profile_picture

Dr.-Ing. Oliver Urbann

Team Leader Autonomous Robotics Group
Robotics Research Institute
TU Dortmund University


Research Interests

Starting in 2007 on the RoboCup project of the Robotics Research Institute at TU Dortmund University in Germany, my central research interest is biped walking, control theory and sensor fusion. The RoboCup is an international robot soccer competition with various leagues, including many leagues using biped robots. As a key to success, biped walking must be fast, stable and precise. We are known as the fastest team (video) in our league with 20-30 competitors and are currently the world champions in the SPL Outdoor Competition, see our paper about playing robot soccer outdoor for details.

My current and future research focusses on walking and manipulating control for cost-effective and leightweight robots like the Nao by SoftBank Robotics which is the official robot of the Standard Platform League. Our approach is currently based on a preview controller with observer while I also considered and implemented other approaches like Model Predictice Control. Extensions of the utilized robot model by a mass-spring-damper system called Flexible Linear Inverted Pendulum Model (FLIPM) improves the performance of the walking controller and is therefore also a research focus. Furthermore, in cooperation with Leonardo Lanari and Seth Hutchinson we combine the Boundedness Control approach and FLIPM.

Besides this my team is working on many aspects required for autonomous robotics: vision in varying lightning conditions, behavioural decisions, localization based on multi-modal kalman and particle filter and path planning algorithms like potential fields.

Education

TU Dortmund University

Dr.-Ing. Computer Science (Ph.D. equivalent, summa cum laude)

Walking of Humanoid Robots Based on Control Theory and Real-Time Reactive Stepping (german), keywords: Model Predictive and Preview Control, Observer

Jan. 2017

TU Dortmund University

Diplom Computer Science (M.Sc. equivalent)

Analysis and Compensation of Walk Related Effects on Biped Robots (german), keywords: Rigid Body Simulation, AI, Genetic Algorithms, Inverse Dynamics

2010

Experience

Team Leader Autonomous Robotics Group as Research Associate

Robotics Research Institute, TU Dortmund University
  • Supervising team consisting of two research associates, up to three student research assistants and up to 15 students.
  • Research in the area of autonomous and humanoid robotics: control, senssor feedback, path planning, vision, image processing, object recognition.
  • Teaching: ”Technical Computer Science” (practice groups, exams, organisation), seminars, project groups, robotic tutorials
  • Participation in RoboCup and different local robot soccer competitions.
  • Project "Virtual Forrest", enabling distributed computation of automatic tree recognition from aerial photographs.
  • "Robotic Analyze" (RobAs) software analyzing data of control elements tested by automated robotic manipulators.
  • Project "EMC²": Embedded multi-core systems for mixed criticality applications in dynamic and changeable real-time environments.
  • 2010-Now

    Student research assistant

    Robotics Research Institute, TU Dortmund University

    Development of robotic framework (Linux, C++)

    2008-2010

    Student research assistant

    Leibniz Research Centre for Working Environment and Human Factors

    Development of high-precision reactions tests (software and hardware, C# and C++), increased accuracy from 5 ms to 10 𝜇s.

    2005-2008

    PocketPC Development Freelancer

    Development of PocketPC software in C/C++ for various companies.

    2001-2004

    Achievements

    Word Champion, SPL-Outdoor and 1st Place, Challenges

    RoboCup Leipzig, Germany
    2016

    3rd Place

    RoboCup European Open, Eindhoven, Netherlands
    2016

    DFG Grant

    Dortmund

    Funding for scientific equipment in collboration with the Sports Institute.

    2016

    Speed Record

    RoboCup Hefei

    Fastest documented walk of humanoid robot Nao till now.

    2015

    Representing City of Dortmund

    Dortmund

    RoboCup team represents city of Dortmund in the field of technology.

    2013-Now

    Most cited journal publication

    Dortmund

    "Observer-based dynamic walking control for biped robots" listed by Elsevier as one of the most cited publications of the journal since 2009. Please note that the site now lists later years.

    2013-Now

    2nd Place, Challenges and 3rd Place Drop-In Competition

    RoboCup Hefei, China
    2015

    3rd Place, Drop-In Competition

    RoboCup João Pessoa, Brazil
    2014

    3rd Place, Challenges

    RoboCup Eindhoven, Netherlands
    2013

    Vice World Champion

    RoboCup Istanbul, Turkey
    2011

    3rd Place

    RoboCup Graz, Austria
    2009

    Publications

    Journal

    Observer-based biped walking control, a sensor fusion approach

    Oliver Urbann, Stefan Tasse, Autonomous Robots 35.1 (2013) pp. 37–49. Springer US, 2013

    Observer-based dynamic walking control for biped robots

    Stefan Czarnetzki, Sören Kerner, Oliver Urbann, Robotics and Autonomous Systems 57.8 (2009) pp. 839–845. North-Holland Publishing Co., 2009

    Bipedal Walking with Dynamic Balance that Involves Three-Dimensional Upper Body Motion

    Jing Liu, Oliver Urbann, Robotics and Autonomous Systems 77 (2016) pp. 39-54. North-Holland Publishing Co., 2016

    Conference, Workshop, Symposium

    An Alternative Derivation of the Linear Inverted Pendulum Model

    Oliver Urbann, Ingmar Schwarz, Workshop on Robust Perception, Planning, and Control for Legged Robot Locomotion in Challenging Domains in conjunction with the 2017 IEEE International Conference on Robotics and Automation (ICRA), 2017

    Playing Soccer Outdoors with Humanoid Robots

    Matthias Hofmann, Arne Moos, Fabian Rensen, Ingmar Schwarz, Oliver Urbann, Proceedings of the 11th Workshop on Humanoid Soccer Robots in conjunction with the 2016 IEEE-RAS International Conference on Humanoid Robots, 2016

    Edge Detection Using Streaming SIMD Extensions on Low Cost Robotic Platforms

    Matthias Hofmann, Fabian Rensen, Ingmar Schwarz, Oliver Urbann, Proceedings of the 11th Workshop on Humanoid Soccer Robots in conjunction with the 2016 IEEE-RAS International Conference on Humanoid Robots, 2016

    A Fall Prediction System for Humanoid Robots Using a Multi-Layer Perceptron

    Matthias Hofmann, Ingmar Schwarz, Oliver Urbann, Florian Ziegler, Proceedings of the 10th Workshop on Cognitive Robotics in conjunction with the IEEE/RSJ International Con- ference on Intelligent Robots and Systems (IROS 2016), 2016

    A Reactive Stepping Algorithm Based on Preview Controller with Observer for Biped Robots

    Oliver Urbann, Matthias Hofmann, Intelligent Robots and Systems (IROS), 2016 IEEE/RSJ International Conference on, 2016, pp. 5324–5331

    Boundedness Approach to Gait Planning for the Flexible Linear Inverted Pendulum Model

    Leonardo Lanari, Oliver Urbann, Seth Hutchinson, Ingmar Schwarz, RoboCup 2016: Robot World Cup XX, 2017, to appear

    Analysis and Compensation of Biped Walking Disturbances Caused by Model Abstractions

    Oliver Urbann, Jan Hendrik Berlin, Matthias Hofmann, Ingmar Schwarz, Proceedings of the 10th Workshop on Humanoid Soccer Robots in conjunction with the 2015 IEEE-RAS International Conference on Humanoid Robots, 2015, Seoul, Korea

    Flexible Linear Inverted Pendulum Model for Cost-Effective Biped Robots

    Oliver Urbann, Ingmar Schwarz, Matthias Hofmann, Humanoid Robots (Humanoids), 2015 15th IEEE-RAS International Conference on, 2015, pp. 128–131. Downloadable source code.

    Modification of Foot Placement for Balancing Using a Preview Controller Based Humanoid Walking Algorithm

    Oliver Urbann, Matthias Hofmann, RoboCup 2013: Robot World Cup XVII, 2014, pp. 420–431

    Rigid and Soft Body Simulation Featuring Realistic Walk Behaviour

    Oliver Urbann, Sören Kerner, Stefan Tasse, RoboCup 2011: Robot Soccer World Cup XV, 2012, pp. 126–136. Downloadable source code.

    Applying Dynamic Walking Control for Biped Robots

    Stefan Czarnetzki, Sören Kerner, Oliver Urbann, RoboCup 2009: Robot Soccer World Cup XIII, 2010, pp. 69–80

    A Robust And Calibration-Free Vision System for Humanoid Soccer Robots

    Ingmar Schwarz, Matthias Hofmann, Oliver Urbann, Stefan Tasse RoboCup 2015: Robot World Cup XIX, 2016, to appear

    Stable walking of a bipedal humanoid robot involving three-dimensional upper body motion

    Jing Liu, U. Schwiegelshohn, O. Urbann, Humanoid Robots (Humanoids), 2014 14th IEEE-RAS International Conference on, 2014, pp. 80–85

    Walking pattern generation involving 3D waist motion for a biped humanoid robot

    Jing Liu, O. Urbann, Control Automation Robotics Vision (ICARCV), 2014 13th International Conference on, 2014, pp. 462–467

    On Sensor Model Design Choices for Humanoid Robot Localization

    Stefan Tasse, Matthias Hofmann, Oliver Urbann, RoboCup 2012: Robot Soccer World Cup XVI, 2013, pp. 380–390

    SLAM in the Dynamic Context of Robot Soccer Games

    Stefan Tasse, Matthias Hofmann, Oliver Urbann, RoboCup 2012: Robot Soccer World Cup XVI, 2013, pp. 368–379

    Multi Body Kalman Filtering with Articulation Constraints for Humanoid Robot Pose and Motion Estimation

    Daniel Hauschildt, Sören Kerner, Stefan Tasse, Oliver Urbann, RoboCup 2011: Robot Soccer World Cup XV, 2012, pp. 415–426

    Efficient Multi-hypotheses Unscented Kalman Filtering for Robust Localization

    Gregor Jochmann, Sören Kerner, Stefan Tasse, Oliver Urbann, RoboCup 2011: Robot Soccer World Cup XV, 2012, pp. 222–233

    Skills

    Biped Walking

    Control Theory

    Sensor Fusion

    Physics Simulation

    Artificial Intelligence

    Path Planning

    C/C++

    C#

    Python

    Assembler

    Java

    Contact

    Please feel free to ask me any question related to my research topics. I will try my best to help you. You can also have a look at my various other social network accounts.