Department of Computer Science Technische Universitaet Muenchen
Home People Teaching Research Publications Info Sitemap
About | Research | Publications | Announcements

Ongoing research projects
3D Semantic Mapping
http://www9.cs.tum.edu/people/rusu/research/semantic_mapping
In my research work I investigate the usage of point cloud based representation and reasoning techniques for building accurate, meaningful maps for mobile robots. While 3D mapping is not a new topic, maps so far have been used solely for the purpose of navigation and localization. Adding true semantic information to a map opens new horizons and makes its usage possible in more complex and comprehensive applications, like robot manipulation for example. Also, even though navigation based on 2D laser maps is considered a solved topic, the complexity of our world can seldomly be simplified in such a manner, and usually needs 3D representations which can deal with dynamic changes. Altogether, this is considered a very active research topic. My work lies at the intersection of several research areas such as: computer graphics, machine learning, probabilistic sensor perception and fusion, statistics and of course, robotics.

Within the scope of our research, we created methods for the acquisition of point cloud data (PCD) models, outlier removal, feature estimation, point cloud resampling, registration, model fitting, segmentation, and many more. Most of these methods have already been described in the publications listed below, with the rest following shortly.

One of the main deployment scenarios for the above mentioned methods is the Assistive Kitchen. However, everything was carefully crafted with generality in mind, therefore they have been also successfully applied to outdoor urban, aerial, and underwater datasets.


Relevant publications
  • Radu Bogdan Rusu, Wim Meeussen, Sachin Chitta, and Michael Beetz. Laser-based Perception for Door and Handle Identification. In Proceedings of the International Conference on Advanced Robotics (ICAR), Munich, Germany, June 22-26 2009. Best paper award. [PDF]
  • Zoltan Csaba Marton, Lucian Goron, Radu Bogdan Rusu, and Michael Beetz. Reconstruction and Verification of 3D Object Models for Grasping. In Proceedings of the 14th International Symposium on Robotics Research (ISRR), Lucernce, Switzerland, August 31 - September 3 2009.
  • Zoltan Csaba Marton, Radu Bogdan Rusu, Dominik Jain, Ulrich Klank, and Michael Beetz. Probabilistic Categorization of Kitchen Objects in Table Settings with a Composite Sensor. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), St. Louis, MO, USA, October 11-15 2009.
  • Radu Bogdan Rusu, Andreas Holzbach, Nico Blodow, and Michael Beetz. Fast Geometric Point Labeling using Conditional Random Fields. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), St. Louis, MO, USA, October 11-15 2009. [PDF]
  • Radu Bogdan Rusu, Ioan Alexandru Sucan, Brian Gerkey, Sachin Chitta, Michael Beetz, and Lydia E. Kavraki. Real-time Perception-Guided Motion Planning for a Personal Robot. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), St. Louis, MO, USA, October 11-15 2009. [PDF]
  • Radu Bogdan Rusu, Zoltan Csaba Marton, Nico Blodow, Andreas Holzbach, and Michael Beetz. Model-based and Learned Semantic Object Labeling in 3D Point Cloud Maps of Kitchen Environments. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), St. Louis, MO, USA, October 11-15 2009.
  • Radu Bogdan Rusu, Nico Blodow, Zoltan Csaba Marton, and Michael Beetz. Close-range Scene Segmentation and Reconstruction of 3D Point Cloud Maps for Mobile Manipulation in Human Environments. In Proceedings of the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), St. Louis, MO, USA, October 11-15 2009. [PDF]
  • Zoltan Csaba Marton, Radu Bogdan Rusu, and Michael Beetz. On Fast Surface Reconstruction Methods for Large and Noisy Datasets. In Proceedings of the International Conference on Robotics and Automation (ICRA), Kobe, Japan, May 12-17, 2009. [PDF]
  • Radu Bogdan Rusu, Nico Blodow, and Michael Beetz. Fast Point Feature Histograms (FPFH) for 3D Registration. In Proceedings of the International Conference on Robotics and Automation (ICRA), Kobe, Japan, May 12-17, 2009. [PDF]
  • Radu Bogdan Rusu, Zoltan Csaba Marton, Nico Blodow, and Michael Beetz. Learning Informative Point Classes for the Acquisition of Object Model Maps. In Proceedings of the 10th International Conference on Control, Automation, Robotics and Vision (ICARCV), Hanoi, Vietnam, December 17-20, 2008. [PDF]
  • Radu Bogdan Rusu, Zoltan Csaba Marton, Nico Blodow, Mihai Emanuel Dolha, and Michael Beetz. Functional Object Mapping of Kitchen Environments. In Proceedings of the 21st IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Nice, France, September 22-26, 2008. [PDF]
  • Radu Bogdan Rusu, Nico Blodow, Zoltan Csaba Marton, and Michael Beetz. Aligning Point Cloud Views using Persistent Feature Histograms. In Proceedings of the 21st IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Nice, France, September 22-26, 2008. [PDF]
  • Radu Bogdan Rusu, Zoltan Csaba Marton, Nico Blodow, Mihai Dolha, and Michael Beetz. Towards 3D Point Cloud Based Object Maps for Household Environments. In Robotics and Autonomous Systems Journal (Special Issue on Semantic Knowledge). [PDF]
  • Radu Bogdan Rusu, Zoltan Csaba Marton, Nico Blodow, and Michael Beetz. Persistent Point Feature Histograms for 3D Point Clouds. In Proceedings of the 10th International Conference on Intelligent Autonomous Systems (IAS-10), Baden-Baden, Germany, 2008. [PDF]
  • Radu Bogdan Rusu, Nico Blodow, Zoltan-Csaba Marton, Alina Soos, and Michael Beetz. Towards 3D Object Maps for Autonomous Household Robots. In Proceedings of the 20th IEEE International Conference on Intelligent Robots and Systems (IROS), San Diego, CA, USA, Oct 29 - 2 Nov., 2007. [PDF]

Player/Stage and its faithful companion, Javaclient(2)
http://playerstage.sourceforge.net/, http://java-player.sourceforge.net/
Player/Stage needs no introduction. If you work with robots and you never heard of it..., hmmm, let's just say that it's time to visit its SF web page ASAP. To make it short, Player/Stage is *the* de facto system for robotics (and ubiquitous computing) projects.

I am a Player/Stage developer. Some of my code has already been committed to the official repository, while other things are available on my personal Player/Stage page .

Javaclient allows development of applications for Player/Stage using the elegancy and power of the Java programming language. The client implements all major devices described in Player manual.

I am the main developer of the Javaclient software package. The project is sustained by a large international comunity of users and is being co-maintained along with Maxim Batalin @ UCLA.

Relevant publications
  • Radu Bogdan Rusu, Brian Gerkey, and Michael Beetz. Robots in the kitchen: Exploiting ubiquitous sensing and actuation. In Robotics and Autonomous Systems Journal (Special Issue on Network Robot Systems), 2008. [PDF]
  • Radu Bogdan Rusu, Alexis Maldonado, Michael Beetz and Brian Gerkey. Extending Player/Stage/Gazebo towards Cognitive Robots Acting in Ubiquitous Sensor-equipped Environments. In IEEE International Conference on Robotics and Automation (ICRA) Workshop for Network Robot System, 2007, April 14, Rome, Italy, 2007. [PDF]
  • Radu Bogdan Rusu, Alexis Maldonado, Michael Beetz, Matthias Kranz, Lorenz Mösenlechner, Paul Holleis, and Albrecht Schmidt. Player/Stage as Middleware for Ubiquitous Computing. In Proceedings of the 8th Annual Conference on Ubiquitous Computing (Ubicomp 2006), Orange County California, September 17-21, 2006.
  • Matthias Kranz, Radu Bogdan Rusu, Alexis Maldonado, Michael Beetz, and Albrecht Schmidt. A Player/Stage System for Context-Aware Intelligent Environments. In Proceedings of UbiSys'06, System Support for Ubiquitous Computing Workshop, at the 8th Annual Conference on Ubiquitous Computing (Ubicomp 2006), Orange County California, September 17-21, 2006.
  • Radu Bogdan Rusu, Radu Robotin, Gheorghe Lazea, and Cosmin Marcu. Towards Open Architectures for Mobile Robots: ZeeRO. In Proceedings of 2006 IEEE-TTTC International Conference on Automation, Quality & Testing, Robotics, AQTR 2006, May 25-28 Cluj-Napoca, Romania, 2006. [PDF]

Past/finished research projects
Context-Aware Intelligent Environments
http://awarekitchen.cs.tum.edu/
Intelligent sensor equipped environments can be of much greater help if they are capable of recognizing the actions and activities of their users, and inferring their intentions.

Understanding human activities and characterizing them into expressive and detailed activity models is one of they key issues of today's current pervasive computing systems. If such a system could recognize and understand automatically its user's behavior, it could interact in a more efficient and friendly manner. Unfortunately, the current model construction techniques are based on supervised learning and require specifications from their human counterparts, such as labeling the acquired sensor data. Our vision is to build technical cognitive systems that create and use models in a straightforward manner, by combining already existing online information with the system's context history.

We are experimenting with fusion of sensor data streams through the usage of wireless sensor networks and Player/Stage, for generating motion blueprints which can be used as a starting point in building and learning complex robotic arm movements. Our tests are being conducted on Powercube 6-DOF arms, mounted on a B21 mobile robot.
Relevant publications
  • Michael Beetz, Freek Stulp, Bernd Radig, Jan Bandouch, Nico Blodow, Mihai Emanuel Dolha, Andreas Fedrizzi, Dominik Jain, Uli Klank, Ingo Kresse, Alexis Maldonado, Zoltan-Csaba Marton, Lorenz Moesenlechner, Federico Ruiz, Radu Bogdan Rusu, and Moritz Tenorth Cognition, Control and Learning for Everyday Manipulation Tasks in Human Environments. In IEEE 17th International Symposium on Robot and Human Interactive Communication (RO-MAN), Muenchen, Germany, 2008.
  • Radu Bogdan Rusu, Jan Bandouch, Zoltan Csaba Marton, Nico Blodow, and Michael Beetz. Action Recognition in Intelligent Environments using Point Cloud Features Extracted from Silhouette Sequences. In IEEE 17th International Symposium on Robot and Human Interactive Communication (RO-MAN), Muenchen, Germany, 2008. [PDF]
  • Radu Bogdan Rusu, Brian Gerkey, and Michael Beetz. Robots in the kitchen: Exploiting ubiquitous sensing and actuation. In Robotics and Autonomous Systems Journal (Special Issue on Network Robot Systems), 2008. [PDF]
  • Michael Beetz, Jan Bandouch, Alexandra Kirsch, Alexis Maldonado, Armin Müller, and Radu Bogdan Rusu. The Assistive Kitchen - A Demonstration Scenario for Cognitive Technical Systems. In Proceedings of the 4th COE Workshop on Human Adaptive Mechatronics (HAM), 2007. [PDF]
  • Radu Bogdan Rusu, Alexis Maldonado, Michael Beetz and Brian Gerkey. Extending Player/Stage/Gazebo towards Cognitive Robots Acting in Ubiquitous Sensor-equipped Environments. In IEEE International Conference on Robotics and Automation (ICRA) Workshop for Network Robot System, 2007, April 14, Rome, Italy, 2007. [PDF]
  • Radu Bogdan Rusu, Alexis Maldonado, Michael Beetz, Matthias Kranz, Lorenz Mösenlechner, Paul Holleis, and Albrecht Schmidt. Player/Stage as Middleware for Ubiquitous Computing. In Proceedings of the 8th Annual Conference on Ubiquitous Computing (Ubicomp 2006), Orange County California, September 17-21, 2006.
  • Matthias Kranz, Radu Bogdan Rusu, Alexis Maldonado, Michael Beetz, and Albrecht Schmidt. A Player/Stage System for Context-Aware Intelligent Environments. In Proceedings of UbiSys'06, System Support for Ubiquitous Computing Workshop, at the 8th Annual Conference on Ubiquitous Computing (Ubicomp 2006), Orange County California, September 17-21, 2006.

Leaving Flatland
http://www.ai.sri.com/project/Leaving Flatland
The next advance for ground-based robots will be the development of perception, planning, and control systems that will enable UGVs to autonomously navigate in the same environments, and at the same speeds, as their human counterparts. This autonomous capability is a necessary prerequisite for the realization of the potential of UGVs as tactical multipliers.

Leaving Flatland is a joint effort between SRI (Stanford Research Institute) International, Stanford University and Boston Dynamics/MIT. The goal of the project is to enable small (e.g., man-packable) UGVs to move alongside their human counterparts through virtually any environment, whether indoor or outdoor.

I am working together with Aravind Sundaresan, Benoit Morisset, and Motilal Agrawal (alumnus: Brian Gerkey, and Kurt Konolige) on the 3d perception part (VO/VSLAM, point clouds, semantic maps).

In support of this new program, we propose an exploratory project to study the key challenges in closing the loop on autonomous perception and action in challenging terrain. The primary objective of the proposed project is to demonstrate the first robot to autonomously traverse challenging 3-D terrain with onboard sensing and online processing. The robot's perception, planning, and execution capabilities will operate in unmodified environments, both indoor and outdoor.
Relevant publications
  • Radu Bogdan Rusu, Aravind Sundaresan, Benoit Morisset, Kris Hauser, Motilal Agrawal, Jean-Claude Latombe, and Michael Beetz. Leaving Flatland: Efficient Real-Time 3D Navigation. In International Journal of Field Robotics (Special Issue on 3D Mapping), 2009.
  • Benoit Morisset, Radu Bogdan Rusu, Aravind Sundaresan, Kris Hauser, Motilal Agrawal, Jean-Claude Latombe, and Michael Beetz. Leaving Flatland: Toward Real-Time 3D Navigation. In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), Kobe, Japan, May 12-17, 2009. [PDF]
  • Radu Bogdan Rusu, Aravind Sundaresan, Benoit Morisset, Motilal Agrawal, Michael Beetz, and Kurt Konolige. Realtime Extended 3D Reconstruction from Stereo for Navigation. In Proceedings of the 21st IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) Workshop on 3D Mapping, Nice, France, September 26, 2008. Invited paper. [PDF]
  • Radu Bogdan Rusu, Aravind Sundaresan, Benoit Morisset, Motilal Agrawal, and Michael Beetz. Leaving Flatland: Realtime 3D Stereo Semantic Reconstruction. In Proceedings of the International Conference on Intelligent Robotics and Applications (ICIRA) 2008, October 15-17, Wuhan, China. [PDF]

COGITO - Plan-based Control of Robotic Agents
http://cogito.cs.tum.edu/
A key challenge for the next generation of autonomous robots is the reliable and efficient accomplishment of prolonged, complex, and dynamically changing tasks in the real world.

One of the most promising approaches to realizing these capabilities is the plan-based approach to robot control. In the plan-based approach, robots produce control actions by generating, maintaining, and executing plans that are tailored for the robots' respective tasks. Plans are robot control programs that a robot can not only execute but also reason about and manipulate. Thus a plan-based controller is able to manage and adapt the robot's intended course of action --- the plan --- while executing it and can thereby better achieve complex and changing goals. The use of plans enables these robots to flexibly interleave complex and interacting tasks, exploit opportunities, quickly plan their courses of action, and, if necessary, revise their intended activities.

One of the grand visions in the area of plan-based robot control is the realization of general autonomous robot control programs that can adapt themselves to the environments they are to operate in and to the distribution of complex tasks they are to perform. An instance of this grand vision is a pre-programmed household robot that knows how to clean a kitchen, how to operate a dishwasher, and so on. Being installed in a new environment it specializes its general plans to the specifics of the household and learns to manage the specific agenda of household chorus that is given to it. The robot also has to learn about the pitfalls of its tasks and its environment and avoid them through foresight. Our research field is still far away from realizing such competent robotic agents.

I am working on data mining/machine learning techniques as part of the COGITO project.

YARP (Yet Another Robot Platform)
http://yarp0.sourceforge.net/
I wrote some code for YARP, together with Alex at the VVV'06 summer camp in Ventimigla, Italy. You are free to do whatever you want with the code (although we are not really sure if it's actually useful for anything), provided that you agree to respect the GPL license that accompanies it (sometimes not explicitly).

Some of the pictures that I took during the camp are up on the Wiki, others on my personal web page.

For more information about YARP, contact Giorgio Metta or Paul Fitzpatrick. I am also keeping an open eye on the project, but due to time constraints, I am not contributing that much these days.
Downloads

  • ACE (Adaptive Communication Environment) enabled IRC chat server and client - download here;
  • SimpleShape tracker (Player driver), patched and modified for YARP - download here;
  • Player/Stage driver for providing YARP images - download here;
  • YARP interface, driver and network wrapper for a XSens MTx/MTi inertial measurement unit - download here.

AGILO - Semi-Automatic Acquisition of Visuomotoric Plans
http://agilo.cs.tum.edu/
Robotic soccer has become a standard 'real-world' testbed for autonomous multi robot control. In robot soccer (mid-size league) two teams of four autonomous robots --- one goal keeper and three field players --- play soccer against each other. The soccer field is four by nine meters big. The key characteristics of mid-size robot soccer are that the robots are completely autonomous. Consequently, all sensing and all action selection is done onboard of the individual robots. Skillful play requires our robots to recognize objects, such as other robots, field lines, and goals, and even entire game situations.

In the AGILO project we investigate how probabilistic visuomotoric autonomous robot controllers that are capable of learning can meet these challenges.

The AGILO robot controllers employ game state estimation and situated action selection based on automatically learned control mechanisms.

We are rebuilding the AGILO team using Roboteq controllers.

The RedBack mobile robot
I was part of the RedBack team in Rome, 2005 at the RoboCup Rescue Robotics camp. The goal was to create five 'tarantula'-like mobile robots, and use Player to control them with a joystick for rescue robotics teleoperations.

I was in charge of the Player driver and together with Alexis we hacked a copy of playernav which we then used to control the RedBack robot. We had a webcam and a Hokuyo laser sensor installed on the robot. To get Hokuyo URG04 working, we had to modify Toby's Player driver.

Please contact Raymond Sheh or Adam Jacoff and/or join the redback mailing list for additional information about the RedBack robot. Raymond also set up a Wiki page.

The members of the team were:
  • Masayuki Arai,
    Tokyo Institute of Technology
  • Michal Baczynski,
    Technical University of Lodz
  • Sharam Bahadori,
    University of Rome 'La Sapienza'
  • Timothy Jump,
    Benilde-St.Margaret's School
  • Alan Lytle,
    NIST
  • Alexis Maldonado (!),
    Techische Universität München
  • Luca Marchetti,
    University of Rome 'La Sapienza'
  • Olof Rensfelt,
    Uppsala University
  • Soraia Rocha,
    Universidade Nova De Lisboa
  • Radu Bogdan Rusu (!),
    Techische Universität München
  • Raymond Sheh,
    University of New South Wales
Downloads

You can download an alpha-stage Player driver here. Don't forget to issue autoreconf -i -s; ./configure; make in the RedBack directory after you unpack the archive.

I uploaded some pictures from the RoboCup event. They are available at www9.cs.tum.edu/people/rusu/events/Rome2005/. I have also mirrored Raymond's time-lapse movie (HQ) here.

Update (18.11.05): New movie added.

The ZeeRO mobile robot
http://www.robotux.info/zeero/
The project started in the Robotics Research Group at the Technical University of Cluj-Napoca, Romania. The goal of the project was to create a sub-1000 Euro mobile robot, which could be used as a standard research platform in the robotics lab.
The following things were taken into consideration:
  • the mobile robot must have a modular hardware and software architecture;
  • it must be programmable from a variety of programming languages (Java, C, Lisp, etc);
  • the sensor system must accommodate a variety of different types of sensors (research on sensor data fusion);
  • the robot should be fairly small, maximum 40cms in diameter;
  • the robot must posses wireless communication capabilities, since it is expected that more robots will be built in the near future, and research subjects such as the interaction of intelligent robotic agents;
  • it must have a layered, distributed central processing unit, capable of handling concurrent operations;
  • and finally, it should provide sufficient I/O ports and/or interfaces for future upgrades
I was the main architect, programmer and developer of the ZeeRO project. Check my publications web page or go directly to ZeeRO's web page for more information.
Downloads

You can download an alpha-stage Player driver here.

Notes:
  • you need to issue autoreconf -i -s; ./configure; make in the ZeeRO directory after you unpack the archive;
  • you need the Acroname libraries and headers installed /usr/local/include/acroname/aCommon;
  • the driver is provided for learning/illustrative purposes only. Look at the code and see if you can find anything useful in there. If you actually want to run it you will probably need a Gumstix connected to an Acroname Brainstem board as shown here.
Relevant publications
  • Gheorghe Lazea, Radu Bogdan Rusu, Radu Robotin, and Romulus Sime. The ZeeRO Mobile Robot - A Modular Architecture. In Proceedings of 15th International Workshop on Robotics in Alpe-Adria-Danube Region, RAAD 2006, June 15-17, Balatonfüred, Hungary, 2006. [PDF]
  • Radu Bogdan Rusu, Radu Robotin, Gheorghe Lazea, and Cosmin Marcu. Towards Open Architectures for Mobile Robots: ZeeRO. In Proceedings of 2006 IEEE-TTTC International Conference on Automation, Quality & Testing, Robotics, AQTR 2006, May 25-28 Cluj-Napoca, Romania, 2006. [PDF]
  • Radu Bogdan Rusu. Modern architectures for mobile robots: Javaclient and ZeeRO. Advanced Master's Thesis, Faculty of Automation and Computer Science, Technical University of Cluj-Napoca Romania, 2005. Supervisor: Prof.Dr.Ing. Gheorghe Lazea (head of the Robotics Research Group, ex-rector of the Technical University of Cluj-Napoca 1996-2004). 10/10 grade awarded. In Romanian. [PDF]

Robotux - a multiagent robot based security system
One of the key problems in the process of controlling the behavior of multirobot systems, is finding the proper algorithms which can generate a robust, coherent and adaptive behavior. If up to a certain (small) number of robots, the system can be easily controlled via client-server centralized architectures, the real-time capability of the communication system fails when the number of robots in the robot society grows above that threshold value. One solution to the above problem, is the use of multiagent systems. The integration process of a multirobot society with a multiagent system into a heterogeneous, scalable, redundant and flexible system becomes a necessary approach for solving some of the complex problems associated with the behavior of autonomous multirobot groups.

While doing my diploma work at TUCN, I made several studies of such a complex hybrid multirobot-multiagent system, the role that an agent plays in a distributed system, the communication mechanisms of the agent with the physical (or simulated) robot, as well as the communication protocols between the multiagent society members. My final contribution was the development of a multiagent multirobot system, by making use of the Java programming language, the ADK agent development toolkit and the Player/Stage platform. In order to prove the effectiveness of the resulted system, a real-life scenario which makes use of Robotux as an integrated solution for both security and 'housekeeping' in a given environment was presented. The application was tested on a small scale using ActivMedia Pioneer 2DX and 3AT robots, as well as on a large scale using the 2D Stage simulator.
The research project is in continuous development at the Robotics Research Group, Technical University of Cluj-Napoca, Romania.
Relevant publications
  • Radu Bogdan Rusu. Robotux - a multiagent robot security system. Master's Thesis, Faculty of Automation and Computer Science, Technical University of Cluj-Napoca, Romania, 2004. Supervisors: Prof.Dr.Ing. Liviu Miclea (head of the Automation and Industrial Informatics Department) and Prof.Dr.Ing. Gheorghe Lazea (head of the Robotics Research Group, ex-rector of the Technical University of Cluj-Napoca 1996-2004). 10/10 grade awarded. In Romanian. [PDF]
  • Radu Bogdan Rusu, Liviu Miclea, and Szilard Enyedi. Robotux - a Multiagent Robot Based Security System. In Proceedings of IEEE-TTTC Automation, Quality & Testing, Robotics International Conference 2004, Cluj-Napoca, Romania, May 13-15, 2004, AQTRJ, pp. 59-64, 2004. [PDF]

Computer Science IX | Technische Universität München