DARPA Subterranean Challenge

We participate in the DARPA competition as CTU-CRAS-NORLAB team selected for DARPA Subterranean or "SubT" Challenge. For sixty years, DARPA has held to a singular and enduring mission: to make pivotal investments in breakthrough technologies for national security with its annual budget: $3.427 billion!

We won the 1st place among the non-funded teams in the
Urban & Tunnel Circuits!



The SubT Challenge is aimed to revolutionize how we operate in the underground domain. Current technologies fail to provide persistent situational awareness of the diverse subterranean operating environments, including tunnels, urban underground, and cave networks. Complex underground settings present significant challenges for military and civilian first responders. The hazards vary drastically across domains that can degrade or change over time and are often too high-risk for personnel to enter. The DARPA SubT Challenge seeks novel approaches to rapidly map, navigate, and search underground environments during time-sensitive disaster response scenarios. DARPA encourages participation from multidisciplinary teams from around the world to address the autonomy, perception, networking, and mobility technologies necessary to map subsurface networks in unpredictable conditions.


CTU Martin Saska, Tomáš Báča, Vojtěch Spurný, Daniel Heřt, Matěj Petrlík, Matouš Vrba, Vít Krátký, Pavel Petráček, Yurii Stasinchuk
UPENN Tomáš Svoboda (team lead), Karel Zimmermann, Tomáš Petříček, Martin Pecka, Vojtěch Šalanský
Jan Faigl, Petr Čížek, Jan Bayer, Tomáš Rouček




  1. Vít Krátký, Pavel Petráček, Tomáš Báča and Martin Saska. An autonomous unmanned aerial vehicle system for fast exploration of large complex indoor environments. Journal of Field Robotics, pages 1-24, May 2021. PDF, DOI BibTeX

    	author = "Krátký, Vít and Petráček, Pavel and Báča, Tomáš and Saska, Martin",
    	title = "An autonomous unmanned aerial vehicle system for fast exploration of large complex indoor environments",
    	journal = "Journal of Field Robotics",
    	pages = "1-24",
    	year = 2021,
    	month = "May",
    	keywords = "aerial robotics, exploration, extreme environments, GPS-denied operation, planning",
    	doi = "https://doi.org/10.1002/rob.22021",
    	pdf = "https://onlinelibrary.wiley.com/doi/epdf/10.1002/rob.22021"
  2. M Petrlík, T Báča, D Heřt, M Vrba, T Krajník and M Saska. A Robust UAV System for Operations in a Constrained Environment. IEEE Robotics and Automation Letters (RAL) 5(2):2169-2176, 2020. URL PDF, DOI BibTeX

    	author = "M. {Petrl\'{i}k} and T. {B\'{a}\v{c}a} and D. {He\v{r}t} and M. {Vrba} and T. {Krajn\'{i}k} and M. {Saska}",
    	journal = "IEEE Robotics and Automation Letters (RAL)",
    	title = "A Robust UAV System for Operations in a Constrained Environment",
    	volume = 5,
    	number = 2,
    	pages = "2169-2176",
    	year = 2020,
    	keywords = "Aerial Systems;Perception and Autonomy;Search and Rescue Robots;Robotics in Hazardous Fields",
    	doi = "10.1109/LRA.2020.2970980",
    	issn = "2377-3774",
    	url = "https://ieeexplore.ieee.org/document/8979150",
    	pdf = "data/papers/petrlik2020ral.pdf"
  3. T Rouček, M Pecka, P Čížek, T Petříček, J Bayer, V Šalanský, D Heřt, M Petrlík, T Báča, V Spurný, F Pomerleau, V Kubelka, J Faigl, K Zimmermann, M Saska, T Svoboda and T Krajník. DARPA Subterranean Challenge: Multi-robotic Exploration of Underground Environments. In Modelling and Simulation for Autonomous Systems. 2019, 274–290. URL PDF, DOI BibTeX

    	title = "DARPA Subterranean Challenge: Multi-robotic Exploration of Underground Environments",
    	author = "T. Rou\v{c}ek and M. Pecka and P. \v{C}\'{i}\v{z}ek and T. Pet\v{r}\'{i}\v{c}ek and J. Bayer and V. \v{S}alansk\'{y} and D. He\v{r}t and M. Petrl\'{i}k and T. B\'{a}\v{c}a and V. Spurn\'{y} and F. Pomerleau and V. Kubelka and J Faigl and K. Zimmermann and M. Saska and T. Svoboda and T. Krajn\'{i}k",
    	booktitle = "Modelling and Simulation for Autonomous Systems",
    	year = 2019,
    	pages = "274--290",
    	isbn = "978-3-030-43890-6",
    	doi = "10.1007/978-3-030-43890-6_22",
    	url = "https://link.springer.com/chapter/10.1007/978-3-030-43890-6_22",
    	pdf = "data/papers/mesas2019darpa.pdf"



DARPA SubT Cave Circuit - 10/2020

As a part of the CTU-CRAS-NORLAB team, we have participated in the virtual competition this time (unfortunately, the COVID-19 pandemic forced the organizers to cancel the system's track of Cave Circuit). We are eagerly awaiting the results of the virtual challenge scheduled for the end of October, 2020.

However, we did not postpone our research and we set the feet and the propellers of our robots into the Býčí skála (cave system in the southern part of Moravian Karst, Czech Republic). See the video below showing our progress in preparation for the final round of the DARPA SubT competition.

DARPA SubT Urban Circuit - 2/2020: 1st place among self-funded teams, 3rd place in total

As a part of the CTU-CRAS-NORLAB team at the DARPA Subterranean Challenge, we have supported the achievements of Tunnel Circuit by winning the 1st place among self-funded teams and secured 3rd place overall. The team brought home a $500 000 prize. Pictures from the preparations and from the competition itself are publicly available within Google Photos album.


DARPA SubT Tunnel Circuit - 8/2019: 1st place among self-funded teams, 3rd place in total

As a part of the CTU-CRAS team at the DARPA Subterranean Challenge, we have achieved 1st place among self-funded teams, and secured a 3rd place overall. The team brough home a $200 000 prize. Our great achievement has been shown during the main news sequence broadcast by Česká televize. The competition was motivated by creation of the search and rescue system, that will be deployed during a mine accident.



Pictures taken during the DARPA Tunnel Circuit

DARPA SubT Integration Exercise (STIX) - 4/2019

During the first round of the competition in a real gold mine in Rocky Mountains close to Denver, USA, our team achieved a valuable result. According to our information, our drones in total came through the longest overall distance from all competitors. Our team achieved the highest scoring from all teams in our group (results of teams from other groups are confidential), which gives us a great chance for the final round of tunnel circuit challenge being held in research mine near Pittsburgh, USA, in 8/2019.


Pictures taken during the first DARPA STIX


Media coverage of our DARPA SubT Challenge participation:

Vědci z FEL ČVUT získali prestižní grant od americké agentury DARPA Česká televize (Studio 6) [Czech]
Čeští experti získali 1,5 milionu dolarů od DARPA na vývoj dronů. Nakoupí psoboty Spot Česká televize [Czech]
Výzkumníci z ČVUT získali 1,5 milionu dolarů od americké agentury DARPA na další vývoj autonomních robotů Věda a výzkum [Czech]
Vědce z ČVUT dotuje americká armádní agentura. Vyvíjejí autonomní záchranné roboty e15 [Czech]
Výzkumníci z Fakulty elektrotechnické ČVUT získali 1,5 milionu dolarů od americké agentury DARPA na další vývoj autonomních robotů Armádní noviny [Czech]
Teams Coordinated Robotics, BARCS, and Dynamo On Top in DARPA Subterranean Challenge Cave Circuit Virtual Competition DARPA [English]
Naši vědci v jeskyni Býčí skála zkouší roboty, kteří najdou zavalené lidi i věci Katedra kybernetiky FEL ČVUT [Czech]
Naše drony a pozemní roboty budou zkoumat jeskynní komplex Býčí skála Katedra kybernetiky FEL ČVUT [Czech]
Úspěch robotů a dronů ČVUT v soutěži ministerstva obrany USA Student HW [Czech]
Naši robotici vyhráli světovou soutěž DARPA Subterranean Challenge Urban Circuit mezi nesponzorovanými týmy a obsadili celkově 3. místo FEL ČVUT [Czech]
Drony z ČVUT v týdnu zvítězily v Abú Zabí i USA. Naši vědci patří ke světové špičce Živě.cz [Czech]
Robots and their masters take over nuclear plant for DARPA’s Subterranean Challenge GeekWire [English]
CTU-CRAS-NORLAB earned the highest score among self-funded teams and the $500,000 prize DARPA [English]
Martin Pecka, Karel Zimmermann (ČVUT): V soutěži DARPA jsme uspěli, protože jsme nebyli líní trénovat Lupa.cz [Czech]

Call for participation in our team:

Multi-robot Systems group is seeking highly motivated students with strong interest in one or more of the following research areas related to the competition:

  • MAV control and stabilization
  • Motion planning and task allocation
  • Computer vision
  • Visual servoing

We require: C, C++ and/or Matlab programming skills, good study results, and good level of English. Experience with embedded programming, hardware and robotics is an advantage. Enthusiasts in aviation and models of multi-rotor helicopters welcome. 

We offer: work in young and ambitious team of master and PhD students, international experience in highly recognized competition among the best universities, guidance and expertise of respected scientists in the given fields including researchers from worldwide top MAV laboratory at UPENN, high-tech equipment, and scholarship for work beyond university duties.

Contact: Martin Saska