part of IEEE World Congress on Computational Intelligence (WCCI 2012), Brisbane, 2012.
[pdf] [bibtex]

Bipedal Walking Energy Minimization by Reinforcement Learning with Evolving Policy Parameterization. Kormushev, P., Ugurlu, B., Calinon, S., Tsagarakis, N., and Caldwell, D.G., Proc. IEEE/RSJ Intl Conf. on Intelligent Robots and Systems (IROS-2011), San Francisco, 2011.
[pdf] [bibtex]

Upper-body Kinesthetic Teaching of a Free-standing Humanoid Robot. Kormushev, P., Nenchev, D.N., Calinon, S., and Caldwell, D.G., IEEE Intl. Conf. on Robotics and Automation (ICRA 2011), pp. 3970-3975, 2011.
[pdf] [bibtex]

Imitation learning of positional and force skills demonstrated via kinesthetic teaching and haptic input. Kormushev, P., Calinon, S., and Caldwell, D.G., Advanced Robotics, Vol. 25, pp. 581-603, 2011.
[pdf] [bibtex]

Time Hopping Technique for Faster Reinforcement Learning in Simulations. Kormushev, P., Nomoto, K., Dong, F., and Hirota, K., International Journal of Cybernetics and Information Technologies, Vol. 11, No. 3, pp. 42-59, 2011.
[pdf] [bibtex]

Whiteboard Cleaning Task Realization with HOAP-2. Sato, F., Nishii, T., Takahashi, J., Yoshida, Y., Mitsuhashi, M., Kormushev, P., Kanamiya, Y., Proc. SICE System Integration (SI-2010) in Sendai, Japan, pp.426-429, 2010.
[pdf] [bibtex]

Approaches for Learning Human-like Motor Skills which Require Variable Stiffness During Execution. Kormushev, P., Calinon, S., and Caldwell, D.G., Workshop on Humanoid Robots Learning from Human Interaction (link), Humanoids-2010, Nashville, USA, 2010.
[pdf] [bibtex]

Learning the skill of archery by a humanoid robot iCub. Kormushev, P., Calinon, S., Saegusa, R. and Metta, G., Proc. IEEE Intl Conf. on Humanoid Robots (Humanoids-2010), pp. 417-423, 2010.
[pdf] [bibtex]

Robot Motor Skill Coordination with EM-based Reinforcement Learning. Kormushev, P., Calinon, S. and Caldwell, D.G., Proc. IEEE/RSJ Intl Conf. on Intelligent Robots and Systems (IROS-2010), 2010.
[pdf] [bibtex] Top-5 most cited paper from IROS’10

Time Hopping Technique for Reinforcement Learning and its Application to Robot Control. Kormushev, P., Dept. of Computational Intelligence and Systems Science, Tokyo Institute of Technology, PhD thesis, September, 2009.
[pdf] [bibtex]

Probability redistribution using time hopping for reinforcement learning. Kormushev, P., K., Dong, F., and Hirota, K., 10th International Symposium on advanced Intelligent Systems ISIS-2009, 2009.
[pdf] [bibtex]

Eligibility propagation to speed up time hopping for reinforcement learning. Kormushev, P., Nomoto, K., Dong, F., and Hirota, K., Journal of Advanced Computational Intelligence and Intelligent Informatics, Vol.13, No.6, 2009.
[pdf] [bibtex]

Time manipulation technique for speeding up reinforcement learning in simulations. Kormushev, P., Nomoto, K., Dong, F., and Hirota, K., International Journal of Cybernetics and Information Technologies, Vol. 8, No. 1, pp. 12-24, January, 2008.
[pdf] [bibtex]

Intent expression using eye robot for mascot robot system. Yamazaki, Y., Dong, F., Masuda, Y., Uehara, Y., Kormushev, P., Vu, H. A., Le, P. Q., and Hirota, K., 8th International Symposium on Advanced Intelligent Systems ISIS-2007, 2007.
[pdf] [bibtex]

Fuzzy inference based mentality estimation for eye robot agent. Yamazaki, Y., Dong, F., Masuda, Y., Uehara, Y., Kormushev, P., Vu, H. A., Le, P. Q., and Hirota, K., Proceedings of 23rd Fuzzy System Symposium FSS-2007, 2007.
[pdf] [bibtex]

INFRAWEBS Axiom editor – a graphical ontology-driven tool for creating complex logical expressions. Agre, G., Kormushev, P. and Dilov, I., International Journal of Information Theories and Applications, Vol. 13, No. 2, pp. 169-178, November, 2006.
[pdf] [bibtex]

Visual approach for data mining on medical information databases using Fastmap algorithm. Kormushev, P., M.Sc. thesis, Faculty of Mathematics and Informatics, Sofia University, March, 2006.
[pdf] [bibtex]

INFRAWEBS Axiom Editor User’s Guide. Agre, G., Kormushev, P. and Dilov, I., 2006.
[pdf] [bibtex]

Design, development and implementation of a tool for construction of declarative functional descriptions of semantic web services based on WSMO methodology. Kormushev, P., M.Sc. thesis, Faculty of Mathematics and Informatics, Sofia University, July, 2005.
[pdf] [bibtex]

INFRAWEBS Capability editor – A graphical ontology-driven tool for creating capabilities of Semantic Web Services. Agre, G., Kormushev, P. and Dilov, I., Third International Conference on Information Research, Applications and Education i.TECH-2005, June, 2005.
[pdf] [bibtex]




Download my full publications list as a BibTeX file



Download my full publications list as a PDF file



Find me on Google Scholar

 

Awards

“John Atanasoff” Award, 2013
Awarded by the President of Bulgaria for scientific excellence and contributions to the development of Information and Communications Technologies (ICT) in Bulgaria and abroad. The award bears the name of Prof. John Atanasoff (who is of Bulgarian descent) – the inventor of the first electronic digital computer. More info here

Japanese Research Fellowship, 2006 – 2009
Awarded by the Japanese Government (MEXT/Monbukagakusho) and Tokyo Institute of Technology, to finance my doctoral research in robotics in Japan.

The “St. Kliment Ohridski” Award, 2005
Awarded by the President of Sofia University Prof. Boyan Biolchev, for exceptional academic achievements and extracurricular activities, in the category “Master student”. More info here

“John Atanasoff” Scholarship, 2002
Awarded by Eureka Foundation in 2002 for outstanding achievements in Computer Science. This was the first “John Atanasoff” scholarship. It was dedicated to the 100th birthday of John Atanasoff – the inventor of Bulgarian origin, who created the first electronic digital computer. Later in 2003, the Bulgarian president Georgi Parvanov created another, presidential award by the same name of “John Atanasoff”.

3rd place in the Bulgarian National Programming Competition, 2001
Awarded by Musala Soft and PC Magazine Bulgaria http://konkurs.musala.com

INFORMATION NOTICE ON COOKIES

IIT's website uses the following types of cookies: browsing/session, analytics, functional and third party cookies. Users can choose whether or not to accept the use of cookies and access the website.
By clicking on further information, the full information notice on the types of cookies used will be displayed and you will be able to choose whether or not to accept cookies whilst browsing on the website.

Try our new site and tell us what you think
Take me there