Publikationen

2016

  • Simulation based selection of actions for a humanoid soccer-robot
    H. Mellmann, B. Schlotter, and C. Blum
    Robocup 2016: robot soccer world cup xx , 2016 .
    [BibTeX] [Abstract] [PDF]
    This paper introduces a method for making fast decisions in a highly dynamic situation, based on forward simulation. This approach is inspired by the decision problem within the RoboCup domain. In this environment, selecting the right action is often a challenging task. The outcome of a particular action may depend on a wide variety of environmental factors, such as the robot’s position on the field or the location of obstacles. In addition, the perception is often heterogeneous, uncertain, and incomplete. In this context, we investigate forward simulation as a versatile and extensible yet simple mechanism for inference of decisions. The outcome of each possible action is simulated based on the estimated state of the situation. The simulation of a single action is split into a number of simple deterministic simulations — samples — based on the uncertainties of the estimated state and of the action model. Each of the samples is then evaluated separately, and the evaluations are combined and compared with those of other actions to inform the overall decision. This allows us to effectively combine heterogeneous perceptual data, calculate a stable decision, and reason about its uncertainty. This approach is implemented for the kick selection task in the RoboCup SPL environment and is actively used in competitions. We present analysis of real game data showing significant improvement over our previous methods.
    @InProceedings{RC-MellmannSchlotterBlum-16,
    author = {Heinrich Mellmann and Benjamin Schlotter and Christian Blum},
    title = {Simulation Based Selection of Actions for a Humanoid Soccer-Robot},
    booktitle = {RoboCup 2016: Robot Soccer World Cup XX},
    year = {2016},
    note = {to appear},
    abstract = {This paper introduces a method for making fast decisions in a highly
    dynamic situation, based on forward simulation. This approach is
    inspired by the decision problem within the RoboCup domain. In this
    environment, selecting the right action is often a challenging task.
    The outcome of a particular action may depend on a wide variety of
    environmental factors, such as the robot's position on the field
    or the location of obstacles. In addition, the perception is often
    heterogeneous, uncertain, and incomplete. In this context, we investigate
    forward simulation as a versatile and extensible yet simple mechanism
    for inference of decisions. The outcome of each possible action is
    simulated based on the estimated state of the situation. The simulation
    of a single action is split into a number of simple deterministic
    simulations -- samples -- based on the uncertainties of the estimated
    state and of the action model. Each of the samples is then evaluated
    separately, and the evaluations are combined and compared with those
    of other actions to inform the overall decision. This allows us to
    effectively combine heterogeneous perceptual data, calculate a stable
    decision, and reason about its uncertainty. This approach is implemented
    for the kick selection task in the RoboCup SPL environment and is
    actively used in competitions. We present analysis of real game data
    showing significant improvement over our previous methods.},
    keywords = {RoboCup, humanoid robots, internal simulation, decision making},
    url = {http://www.naoteamhumboldt.de/wp-content/papercite-data/pdf/2016-mellmannschlotterblum.pdf},
    }
  • Berlin United – Nao Team Humboldt: Team Report 2016
    H. Mellmann, B. Schlotter, S. Kaden, P. Strobel, T. Krause, and C. Ritter
    Humboldt-Universität zu Berlin, Adaptive Systems Group , 2016 .
    [BibTeX] [PDF]
    @TechReport{naoth-report16,
    author = {Heinrich Mellmann and Benjamin Schlotter and Steffen Kaden and Philipp Strobel and Thomas Krause and Claas-Norman Ritter},
    title = {{B}erlin {U}nited - {N}ao {T}eam {H}umboldt: {T}eam {R}eport 2016},
    institution = {Humboldt-Universität zu Berlin, Adaptive Systems Group},
    year = {2016},
    url = {http://www.naoteamhumboldt.de/wp-content/papercite-data/pdf/2016-Naoth-Report.pdf},
    }

2014

  • Multi-hypothesis goal modeling for a humanoid soccer robot
    M. M. Scheunemann and H. Mellmann
    Proceedings of the 9th workshop on humanoid soccer robots, 14th ieee-ras international conference on humanoid robots (humanoids), madrid, spain. , 2014 .
    [BibTeX] [PDF]
    @INPROCEEDINGS{HSR-ScheunemannMellmann-14,
    author = {Scheunemann, Marcus M and Mellmann, Heinrich},
    title = {Multi-Hypothesis Goal Modeling for a Humanoid Soccer Robot},
    booktitle = {Proceedings of the 9th Workshop on Humanoid Soccer Robots, 14th IEEE-RAS
    International Conference on Humanoid Robots (Humanoids), Madrid,
    Spain.},
    year = {2014},
    doi = {10.13140/2.1.4830.4963},
    url = {http://www.ais.uni-bonn.de/humanoidsoccer/ws14/papers/HSR14_Scheunemann.pdf}
    }

2013

  • Adaptive grasping for a small humanoid robot utilizing force- and electric current sensors
    H. Mellmann, M. Scheunemann, and O. Stadie
    Proceedings of the 22nd international workshop on concurrency, specification and programming (cs&p) , Warsaw, Poland , 2013 , pp. 283-293.
    [BibTeX] [Abstract] [PDF]
    The ability to grasp objects of different size and shape is one of the most important skills of a humanoid robot. Human grasping integrates a lot of different senses. In particular, the tactile sensing is very important for a stable grasping motion. When we lift a box without knowing what is inside, we do it carefully using our tactile and proprioceptive senses to estimate the weight and thus, the force necessary to hold and to lift this box. In this paper we present an adaptive controlling mechanism which enables a robot to grasp objects of different weights. Thereby, we only use the proprioceptive sensors like positions and electric current at the joints and force sensors at the end-effectors providing the robot with tactile feedback. We implemented and tested our approach on a humanoid robot.
    @InProceedings{CSP-MellmannScheunemannEtAl-13,
    author = {Heinrich Mellmann and Marcus Scheunemann and Oliver Stadie},
    title = {Adaptive Grasping for a Small Humanoid Robot Utilizing Force- and Electric Current Sensors},
    booktitle = {Proceedings of the 22nd International Workshop on Concurrency, Specification and Programming (CS\&P)},
    year = {2013},
    editor = {Marcin S. Szczuka and Ludwik Czaja and Magdalena Kacprzak},
    volume = {1032},
    series = {CEUR Workshop Proceedings},
    pages = {283-293},
    address = {Warsaw, Poland},
    publisher = {CEUR-WS.org},
    abstract = {The ability to grasp objects of different size and shape is one of
    the most important skills of a humanoid robot. Human grasping integrates
    a lot of different senses. In particular, the tactile sensing is
    very important for a stable grasping motion. When we lift a box without
    knowing what is inside, we do it carefully using our tactile and
    proprioceptive senses to estimate the weight and thus, the force
    necessary to hold and to lift this box. In this paper we present
    an adaptive controlling mechanism which enables a robot to grasp
    objects of different weights. Thereby, we only use the proprioceptive
    sensors like positions and electric current at the joints and force
    sensors at the end-effectors providing the robot with tactile feedback.
    We implemented and tested our approach on a humanoid robot.},
    bibsource = {DBLP, http://dblp.uni-trier.de},
    crossref = {DBLP:conf/csp/2013},
    ee = {http://ceur-ws.org/Vol-1032/paper-24.pdf},
    timestamp = {2013.11.03},
    }
  • Experiments with simulated humanoid robots
    H. Burkhard and M. Domańska
    2013 , pp. 27-38.
    [BibTeX] [Abstract] [PDF]
    Experimenting with real robots is limited by the available ressources: Complex hardware is costly, and it needs time and experience for setup and maintenance. Simulated robots can be used as alternative. Our RoboNewbie project is a basic framework for experimenting with simulated robots. It serves as an inspiration for beginners, and it provides room for many challenging experiments. The RoboNewbie agents run in the simulation environment of SimSpark RCSS, the official RoboCup 3D simulator, where the simulated robots are models of the humanoid Robot NAO of the French Company Aldebaran. Different example agents provide easily understandable interfaces to simulated sensors and effectors of the robot as well as simple control structures. The framework has been successfully used at different courses where the participants needed only few hours to understand the usage of the framework and to develop own agents for different tasks.
    @InProceedings{BurkhardDomanska-CSP2013,
    author = {Hans-Dieter Burkhard and Monika Domańska},
    title = {Experiments with Simulated Humanoid Robots},
    year = {2013},
    pages = {27--38},
    abstract = {Experimenting with real robots is limited by the available ressources: Complex hardware is costly, and it needs time and experience for setup and maintenance. Simulated robots can be used as alternative. Our RoboNewbie project is a basic framework for experimenting with simulated robots. It serves as an inspiration for beginners, and it provides room for many challenging experiments. The RoboNewbie agents run in the simulation environment of SimSpark RCSS, the official RoboCup 3D simulator, where the simulated robots are models of the humanoid Robot NAO of the French Company Aldebaran. Different example agents provide easily understandable interfaces to simulated sensors and effectors of the robot as well as simple control structures. The framework has been successfully used at different courses where the participants needed only few hours to understand the usage of the framework and to develop own agents for different tasks.},
    crossref = {CSP2013},
    url = {http://ceur-ws.org/Vol-1032/paper-03.pdf},
    }
  • Voronoi based strategic positioning for robot soccer
    S. Kaden, H. Mellmann, M. Scheunemann, and H. Burkhard
    Proceedings of the 22nd international workshop on concurrency, specification and programming (cs&p) , Warsaw, Poland , 2013 , pp. 271-282.
    [BibTeX] [Abstract] [PDF]
    Strategic positioning is a decisive part of the team play within a soccer game. In most solutions the positioning techniques are treated as a constituent of a complete team play strategy. In a comprehensive overview we discuss the team play and positioning methods used within RoboCup and extract the essential requirements for player positioning. In this work, we propose an approach for strategic positioning allowing for flexible formulation of arbitrary strategies. Based on the conditions of a specific strategy, the field is subdivided in regions by a Voronoi tessellation and each region is assigned a weight. Those weights influence the calculation of the optimal robot position as well as the path. A team play strategy can be expressed by the choice of the tessellation as well as the choice of the weights. This provides a powerful abstraction layer simplifying the design of the actual play strategy. We also present an implementation of an example strategy based on this approach and analyse the performance of our approach in simulation.
    @InProceedings{CSP-KadenMellmannEtAl-13,
    author = {Steffen Kaden and Heinrich Mellmann and Marcus Scheunemann and Hans-Dieter Burkhard},
    title = {Voronoi Based Strategic Positioning for Robot Soccer},
    booktitle = {Proceedings of the 22nd International Workshop on Concurrency, Specification and Programming (CS\&P)},
    year = {2013},
    editor = {Marcin S. Szczuka and Ludwik Czaja and Magdalena Kacprzak},
    volume = {1032},
    series = {CEUR Workshop Proceedings},
    pages = {271-282},
    address = {Warsaw, Poland},
    publisher = {CEUR-WS.org},
    abstract = {Strategic positioning is a decisive part of the team play within a
    soccer game. In most solutions the positioning techniques are treated
    as a constituent of a complete team play strategy. In a comprehensive
    overview we discuss the team play and positioning methods used within
    RoboCup and extract the essential requirements for player positioning.
    In this work, we propose an approach for strategic positioning allowing
    for flexible formulation of arbitrary strategies. Based on the conditions
    of a specific strategy, the field is subdivided in regions by a Voronoi
    tessellation and each region is assigned a weight. Those weights
    influence the calculation of the optimal robot position as well as
    the path. A team play strategy can be expressed by the choice of
    the tessellation as well as the choice of the weights. This provides
    a powerful abstraction layer simplifying the design of the actual
    play strategy. We also present an implementation of an example strategy
    based on this approach and analyse the performance of our approach
    in simulation.},
    bibsource = {DBLP, http://dblp.uni-trier.de},
    ee = {http://ceur-ws.org/Vol-1032/paper-23.pdf},
    timestamp = {2013.11.03},
    }

2011

  • Dynamic motion control: adaptive bimanual grasping for a humanoid robot
    H. Mellmann and G. Cotugno
    Fundamenta informaticae , vol. 112 , iss. 1 , pp. 89-101 , 2011 .
    [BibTeX] [Abstract]
    The ability to grasp objects of different size and shape is one of the most important skills of a humanoid robot. There are a lot of different approaches tackling this problem; however, there is no general solution. The complexity and the skill of a possible grasping motion depend hardly on a particular robot. In this paper we analyze the kinematic and sensory grasping abilities of the humanoid robot Nao. Its kinematic constraints and hand’s mechanical structure represent an interesting case of study due to lack of actuators for fingers and the limited computation power. After describing the platform and studying its capabilities, we propose some simple controllers and we present a benchmark based on some experimental data.
    @Article{FI-MellmannCotugno-11,
    author = {Heinrich Mellmann and Giuseppe Cotugno},
    title = {Dynamic Motion Control: Adaptive Bimanual Grasping for a Humanoid Robot},
    journal = {Fundamenta Informaticae},
    year = {2011},
    volume = {112},
    number = {1},
    pages = {89-101},
    abstract = {The ability to grasp objects of different size and shape is one of
    the most important skills of a humanoid robot. There are a lot of
    different approaches tackling this problem; however, there is no
    general solution. The complexity and the skill of a possible grasping
    motion depend hardly on a particular robot. In this paper we analyze
    the kinematic and sensory grasping abilities of the humanoid robot
    Nao. Its kinematic constraints and hand's mechanical structure represent
    an interesting case of study due to lack of actuators for fingers
    and the limited computation power. After describing the platform
    and studying its capabilities, we propose some simple controllers
    and we present a benchmark based on some experimental data.},
    timestamp = {2011.06.13},
    }
  • Local goal model for a humanoid soccer robot
    H. Mellmann and M. Scheunemann
    Proceedings of the workshop on concurrency, specification, and programming cs&p 2011 , Pułtusk, Poland , 2011 , pp. 353-360.
    [BibTeX] [PDF]
    @InProceedings{CSP-MellmannScheunemann-11,
    author = {Heinrich Mellmann and Marcus Scheunemann},
    title = {Local Goal Model for a Humanoid Soccer Robot},
    booktitle = {Proceedings of the Workshop on Concurrency, Specification, and Programming CS\&P 2011},
    year = {2011},
    editor = {Marcin Szczuka, Ludwik Czaja, Andrzej Skowron, Magdalena Kacprzak},
    pages = {353-360},
    address = {Pułtusk, Poland},
    month = {September},
    publisher = {Bia{\l}ystok University of Technology},
    timestamp = {2011.11.12},
    }
  • Qualitative measure for particle clustering and object classification in context of the world modeling for a mobile robot
    H. Mellmann and M. Scheunemann
    Proceedings of the workshop on concurrency, specification, and programming cs&p 2011 , Pułtusk, Poland , 2011 , pp. 353-360.
    [BibTeX] [PDF]
    @InProceedings{CSP-MellmannScheunemann-11-2,
    author = {Heinrich Mellmann and Marcus Scheunemann},
    title = {Qualitative Measure for Particle Clustering and Object Classification in Context of the World Modeling for a Mobile Robot},
    booktitle = {Proceedings of the Workshop on Concurrency, Specification, and Programming CS\&P 2011},
    year = {2011},
    editor = {Marcin Szczuka, Ludwik Czaja, Andrzej Skowron, Magdalena Kacprzak},
    pages = {353-360},
    address = {Pułtusk, Poland},
    month = {September},
    publisher = {Bia{\l}ystok University of Technology},
    timestamp = {2011.11.12},
    }

2010

  • Nao-team humboldt 2010
    H. Burkhard, F. Holzhauer, T. Krause, H. Mellmann, C. N. Ritter, O. Welter, and Y. Xu
    Humboldt Universität zu Berlin , 2010 .
    [BibTeX] [PDF]
    @TechReport{tdp10,
    author = {Hans-Dieter Burkhard and Florian Holzhauer and Thomas Krause and Heinrich Mellmann and Claas Norman Ritter and Oliver Welter and Yuan Xu},
    title = {NAO-Team Humboldt 2010},
    institution = {Humboldt Universität zu Berlin},
    year = {2010},
    owner = {Heinrich Mellmann},
    timestamp = {2010.02.03},
    url = {http://www.naoteamhumboldt.de/wp-content/papercite-data/pdf/2010-NaoTH-TeamDescription.pdf},
    }
  • Dynamic motion control: adaptive bimanual grasping for a humanoid robot
    G. Cotugno and H. Mellmann
    Workshop on concurrency, specification, and programming (cs&p) , Börnicke (near Berlin), Germany , 2010 .
    [BibTeX] [Abstract]
    In this paper we analyze the kinematic and sensory grasping abilities of the humanoid robot Nao. Its kinematic constraints and hand’s mechanical structure represent an interesting case of study due to lack of actuators for fingers and the limited computation power. After describing the platform and studying its capabilities, we propose some simple controllers and we present a benchmark based on some experimental data.
    @InProceedings{CSP-CotugnoMellmann-10,
    author = {Giuseppe Cotugno and Heinrich Mellmann},
    title = {Dynamic Motion Control: Adaptive Bimanual Grasping for a Humanoid Robot},
    booktitle = {Workshop on Concurrency, Specification, and Programming (CS\&P)},
    year = {2010},
    volume = {Volume 2},
    address = {Börnicke (near Berlin), Germany},
    month = {September},
    abstract = {In this paper we analyze the kinematic and sensory grasping abilities
    of the humanoid robot Nao. Its kinematic constraints and hand's mechanical
    structure represent an interesting case of study due to lack of actuators
    for fingers and the limited computation power. After describing the
    platform and studying its capabilities, we propose some simple controllers
    and we present a benchmark based on some experimental data.},
    timestamp = {2010.09.29},
    }
  • Ein anderes modell der welt: alternative methoden zur lokalisierung mobiler roboter
    H. Mellmann , Diploma Thesis , Humboldt Universität zu Berlin , 2010 , (in German).
    [BibTeX] [Abstract] [PDF]
    Many tasks of a mobile robot, e.g., navigation, require the knowledge of the positions of the objects in the surrounding environment. This task is especially challenging for the robots which perception is based on a directed visual system, e.g., a camera with a limited view angle. The incomplete and noisy sensor information leads to the uncertainty in the robots belief of the world. An appropriate model of the world may enable the robot to make plans and to realize complex behavior. The state of the art modeling methods use often only a small part of the available information. In particular the redundant information remain unused. In this work we investigate methods to exploit effectively the redundant information in order to get a better model of the world. In the first part we discuss a number of possibilities to use of specific properties of the objects to estimate the parameters of the camera matrix. In the second part we present a constraint based approach for the world modeling.
    @MASTERSTHESIS{DiplomaThesis-Mellmann-10,
    author = {Heinrich Mellmann},
    title = {Ein anderes Modell der Welt: Alternative Methoden zur Lokalisierung
    Mobiler Roboter},
    school = {Humboldt Universität zu Berlin},
    year = {2010},
    type = {Diploma Thesis},
    month = {April},
    note = {(in German)},
    abstract = {Many tasks of a mobile robot, e.g., navigation, require the knowledge
    of the positions of the objects in the surrounding environment. This
    task is especially challenging for the robots which perception is
    based on a directed visual system, e.g., a camera with a limited
    view angle. The incomplete and noisy sensor information leads to
    the uncertainty in the robots belief of the world. An appropriate
    model of the world may enable the robot to make plans and to realize
    complex behavior. The state of the art modeling methods use often
    only a small part of the available information. In particular the
    redundant information remain unused. In this work we investigate
    methods to exploit effectively the redundant information in order
    to get a better model of the world. In the first part we discuss
    a number of possibilities to use of specific properties of the objects
    to estimate the parameters of the camera matrix. In the second part
    we present a constraint based approach for the world modeling.},
    owner = {Heinrich Mellmann},
    timestamp = {2010.07.27},
    url = {http://www.heinrich-mellmann.de/content/publications/data/2011-05-13-diplomarbeit-mellmann-very-final.pdf}
    }
  • Adaptive motion control with visual feedback for a humanoid robot
    H. Mellmann and Y. Xu
    Ieee/rsj international conference on intelligent robots and systems (iros 2010) , 2010 .
    [BibTeX] [Abstract] [PDF]
    The performance of a soccer robot is highly dependent on its motion ability. The kicking motion is one of the most important motions in a soccer game. However, automatic, full body motion generation for humanoid robots presents a formidable computational challenge. At the current state the most common approaches of implementing this motion are based on key frame technique. Such solutions are inflexible, i.e., in order to adjust the aimed direction of the kick the robot has to walk around the ball. The adjustment costs a lot of time especially if some precise adjustments have to be done, e.g., for a penalty kick. In this paper we present an approach for adaptive control of the motions. We implemented our approach in order to solve the task of kicking the ball on a humanoid robot Nao. The approach was tested both in simulation and on a real robot.
    @InProceedings{IROS-MellmannXu-10,
    author = {Heinrich Mellmann and Yuan Xu},
    title = {Adaptive Motion Control with Visual Feedback for a Humanoid Robot},
    booktitle = {IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2010)},
    year = {2010},
    abstract = {The performance of a soccer robot is highly dependent on its motion
    ability. The kicking motion is one of the most important motions
    in a soccer game. However, automatic, full body motion generation
    for humanoid robots presents a formidable computational challenge.
    At the current state the most common approaches of implementing this
    motion are based on key frame technique. Such solutions are inflexible,
    i.e., in order to adjust the aimed direction of the kick the robot
    has to walk around the ball. The adjustment costs a lot of time especially
    if some precise adjustments have to be done, e.g., for a penalty
    kick. In this paper we present an approach for adaptive control of
    the motions. We implemented our approach in order to solve the task
    of kicking the ball on a humanoid robot Nao. The approach was tested
    both in simulation and on a real robot.},
    timestamp = {2010.07.27},
    }
  • Naoth software architecture for an autonomous agent
    H. Mellmann, Y. Xu, T. Krause, and F. Holzhauer
    Proceedings of the international workshop on standards and common platforms for robotics (scpr 2010) , Darmstadt , 2010 , pp. 316-327.
    [BibTeX] [Abstract] [PDF]
    An appropriate architecture (i.e., framework) is the base of each successful heterogeneous software project. It enables a group of developers to work at the same project and to organize their solutions. From this point of view, the artificial intelligence and/or robotics related research projects are usually more complicated, since the actual result of the project is often not clear. In particular, a strong organization of the software is necessary if the project is involved in education. Obviously, there is no perfect framework which could satisfy all the needs of the developers. In this paper we present a modular software architecture designed to implement an autonomous agent. In particular, it is used to develop software which is used simultaneously at several platforms (e.g., humanoid robot, simulated agent). One of the main aspects considered in our design is a strong code modularization which allows for re-usability, transparency and easily testing. Other important aspects are real-time applicability and simple usage. This paper presents the main concepts and the particular implementation of the important parts. We also provide a qualitative comparison with other existing robotics frameworks.
    @InProceedings{SCPR-MellmannXuEtAl-10,
    author = {Heinrich Mellmann and Yuan Xu and Thomas Krause and Florian Holzhauer},
    title = {NaoTH Software Architecture for an Autonomous Agent},
    booktitle = {Proceedings of the International Workshop on Standards and Common Platforms for Robotics (SCPR 2010)},
    year = {2010},
    pages = {316--327},
    address = {Darmstadt},
    month = {November},
    abstract = {An appropriate architecture (i.e., framework) is the base of each
    successful heterogeneous software project. It enables a group of
    developers to work at the same project and to organize their solutions.
    From this point of view, the artificial intelligence and/or robotics
    related research projects are usually more complicated, since the
    actual result of the project is often not clear. In particular, a
    strong organization of the software is necessary if the project is
    involved in education. Obviously, there is no perfect framework which
    could satisfy all the needs of the developers. In this paper we present
    a modular software architecture designed to implement an autonomous
    agent. In particular, it is used to develop software which is used
    simultaneously at several platforms (e.g., humanoid robot, simulated
    agent). One of the main aspects considered in our design is a strong
    code modularization which allows for re-usability, transparency and
    easily testing. Other important aspects are real-time applicability
    and simple usage. This paper presents the main concepts and the particular
    implementation of the important parts. We also provide a qualitative
    comparison with other existing robotics frameworks.},
    timestamp = {2010.11.21},
    }
  • Narrowing reality gap and validation: improving the simulator for humanoid soccer robot
    Y. Xu and H. Burkhard
    Concurrency, specification and programming cs&p’2010 , Helenenau, Germany , 2010 .
    [BibTeX]
    @InProceedings{xucsp2010,
    author = {Yuan Xu and Hans-Dieter Burkhard},
    title = {Narrowing Reality Gap and Validation: Improving the Simulator for Humanoid Soccer Robot},
    booktitle = {Concurrency, Specification and Programming CS\&P'2010},
    year = {2010},
    address = {Helenenau, Germany},
    month = {September},
    timestamp = {2014.01.27},
    }
  • Adaptive motion control: dynamic kick for a humanoid robot
    Y. Xu and H. Mellmann
    Proceedings of the 33rd annual german conference on artificial intelligence (ki 2010) , 2010 , pp. 392-399.
    [BibTeX] [Abstract] [PDF]
    Automatic, full body motion generation for humanoid robots presents a formidable computational challenge. The kicking motion is one of the most important motions in a soccer game. However, at the current state the most common approaches of implementing this motion are based on key frame technique. Such solutions are inflexible and cost a lot of time to adjust robot’s position. In this paper we present an approach for adaptive control of the motions. We implemented our approach in order to solve the task of kicking the ball on a humanoid robot Nao. The approach was tested both in simulation and on a real robot.
    @InProceedings{KI-XuMellmann-10,
    author = {Xu, Yuan and Mellmann, Heinrich},
    title = {Adaptive Motion Control: Dynamic Kick for a Humanoid Robot},
    booktitle = {Proceedings of the 33rd Annual German Conference on Artificial Intelligence (KI 2010)},
    year = {2010},
    editor = {Dillmann, Rüdiger and Beyerer, Jürgen and Hanebeck, Uwe and Schultz, Tanja},
    volume = {6359},
    series = {Lecture Notes in Computer Science},
    pages = {392--399},
    publisher = {Springer Berlin / Heidelberg},
    abstract = {Automatic, full body motion generation for humanoid robots presents
    a formidable computational challenge. The kicking motion is one of
    the most important motions in a soccer game. However, at the current
    state the most common approaches of implementing this motion are
    based on key frame technique. Such solutions are inflexible and cost
    a lot of time to adjust robot’s position. In this paper we present
    an approach for adaptive control of the motions. We implemented our
    approach in order to solve the task of kicking the ball on a humanoid
    robot Nao. The approach was tested both in simulation and on a real
    robot.},
    affiliation = {Institut für Informatik, LFG Künstliche Intelligenz, Humboldt-Universität zu Berlin, Germany},
    doi = {10.1007/978-3-642-16111-7_45},
    timestamp = {2011.01.04},
    url = {http://dx.doi.org/10.1007/978-3-642-16111-7_45},
    }
  • An approach to close the gap between simulation and real robots
    Y. Xu, H. Mellmann, and H. Burkhard
    2nd international conference on simulation, modeling and programming for autonomous robots (simpar) , 2010 , pp. 533-544.
    [BibTeX] [Abstract] [PDF]
    Numerous simulators have been developed over the years to assist robotics research in the development, testing, and evaluation. Nevertheless, there is still a big gap between the simulation and the reality. This makes it difficult to transfer methods and code. The 3D simulator | SimSpark is developed and used by a big community of AI researchers in RoboCup. But up to now there are only few applications to real robots. In this paper, we discuss the general possibilities how the SimSpark simulator can be used to support research in cognitive robotics and present applications on the humanoid robot Nao. As a result of our investigation we have developed a unified team playing both in Simulation League and Standard Platform League in RoboCup.
    @InProceedings{SIMPAR-XuMellmannEtAl-10,
    author = {Xu, Yuan and Mellmann, Heinrich and Burkhard, Hans-Dieter},
    title = {An Approach to Close the Gap between Simulation and Real Robots},
    booktitle = {2nd International Conference on Simulation, Modeling and Programming for Autonomous Robots (SIMPAR)},
    year = {2010},
    editor = {Ando, Noriaki and Balakirsky, Stephen and Hemker, Thomas and Reggiani, Monica and von Stryk, Oskar},
    series = {Lecture Notes in Computer Science},
    pages = {533--544},
    publisher = {Springer Berlin / Heidelberg},
    note = {(german)},
    abstract = {Numerous simulators have been developed over the years to assist robotics
    research in the development, testing, and evaluation. Nevertheless,
    there is still a big gap between the simulation and the reality.
    This makes it difficult to transfer methods and code. The 3D simulator
    | SimSpark is developed and used by a big community of AI researchers
    in RoboCup. But up to now there are only few applications to real
    robots. In this paper, we discuss the general possibilities how the
    SimSpark simulator can be used to support research in cognitive robotics
    and present applications on the humanoid robot Nao. As a result of
    our investigation we have developed a unified team playing both in
    Simulation League and Standard Platform League in RoboCup.},
    affiliation = {Institut für Informatik, LFG Künstliche Intelligenz, Humboldt-Universität zu Berlin, Rudower Chaussee 25, 12489 Berlin, Germany},
    doi = {10.1007/978-3-642-17319-6_48},
    timestamp = {2011.01.04},
    url = {http://dx.doi.org/10.1007/978-3-642-17319-6_48},
    }

2009

  • Nao-team humboldt 2009
    H. Burkhard, F. Holzhauer, T. Krause, H. Mellmann, C. N. Ritter, O. Welter, and Y. Xu
    Humboldt Universität zu Berlin , 2009 .
    [BibTeX] [PDF]
    @TechReport{tdp09,
    author = {Hans-Dieter Burkhard and Florian Holzhauer and Thomas Krause and Heinrich Mellmann and Claas Norman Ritter and Oliver Welter and Yuan Xu},
    title = {NAO-Team Humboldt 2009},
    institution = {Humboldt Universität zu Berlin},
    year = {2009},
    owner = {Heinrich Mellmann},
    timestamp = {2010.02.03},
    }
  • Constraint based world modeling for multi agent systems in dynamic environments
    D. Göhring , PhD Thesis , Humboldt University Berlin , 2009 , [Online: Stand 2010-05-23T15:08:02Z].
    [BibTeX] [Abstract] [PDF]
    % \selectlanguage{ngerman} Die mobile Robotik stellt ein sehr junges und komplexes Forschungsfelder unserer Zeit dar. Innerhalb der letzten Jahrzehnte wurde es Robotern möglich, sich innerhalb ihrer Umgebung zu bewegen, zu navigieren und mit ihrer Umwelt zu interagieren. Aufgrund der Tatsache, dass die Welt von Unsicherheit geprägt ist und ein Roboter immer nur partielle Information über sie erhalten kann, wurden probabilistische Navigationsverfahren entwickelt, mit denen sich Roboter lokalisieren und Objekte ihrer Umgebung modellieren können. Weiterhin wurden in letzter Zeit Verfahren untersucht, die die kooperative Exploration der Umgebung durch eine Gruppe von Robotern zum Ziel haben. In der vorliegenden Arbeit wird ein neuartiges Konzept, welches sich Perzeptrelationen für die kooperative Umweltmodellierung zu Nutze macht, vorgestellt und evaluiert. Einen zweiten Beitrag der Arbeit stellen constraintbasierte Lokalisierungstechniken dar, die es einem oder mehreren Robotern auf effiziente Art und Weise ermöglichen, sich zu lokalisieren und ihre Umwelt zu modellieren. % \selectlanguage{english} Mobile autonomous robotics is a very young and complex field of research. Only in recent decades have robots become able to explore, to move, navigate and to interact with their environment. Since the world is uncertain and since robots can only gain partial information about it, probabilistic navigation algorithms have become very popular whenever a robot has to localize itself or surrounding objects. Furthermore, cooperative exploration and localization approaches have become very relevant lately, as robots begin to act not just alone but in groups. Within this thesis a new approach using the concept of spatial percept-relations for cooperative environment modeling is presented and evaluated. As a second contribution, constraint based localization techniques will be introduced for having a robot or a group of robots efficiently localized and to model their environment.
    @PHDTHESIS{Goehring10Dissertation,
    author = {Daniel Göhring},
    title = {Constraint based world modeling for multi agent systems in dynamic
    environments},
    school = {Humboldt University Berlin},
    year = {2009},
    note = {[Online: Stand 2010-05-23T15:08:02Z]},
    abstract = {% \selectlanguage{ngerman} Die mobile Robotik stellt ein sehr junges
    und komplexes Forschungsfelder unserer Zeit dar. Innerhalb der letzten
    Jahrzehnte wurde es Robotern möglich, sich innerhalb ihrer Umgebung
    zu bewegen, zu navigieren und mit ihrer Umwelt zu interagieren. Aufgrund
    der Tatsache, dass die Welt von Unsicherheit geprägt ist und ein
    Roboter immer nur partielle Information über sie erhalten kann, wurden
    probabilistische Navigationsverfahren entwickelt, mit denen sich
    Roboter lokalisieren und Objekte ihrer Umgebung modellieren können.
    Weiterhin wurden in letzter Zeit Verfahren untersucht, die die kooperative
    Exploration der Umgebung durch eine Gruppe von Robotern zum Ziel
    haben. In der vorliegenden Arbeit wird ein neuartiges Konzept, welches
    sich Perzeptrelationen für die kooperative Umweltmodellierung zu
    Nutze macht, vorgestellt und evaluiert. Einen zweiten Beitrag der
    Arbeit stellen constraintbasierte Lokalisierungstechniken dar, die
    es einem oder mehreren Robotern auf effiziente Art und Weise ermöglichen,
    sich zu lokalisieren und ihre Umwelt zu modellieren. % \selectlanguage{english}
    Mobile autonomous robotics is a very young and complex field of research.
    Only in recent decades have robots become able to explore, to move,
    navigate and to interact with their environment. Since the world
    is uncertain and since robots can only gain partial information about
    it, probabilistic navigation algorithms have become very popular
    whenever a robot has to localize itself or surrounding objects. Furthermore,
    cooperative exploration and localization approaches have become very
    relevant lately, as robots begin to act not just alone but in groups.
    Within this thesis a new approach using the concept of spatial percept-relations
    for cooperative environment modeling is presented and evaluated.
    As a second contribution, constraint based localization techniques
    will be introduced for having a robot or a group of robots efficiently
    localized and to model their environment.},
    file = {:http\://edoc.hu-berlin.de/dissertationen/goehring-daniel-2009-11-10/PDF/goehring.pdf:PDF},
    keywords = {% \selectlanguage{ngerman}Künstliche Intelligenz, Robotik, Lokalisierungsverfahren,
    Constraints % \selectlanguage{english}Robotics, Artificial Intelligence,
    Localization Approaches, Constraints},
    owner = {gxy},
    publisher = { Humboldt-Universität zu Berlin, Mathematisch-Naturwissenschaftliche
    Fakultät II },
    timestamp = {2014.01.27},
    url = {http://edoc.hu-berlin.de/docviews/abstract.php?id=30348}
    }
  • Constraint based world modeling in mobile robotics
    D. Göhring, H. Mellmann, and H. Burkhard
    Proc. ieee international conference on robotics and automation (icra 2009) , 2009 , pp. 2538-2543.
    [BibTeX] [Abstract] [PDF]
    In this paper we present a novel approach using constraint based techniques for world modeling, i.e. self localization and object modeling. Within the last years, we have seen a reduction of landmarks such as beacons or colored goals within the RoboCup domain. Using other features as line information becomes more important. Using such sensor data is tricky, especially when the resulting position belief is stretched over a larger area. Constraints can overcome this limitations, as they have several advantages: they can represent large distributions and are easy to store and to communicate to other robots. Propagation of several constraints can be computationally cheap. Even high dimensional belief functions can be used. We will describe a sample implementation and show experimental results.
    @InProceedings{ICRA-GoehringMellmann-09,
    author = {Daniel Göhring and Heinrich Mellmann and Hans-Dieter Burkhard},
    title = {Constraint Based World Modeling in Mobile Robotics},
    booktitle = {Proc. IEEE International Conference on Robotics and Automation (ICRA 2009)},
    year = {2009},
    pages = {2538--2543},
    abstract = {In this paper we present a novel approach using constraint based techniques
    for world modeling, i.e. self localization and object modeling. Within
    the last years, we have seen a reduction of landmarks such as beacons
    or colored goals within the RoboCup domain. Using other features
    as line information becomes more important. Using such sensor data
    is tricky, especially when the resulting position belief is stretched
    over a larger area. Constraints can overcome this limitations, as
    they have several advantages: they can represent large distributions
    and are easy to store and to communicate to other robots. Propagation
    of several constraints can be computationally cheap. Even high dimensional
    belief functions can be used. We will describe a sample implementation
    and show experimental results.},
    doi = {10.1109/ROBOT.2009.5152208},
    file = {:ICRA-GoehringMellmann-09.pdf:PDF},
    keywords = {Autonomous Agents, Humanoid Robots, Localization},
    timestamp = {2009.05.14},
    }
  • Active landmark selection for vision-based self-localization
    H. Mellmann
    Proceedings of the workshop on concurrency, specification, and programming cs&p 2009 , Kraków-Przegorzaly, Poland , 2009 , pp. 398-405.
    [BibTeX] [Abstract] [PDF]
    The most of the vision based self-localization methods are using landmarks to estimate the position of the robot. The results of those methods depend highly on the precision of the perceptual information provided by the vision system. Depending on situation, some landmarks provide more certain measurements than others. We present a general criterion to predict the sensitivity of measurements concerning errors. In addition an algorithm is presented for automatic selection of optimal landmarks. This enables the robot to choose actively those landmarks that provide the highest certainty, which leads to better localization results. We demonstrate the performance and accuracy of the algorithm through a series of experiments.
    @InProceedings{CSP-Mellmann-09,
    author = {Heinrich Mellmann},
    title = {Active Landmark Selection for Vision-Based Self-Localization},
    booktitle = {Proceedings of the Workshop on Concurrency, Specification, and Programming CS\&P 2009},
    year = {2009},
    volume = {Volume 2},
    pages = {398--405},
    address = {Kraków-Przegorzaly, Poland},
    month = {September},
    abstract = {The most of the vision based self-localization methods are using landmarks
    to estimate the position of the robot. The results of those methods
    depend highly on the precision of the perceptual information provided
    by the vision system. Depending on situation, some landmarks provide
    more certain measurements than others. We present a general criterion
    to predict the sensitivity of measurements concerning errors. In
    addition an algorithm is presented for automatic selection of optimal
    landmarks. This enables the robot to choose actively those landmarks
    that provide the highest certainty, which leads to better localization
    results. We demonstrate the performance and accuracy of the algorithm
    through a series of experiments.},
    keywords = {RoboCup, localization, landmark selection, active vision, humanoid robots, Aibo},
    timestamp = {2009.10.04},
    url = {http://csp2009.mimuw.edu.pl/proc.php},
    }
  • Reinforcement learning in continuous state and action spaces
    V. Uc-Cetina , PhD Thesis , Humboldt-Universität zu Berlin , 2009 .
    [BibTeX]
    @PhdThesis{Uc-Cetina09thesis,
    author = {Vìctor Uc-Cetina},
    title = {Reinforcement Learning in Continuous State and Action Spaces},
    school = {Humboldt-Universität zu Berlin},
    year = {2009},
    timestamp = {2014.01.27},
    }

2008

  • Cooperative object localization using line-based percept communication
    D. Göhring
    in Robocup 2007: robot soccer world cup xi , U. Visser, F. Ribeiro, T. Ohashi, and F. Dellaert, Eds. , Berlin, Heidelberg: Springer Berlin Heidelberg , 2008 , pp. 53-64 .
    [BibTeX] [Abstract] [PDF]
    In this paper we present a novel approach to estimate the position of objects tracked by a team of robots. Moving objects are commonly modeled in an egocentric frame of reference, because this is sufficient for most robot tasks as following an object, and it is independent of the robots localization within its environment. But for multiple robots, to communicate and to cooperate the robots have to agree on an allocentric frame of reference. Instead of transforming egocentric models into allocentric ones by using self localization information, we will show how relations between different objects within the same camera image can be used as a basis for estimating an object’s position. The spacial relation of objects with respect to stationary objects yields several advantages: a) Errors in feature detections are correlated. The error of relative positions of objects within a single camera frame is comparably small. b) The information is independent of robot localization and odometry. c) Object relations can help to detect inconsistent sensor data. We present experimental evidence that shows how two non-localized robots are capable to infer the position of an object by communication on a RoboCup Four-Legged soccer field.
    @InBook{goehring2008,
    pages = {53--64},
    title = {Cooperative Object Localization Using Line-Based Percept Communication},
    publisher = {Springer Berlin Heidelberg},
    year = {2008},
    author = {G{\"o}hring, Daniel},
    editor = {Visser, Ubbo and Ribeiro, Fernando and Ohashi, Takeshi and Dellaert, Frank},
    address = {Berlin, Heidelberg},
    isbn = {978-3-540-68847-1},
    abstract = {In this paper we present a novel approach to estimate the position of objects tracked by a team of robots. Moving objects are commonly modeled in an egocentric frame of reference, because this is sufficient for most robot tasks as following an object, and it is independent of the robots localization within its environment. But for multiple robots, to communicate and to cooperate the robots have to agree on an allocentric frame of reference. Instead of transforming egocentric models into allocentric ones by using self localization information, we will show how relations between different objects within the same camera image can be used as a basis for estimating an object's position. The spacial relation of objects with respect to stationary objects yields several advantages: a) Errors in feature detections are correlated. The error of relative positions of objects within a single camera frame is comparably small. b) The information is independent of robot localization and odometry. c) Object relations can help to detect inconsistent sensor data. We present experimental evidence that shows how two non-localized robots are capable to infer the position of an object by communication on a RoboCup Four-Legged soccer field.},
    booktitle = {RoboCup 2007: Robot Soccer World Cup XI},
    doi = {10.1007/978-3-540-68847-1_5},
    url = {https://doi.org/10.1007/978-3-540-68847-1_5},
    }
  • Constraint based object state modeling
    D. Göhring, H. Mellmann, and H. Burkhard
    European robotics symposium 2008 (euros) , Prague, Chech Republic , 2008 , pp. 63-72.
    [BibTeX] [Abstract] [PDF]
    Modeling the environment is crucial for a mobile robot. Common approaches use Bayesian filters like particle filters, Kalman filters and their extended forms. We present an alternative and supplementing approach using constraint techniques based on spatial constraints between object positions. This yields several advantages: a) the agent can choose from a variety of belief functions, b) the computational complexity is decreased by efficient algorithms. The focus of the paper are constraint propagation techniques under the special requirements of navigation tasks.
    @InProceedings{EUROS-GoehringMellmann-08,
    author = {Daniel Göhring and Heinrich Mellmann and Hans-Dieter Burkhard},
    title = {Constraint Based Object State Modeling},
    booktitle = {European Robotics Symposium 2008 (EUROS)},
    year = {2008},
    editor = {Bruyninckx Herman and Preucil Libor and Kulich Miroslav},
    volume = {Volume 44/2008},
    series = {Springer Tracts in Advanced Robotics},
    pages = {63--72},
    address = {Prague, Chech Republic},
    publisher = {Springer Berlin / Heidelberg},
    abstract = {Modeling the environment is crucial for a mobile robot. Common approaches
    use Bayesian filters like particle filters, Kalman filters and their
    extended forms. We present an alternative and supplementing approach
    using constraint techniques based on spatial constraints between
    object positions. This yields several advantages: a) the agent can
    choose from a variety of belief functions, b) the computational complexity
    is decreased by efficient algorithms. The focus of the paper are
    constraint propagation techniques under the special requirements
    of navigation tasks.},
    doi = {10.1007/978-3-540-78317-6_7},
    timestamp = {Dienstag, 12. Februar 2008},
    url = {http://www.springerlink.com/content/th6218453434x817},
    }
  • Constraint based belief modeling
    D. Göhring, H. Mellmann, and H. Burkhard
    Robocup 2008: robot soccer world cup xii , 2008 .
    [BibTeX] [Abstract] [PDF]
    In this paper we present a novel approach using constraint based techniques for world modeling, i.e. self localization and object modeling. Within the last years, we have seen a reduction of landmarks as beacons, colored goals, within the RoboCup domain. Using other fea- tures as line information becomes more important. Using such sensor data is tricky, especially when the resulting position belief is stretched over a larger area. Constraints can overcome this limitations, as they have several advantages: They can represent large distributions and are easy to store and to communicate to other robots. Propagation of a several constraints can be computationally cheap. Even high dimensional belief functions can be used. We will describe a sample implementation and show experimental results.
    @InProceedings{RC-GoehringMellmann-08,
    author = {Daniel Göhring and Heinrich Mellmann and Hans-Dieter Burkhard},
    title = {Constraint Based Belief Modeling},
    booktitle = {RoboCup 2008: Robot Soccer World Cup XII},
    year = {2008},
    editor = {Luca Iocchi and Hitoshi Matsubara and Alfredo Weitzenfeld and Changjiu Zhou},
    series = {Lecture Notes in Artificial Intelligence},
    publisher = {Springer},
    abstract = {In this paper we present a novel approach using constraint based techniques
    for world modeling, i.e. self localization and object modeling. Within
    the last years, we have seen a reduction of landmarks as beacons,
    colored goals, within the RoboCup domain. Using other fea- tures
    as line information becomes more important. Using such sensor data
    is tricky, especially when the resulting position belief is stretched
    over a larger area. Constraints can overcome this limitations, as
    they have several advantages: They can represent large distributions
    and are easy to store and to communicate to other robots. Propagation
    of a several constraints can be computationally cheap. Even high
    dimensional belief functions can be used. We will describe a sample
    implementation and show experimental results.},
    timestamp = {2008.08.05},
    }
  • Constraint based localization on a humanoid robot
    D. Göhring, H. Mellmann, and H. Burkhard
    Proceedings of the workshop on concurrency, specification, and programming cs&p 2008 , 2008 .
    [BibTeX] [Abstract] [PDF]
    In this paper we will present an application for constraint based methods to self localize within the RoboCup domain. During a robotic soccer game, robots of a team need to know where they and their team mates are on the Field, therefore they need to localize themselves. For self localization, constraint based methods can be an e?ective alternative to classic Bayesian approaches as Kalman ?lters or Monte-Carlo methods. In this paper we will present, how constraint based techniques can be applied to a humanoid robot. Therefore we will implement constraint based methods in a humanoid robot "NAO" and see how the constraint based approach works within the Standard Platform League.
    @InProceedings{CSP-GoehringMellmannEtAl-08,
    author = {Daniel Göhring and Heinrich Mellmann and Hans-Dieter Burkhard},
    title = {Constraint Based Localization on a Humanoid Robot},
    booktitle = {Proceedings of the Workshop on Concurrency, Specification, and Programming CS\&P 2008},
    year = {2008},
    abstract = {In this paper we will present an application for constraint based
    methods to self localize within the RoboCup domain. During a robotic
    soccer game, robots of a team need to know where they and their team
    mates are on the Field, therefore they need to localize themselves.
    For self localization, constraint based methods can be an e?ective
    alternative to classic Bayesian approaches as Kalman ?lters or Monte-Carlo
    methods. In this paper we will present, how constraint based techniques
    can be applied to a humanoid robot. Therefore we will implement constraint
    based methods in a humanoid robot "NAO" and see how the constraint
    based approach works within the Standard Platform League.},
    timestamp = {2009.10.05},
    }
  • Constraint based world modeling
    D. Göhring, H. Mellmann, K. Gerasymova, and H. Burkhard
    Fundamenta informaticae , vol. Volume 85 , iss. 1-4 , pp. 123-137 , 2008 .
    [BibTeX] [Abstract] [PDF]
    Common approaches for robot navigation use Bayesian filters like particle filters, Kalman filters and their extended forms. We present an alternative and supplementing approach using constraint techniques based on spatial constraints between object positions. This yields several advantages. The robot can choose from a variety of belief functions, and the computational complexity is decreased by efficient algorithms. The paper investigates constraint propagation techniques under the special requirements of navigation tasks. Sensor data are noisy, but a lot of redundancies can be exploited to improve the quality of the result. We introduce two quality measures: The ambiguity measure for constraint sets defines the precision, while inconsistencies are measured by the inconsistency measure. The measures can be used for evaluating the available data and for computing best fitting hypothesis. A constraint propagation algorithm is presented.
    @Article{FI-GoehringMellmannGerasimova-08,
    author = {Daniel Göhring and Heinrich Mellmann and Kataryna Gerasymova and Hans-Dieter Burkhard},
    title = {Constraint Based World Modeling},
    journal = {Fundamenta Informaticae},
    year = {2008},
    volume = {Volume 85},
    number = {1-4},
    pages = {123-137},
    abstract = {Common approaches for robot navigation use Bayesian filters like particle
    filters, Kalman filters and their extended forms. We present an alternative
    and supplementing approach using constraint techniques based on spatial
    constraints between object positions. This yields several advantages.
    The robot can choose from a variety of belief functions, and the
    computational complexity is decreased by efficient algorithms. The
    paper investigates constraint propagation techniques under the special
    requirements of navigation tasks. Sensor data are noisy, but a lot
    of redundancies can be exploited to improve the quality of the result.
    We introduce two quality measures: The ambiguity measure for constraint
    sets defines the precision, while inconsistencies are measured by
    the inconsistency measure. The measures can be used for evaluating
    the available data and for computing best fitting hypothesis. A constraint
    propagation algorithm is presented.},
    file = {:http\://nbn-resolving.de/urn\:nbn\:de\:kobv\:11-10094810:PDF},
    timestamp = {2008.08.05},
    url = {http://iospress.metapress.com/content/2051310891588554/},
    }
  • Self-localization using odometry and horizontal bearings to landmarks
    M. Jüngel and M. Risler
    in Robocup 2007: robot soccer world cup xi , U. Visser, F. Ribeiro, T. Ohashi, and F. Dellaert, Eds. , Berlin, Heidelberg: Springer Berlin Heidelberg , 2008 , pp. 393-400 .
    [BibTeX] [Abstract] [PDF]
    On the way to the big goal – the game against the human world champion on a real soccer field – the configuration of the soccer fields in RoboCup has changed during the last years. There are two main modification trends: The fields get larger and the number of artificial landmarks around the fields decreases. The result is that a lot of the methods for self-localization developed during the last years do not work in the new scenarios without modifications. This holds especially for robots with a limited range of view as the probability for a robot to detect a landmark inside its viewing angle is significantly lower than on the old fields. On the other hand the robots have more space to play and do not collide as often as on the small fields. Thus the robots have a better idea of the courses they cover (odometry has higher reliability). This paper shows a method for self-localization that is based on bearings to horizontal landmarks and the knowledge about the robots movement between the observation of the features.
    @InBook{Jüngel2008,
    pages = {393--400},
    title = {Self-localization Using Odometry and Horizontal Bearings to Landmarks},
    publisher = {Springer Berlin Heidelberg},
    year = {2008},
    author = {J{\"u}ngel, Matthias and Risler, Max},
    editor = {Visser, Ubbo and Ribeiro, Fernando and Ohashi, Takeshi and Dellaert, Frank},
    address = {Berlin, Heidelberg},
    isbn = {978-3-540-68847-1},
    abstract = {On the way to the big goal - the game against the human world champion on a real soccer field - the configuration of the soccer fields in RoboCup has changed during the last years. There are two main modification trends: The fields get larger and the number of artificial landmarks around the fields decreases. The result is that a lot of the methods for self-localization developed during the last years do not work in the new scenarios without modifications. This holds especially for robots with a limited range of view as the probability for a robot to detect a landmark inside its viewing angle is significantly lower than on the old fields. On the other hand the robots have more space to play and do not collide as often as on the small fields. Thus the robots have a better idea of the courses they cover (odometry has higher reliability). This paper shows a method for self-localization that is based on bearings to horizontal landmarks and the knowledge about the robots movement between the observation of the features.},
    booktitle = {RoboCup 2007: Robot Soccer World Cup XI},
    doi = {10.1007/978-3-540-68847-1_40},
    url = {https://doi.org/10.1007/978-3-540-68847-1_40},
    }
  • Memory-based state-estimation
    M. Jüngel and H. Mellmann
    Fundamenta informaticae , vol. Volume 85 , iss. Number 1-4 , pp. 297-311 , 2008 .
    [BibTeX] [Abstract] [PDF]
    In this paper we introduce a state-estimation method that uses a short-term memory to calculate the current state. A common way to solve state estimation problems is to use implementations of the Bayesian algorithm like Kalman filters or particle filters. When implementing a Bayesian filter several problems can arise. One difficulty is to obtain error models for the sensors and for the state transitions. The other difficulty is to find an adequate compromise between the accuracy of the belief probability distribution and the computational costs that are needed to update it. In this paper we show how a short-term memory of perceptions and actions can be used to calculate the state. In contrast to the Bayesian filter, this method does not need an internal representation of the state which is updated by the sensor and motion information. It is shown that this is especially useful when information of sparse sensors (sensors with non-unique measurements with respect of the state) has to be integrated.
    @Article{FI-JuengelMellmann-08,
    author = {Matthias Jüngel and Heinrich Mellmann},
    title = {Memory-Based State-Estimation},
    journal = {Fundamenta Informaticae},
    year = {2008},
    volume = {Volume 85},
    number = {Number 1-4},
    pages = {297--311},
    abstract = {In this paper we introduce a state-estimation method that uses a short-term
    memory to calculate the current state. A common way to solve state
    estimation problems is to use implementations of the Bayesian algorithm
    like Kalman filters or particle filters. When implementing a Bayesian
    filter several problems can arise. One difficulty is to obtain error
    models for the sensors and for the state transitions. The other difficulty
    is to find an adequate compromise between the accuracy of the belief
    probability distribution and the computational costs that are needed
    to update it. In this paper we show how a short-term memory of perceptions
    and actions can be used to calculate the state. In contrast to the
    Bayesian filter, this method does not need an internal representation
    of the state which is updated by the sensor and motion information.
    It is shown that this is especially useful when information of sparse
    sensors (sensors with non-unique measurements with respect of the
    state) has to be integrated.},
    timestamp = {2008.08.05},
    url = {http://iospress.metapress.com/content/2051310891588554/},
    }
  • Using reference objects to improve vision-based bearing measurements
    H. Mellmann, M. Jüngel, and M. Spranger
    Proc. ieee/rsj international conference on intelligent robots and systems iros 2008 , Acropolis Convention Center, Nice, France , 2008 , pp. 3939-3945.
    [BibTeX] [Abstract] [PDF]
    Robots perceiving its environment using cameras usually need a good representation of how the camera is aligned to the body and how the camera is rotated relative to the ground. This is especially important for bearing-based distance measurement. In this paper we show how to use reference objects to improve vision-based distance measurements to objects of unknown size. Several methods for different kinds of reference objects are introduced. These are objects of known size (like a ball), objects extending over the horizon (like goals and beacons), and objects with known shape on the ground (like field lines). We give a detailed description how to determine the rotation of the robot’s camera relative to the ground, provide an error-estimation for all methods and describe the experiments we performed on an Aibo robot.
    @InProceedings{IROS-MellmannJuengelSpranger-08,
    author = {Heinrich Mellmann and Matthias Jüngel and Michael Spranger},
    title = {Using Reference Objects to Improve Vision-Based Bearing Measurements},
    booktitle = {Proc. IEEE/RSJ International Conference on Intelligent Robots and Systems IROS 2008},
    year = {2008},
    pages = {3939--3945},
    address = {Acropolis Convention Center, Nice, France},
    month = {22--26 Sept.},
    publisher = {IEEE},
    abstract = {Robots perceiving its environment using cameras usually need a good
    representation of how the camera is aligned to the body and how the
    camera is rotated relative to the ground. This is especially important
    for bearing-based distance measurement. In this paper we show how
    to use reference objects to improve vision-based distance measurements
    to objects of unknown size. Several methods for different kinds of
    reference objects are introduced. These are objects of known size
    (like a ball), objects extending over the horizon (like goals and
    beacons), and objects with known shape on the ground (like field
    lines). We give a detailed description how to determine the rotation
    of the robot's camera relative to the ground, provide an error-estimation
    for all methods and describe the experiments we performed on an Aibo
    robot.},
    doi = {10.1109/IROS.2008.4651128},
    keywords = {Localization, Computer Vision, Recognition},
    timestamp = {2008.08.05},
    }

2007

  • Exploiting past experience — case-based decision support for soccer agents
    R. Berger and G. Lämmel
    in Ki 2007: advances in artificial intelligence: 30th annual german conference on ai, ki 2007, osnabrück, germany, september 10-13, 2007. proceedings , J. Hertzberg, M. Beetz, and R. Englert, Eds. , Berlin, Heidelberg: Springer Berlin Heidelberg , 2007 , pp. 440-443 .
    [BibTeX] [Abstract] [PDF]
    Selecting and initiating an appropriate (possibly cooperative) behavior in a given context is one of the most important and difficult tasks for soccer playing robots or software agents. Of course, this applies to other complex robot environments as well.
    @InBook{Berger2007,
    pages = {440--443},
    title = {Exploiting Past Experience -- Case-Based Decision Support for Soccer Agents},
    publisher = {Springer Berlin Heidelberg},
    year = {2007},
    author = {Berger, Ralf and L{\"a}mmel, Gregor},
    editor = {Hertzberg, Joachim and Beetz, Michael and Englert, Roman},
    address = {Berlin, Heidelberg},
    isbn = {978-3-540-74565-5},
    abstract = {Selecting and initiating an appropriate (possibly cooperative) behavior in a given context is one of the most important and difficult tasks for soccer playing robots or software agents. Of course, this applies to other complex robot environments as well.},
    booktitle = {KI 2007: Advances in Artificial Intelligence: 30th Annual German Conference on AI, KI 2007, Osnabr{\"u}ck, Germany, September 10-13, 2007. Proceedings},
    doi = {10.1007/978-3-540-74565-5_35},
    url = {https://doi.org/10.1007/978-3-540-74565-5_35},
    }
  • Cases in robotic soccer
    H. Burkhard and R. Berger
    in Case-based reasoning research and development: 7th international conference on case-based reasoning, iccbr 2007 belfast, northern ireland, uk, august 13-16, 2007 proceedings , R. O. Weber and M. M. Richter, Eds. , Berlin, Heidelberg: Springer Berlin Heidelberg , 2007 , pp. 1-15 .
    [BibTeX] [Abstract] [PDF]
    Soccer playing robots are a well established test bed for the development of artificial intelligence for use in real environments. The challenges include perception, decision making and acting in a dynamic environment with only unreliable and partial information. Behaviors and skills for such environments must be optimized by experiences. Case Based Reasoning provides an excellent framework for learning as discussed in this paper.
    @InBook{Burkhard2007,
    pages = {1--15},
    title = {Cases in Robotic Soccer},
    publisher = {Springer Berlin Heidelberg},
    year = {2007},
    author = {Burkhard, Hans-Dieter and Berger, Ralf},
    editor = {Weber, Rosina O. and Richter, Michael M.},
    address = {Berlin, Heidelberg},
    isbn = {978-3-540-74141-1},
    abstract = {Soccer playing robots are a well established test bed for the development of artificial intelligence for use in real environments. The challenges include perception, decision making and acting in a dynamic environment with only unreliable and partial information. Behaviors and skills for such environments must be optimized by experiences. Case Based Reasoning provides an excellent framework for learning as discussed in this paper.},
    booktitle = {Case-Based Reasoning Research and Development: 7th International Conference on Case-Based Reasoning, ICCBR 2007 Belfast, Northern Ireland, UK, August 13-16, 2007 Proceedings},
    doi = {10.1007/978-3-540-74141-1_1},
    url = {https://doi.org/10.1007/978-3-540-74141-1_1},
    }
  • Cooperative world modeling in dynamic multi-robot environments
    D. Göhring and H. Burkhard
    Fundamenta informaticae , vol. 75 , iss. 1–4 , pp. 281-294 , 2007 .
    [BibTeX] [Abstract] [PDF]
    In this paper we describe how a group of agents can commonly estimate the position of objects. Furthermore we will show how these modeled object positions can be used for an improved self localization. Modeling of moving objects is commonly done by a single agent and in a robo- centric coordinate frame because this information is sufficient for most low level robot control and it is independent of the quality of the current robot localization. Especially when many robots coop- erate with each other in a partially observable environment they have to share and to communicate information. For multiple robots to cooperate and share information, though, they need to agree on a global, allocentric frame of reference. But when transforming the egocentric object model into a global one, it inherits the localization error of the robot in addition to the error associated with the egocentric model. We propose using the relation of objects detected in camera images to other objects in the same camera image as a basis for estimating the position of the object in a global coordinate system. The spacial relation of objects with respect to stationary objects (e.g., landmarks) offers several advantages: The information is independent of robot localization and odometry and it can easily be communicated. We present experimental evidence that shows how two robots are able to infer the position of an object within a global frame of reference, even though they are not localized themselves. We will also show, how to use this object information for self localization. A third aspect of this work will cope with the communication delay, therefore we will show how the Hidden Markov Model can be extended for distributed object tracking.
    @Article{goehring07relation-funamenta,
    author = {Daniel Göhring and Hans-Dieter Burkhard},
    title = {Cooperative World Modeling in Dynamic Multi-Robot Environments},
    journal = {Fundamenta Informaticae},
    year = {2007},
    volume = {75},
    number = {1--4},
    pages = {281--294},
    abstract = {In this paper we describe how a group of agents can commonly estimate the position of
    objects. Furthermore we will show how these modeled object positions can be used for an improved
    self localization. Modeling of moving objects is commonly done by a single agent and in a robo-
    centric coordinate frame because this information is sufficient for most low level robot control and
    it is independent of the quality of the current robot localization. Especially when many robots coop-
    erate with each other in a partially observable environment they have to share and to communicate
    information. For multiple robots to cooperate and share information, though, they need to agree on
    a global, allocentric frame of reference. But when transforming the egocentric object model into a
    global one, it inherits the localization error of the robot in addition to the error associated with the
    egocentric model. We propose using the relation of objects detected in camera images to other objects in the same
    camera image as a basis for estimating the position of the object in a global coordinate system.
    The spacial relation of objects with respect to stationary objects (e.g., landmarks) offers several
    advantages: The information is independent of robot localization and odometry and it can easily
    be communicated. We present experimental evidence that shows how two robots are able to infer
    the position of an object within a global frame of reference, even though they are not localized
    themselves. We will also show, how to use this object information for self localization. A third
    aspect of this work will cope with the communication delay, therefore we will show how the Hidden
    Markov Model can be extended for distributed object tracking.},
    publisher = {IOS Press},
    timestamp = {2007.05.04},
    }
  • Sensor modeling using visual object relation in multi robot object tracking
    D. Göhring and J. Hoffmann
    in Robocup 2006: robot soccer world cup x , G. Lakemeyer, E. Sklar, D. G. Sorrenti, and T. Takahashi, Eds. , Berlin, Heidelberg: Springer Berlin Heidelberg , 2007 , pp. 279-286 .
    [BibTeX] [Abstract] [PDF]
    In this paper we present a novel approach to estimating the position of objects tracked by a team of mobile robots. Modeling of moving objects is commonly done in a robo-centric coordinate frame because this information is sufficient for most low level robot control and it is independent of the quality of the current robot localization. For multiple robots to cooperate and share information, though, they need to agree on a global, allocentric frame of reference. When transforming the egocentric object model into a global one, it inherits the localization error of the robot in addition to the error associated with the egocentric model.
    @InBook{Goehring2007,
    pages = {279--286},
    title = {Sensor Modeling Using Visual Object Relation in Multi Robot Object Tracking},
    publisher = {Springer Berlin Heidelberg},
    year = {2007},
    author = {G{\"o}hring, Daniel and Hoffmann, Jan},
    editor = {Lakemeyer, Gerhard and Sklar, Elizabeth and Sorrenti, Domenico G. and Takahashi, Tomoichi},
    address = {Berlin, Heidelberg},
    isbn = {978-3-540-74024-7},
    abstract = {In this paper we present a novel approach to estimating the position of objects tracked by a team of mobile robots. Modeling of moving objects is commonly done in a robo-centric coordinate frame because this information is sufficient for most low level robot control and it is independent of the quality of the current robot localization. For multiple robots to cooperate and share information, though, they need to agree on a global, allocentric frame of reference. When transforming the egocentric object model into a global one, it inherits the localization error of the robot in addition to the error associated with the egocentric model.},
    booktitle = {RoboCup 2006: Robot Soccer World Cup X},
    doi = {10.1007/978-3-540-74024-7_24},
    url = {https://doi.org/10.1007/978-3-540-74024-7_24},
    }
  • Proprioceptive motion modeling for monte carlo localization
    J. Hoffmann
    in Robocup 2006: robot soccer world cup x , G. Lakemeyer, E. Sklar, D. G. Sorrenti, and T. Takahashi, Eds. , Berlin, Heidelberg: Springer Berlin Heidelberg , 2007 , pp. 258-269 .
    [BibTeX] [Abstract] [PDF]
    This paper explores how robot localization can be improved and made more reactive by using an adaptive motion model based on proprioception. The motion model of mobile robots is commonly assumed to be constant or a function of the robot speed. We extend this model by explicitly modeling possible states of locomotion caused by interactions of the robot with its environment, such as collisions. The motion model thus behaves according to which state the robot is in. State transitions are based on proprioception, which in our case describes how well the robot’s limbs are able to follow their respective motor commands. The extended, adaptive motion model yields a better, more reactive model of the current robot belief, which is shown in experiments. The improvement is due to the fact that the motion noise no longer has to subsume any possible outcome of actions including failure. In contrast, a clear distinction between failure and normal, desired operation is possible, which is reflected in the motion model.
    @InBook{Hoffmann2007,
    pages = {258--269},
    title = {Proprioceptive Motion Modeling for Monte Carlo Localization},
    publisher = {Springer Berlin Heidelberg},
    year = {2007},
    author = {Hoffmann, Jan},
    editor = {Lakemeyer, Gerhard and Sklar, Elizabeth and Sorrenti, Domenico G. and Takahashi, Tomoichi},
    address = {Berlin, Heidelberg},
    isbn = {978-3-540-74024-7},
    abstract = {This paper explores how robot localization can be improved and made more reactive by using an adaptive motion model based on proprioception. The motion model of mobile robots is commonly assumed to be constant or a function of the robot speed. We extend this model by explicitly modeling possible states of locomotion caused by interactions of the robot with its environment, such as collisions. The motion model thus behaves according to which state the robot is in. State transitions are based on proprioception, which in our case describes how well the robot's limbs are able to follow their respective motor commands. The extended, adaptive motion model yields a better, more reactive model of the current robot belief, which is shown in experiments. The improvement is due to the fact that the motion noise no longer has to subsume any possible outcome of actions including failure. In contrast, a clear distinction between failure and normal, desired operation is possible, which is reflected in the motion model.},
    booktitle = {RoboCup 2006: Robot Soccer World Cup X},
    doi = {10.1007/978-3-540-74024-7_22},
    url = {https://doi.org/10.1007/978-3-540-74024-7_22},
    }
  • Bearing-only localization for mobile robots
    M. Jüngel
    Proceedings of the 2007 international conference on advanced robotics (icar 2007), jeju, korea, , 2007 .
    [BibTeX] [Abstract] [PDF]
    In this paper we describe a new localization-method. It is a bearing-only approach which only uses horizontal bearings to landmarks and incorporates odometry. The approach is perfectly suited for mobile robots equipped with a camera because bearings can be extracted from images with high accuracy. The method itself does not need any internal representation of the robot’s position which is updated by alternating motion and sensor updates. The position is calculated directly using a short term memory of the observations. Our approach is also able to generate template positions for Monte-Carlo localization. The distribution of the template positions reflects the accuracy of the position calculation which depends on the configuration of the selected landmarks. In this paper we give a detailed description of the method and show the results of experiments conducted on a Sony Aibo robot demonstrating the precision.
    @InProceedings{juengel-bearingonlyocalizationformobilerobots,
    author = {Matthias Jüngel},
    title = {Bearing-Only Localization for Mobile Robots},
    booktitle = {Proceedings of the 2007 International Conference on Advanced Robotics (ICAR 2007), Jeju, Korea,},
    year = {2007},
    month = {August},
    abstract = {In this paper we describe a new localization-method.
    It is a bearing-only approach which only uses horizontal bearings
    to landmarks and incorporates odometry. The approach is perfectly suited for mobile robots equipped with a camera because
    bearings can be extracted from images with high accuracy. The
    method itself does not need any internal representation of the
    robot’s position which is updated by alternating motion and
    sensor updates. The position is calculated directly using a short
    term memory of the observations. Our approach is also able
    to generate template positions for Monte-Carlo localization. The distribution of the template positions reflects the accuracy of the
    position calculation which depends on the configuration of the
    selected landmarks. In this paper we give a detailed description
    of the method and show the results of experiments conducted on a Sony Aibo robot demonstrating the precision.
    },
    timestamp = {2007.08.02},
    }
  • Improving vision-based distance measurements using reference objects
    M. Jüngel, H. Mellmann, and M. Spranger
    Robocup 2007: robot soccer world cup xi , 2007 , pp. 89-100.
    [BibTeX] [Abstract] [PDF]
    Robots perceiving their environment using cameras usually need a good representation of how the camera is aligned to the body and how the camera is rotated relative to the ground. This is especially important for bearing-based distance measurements. In this paper we show how to use reference objects to improve vision-based distance measurements to objects of unknown size. Several methods for different kinds of reference objects are introduced. These are objects of known size (like a ball), objects extending over the horizon (like goals and beacons), and objects with known shape on the ground (like field lines). We give a detailed description how to determine the rotation of the robot’s camera relative to the ground, provide an error-estimation for all methods and describe the experiments we performed on an Aibo robot.
    @InProceedings{RC-JuengelMellmannSpranger-07,
    author = {Matthias Jüngel and Heinrich Mellmann and Michael Spranger},
    title = {Improving Vision-Based Distance Measurements using Reference Objects},
    booktitle = {RoboCup 2007: Robot Soccer World Cup XI},
    year = {2007},
    editor = {Ubbo Visser and Fernando Ribeiro and Takeshi Ohashi and Frank Dellaert},
    volume = {Volume 5001/2008},
    series = {Lecture Notes in Computer Science},
    pages = {89--100},
    publisher = {Springer Berlin / Heidelberg},
    abstract = {Robots perceiving their environment using cameras usually need a good
    representation of how the camera is aligned to the body and how the
    camera is rotated relative to the ground. This is especially important
    for bearing-based distance measurements. In this paper we show how
    to use reference objects to improve vision-based distance measurements
    to objects of unknown size. Several methods for different kinds of
    reference objects are introduced. These are objects of known size
    (like a ball), objects extending over the horizon (like goals and
    beacons), and objects with known shape on the ground (like field
    lines). We give a detailed description how to determine the rotation
    of the robot's camera relative to the ground, provide an error-estimation
    for all methods and describe the experiments we performed on an Aibo
    robot.},
    doi = {10.1007/978-3-540-68847-1},
    keywords = {RoboCup, humanoid robots, Aibo, camera matrix, reference objects},
    timestamp = {Freitag, 18. Juli 2008},
    url = {http://www.springerlink.com/content/y4730241r836k4l5},
    }
  • GermanTeam 2007 – The German national RoboCup team
    T. Röfer, J. Brose, D. Göhring, M. Jüngel, T. Laue, and M. Risler
    Robocup 2007: robot soccer world cup xi preproceedings , 2007 .
    [BibTeX]
    @InProceedings{2007:RoboCup-TDP-GermanTeam,
    author = {T. R\"ofer and J. Brose and D. G\"ohring and M. J\"ungel and T. Laue and M. Risler},
    title = {German{T}eam 2007 - {T}he {G}erman National {R}obo{C}up Team},
    booktitle = {RoboCup 2007: Robot Soccer World Cup XI Preproceedings},
    year = {2007},
    publisher = {RoboCup Federation},
    timestamp = {2014.01.27},
    }

2006

  • Die Doppelpass-Architektur. Verhaltenssteuerung autonomer Agenten in dynamischen Umgebungen
    R. Berger , Diploma Thesis , Humboldt-Universität zu Berlin, Institut für Informatik , 2006 , (in German).
    [BibTeX]
    @MastersThesis{diplom-berger,
    author = {Ralf Berger},
    title = {Die {D}oppelpass-{A}rchitektur. {V}erhaltenssteuerung autonomer {A}genten in dynamischen {U}mgebungen},
    school = {Humboldt-Universität zu Berlin, Institut für Informatik},
    year = {2006},
    type = {Diploma Thesis},
    note = {(in German)},
    timestamp = {2014.01.27},
    }
  • Distributed object modeling using object relations in dynamic environments
    D. Göhring
    Proceedings of concurrency, specification and programming (cs&p 2006) , 2006 .
    [BibTeX] [Abstract] [PDF]
    In this paper we describe how a group of agents can commonly estimate the position of objects. Furthermore we will show how these modeled object positions can be used for an improved self localization. Modeling of moving objects is commonly done by a single agent and in a robo-centric coordinate frame because this information is sufficient for most low level robot control and it is independent of the quality of the current robot localization. Especially when many robots cooperate with each other in a partially observable environment they have to share and to communicate information. For multiple robots to cooperate and to share information, though, they need to agree on a global, allocentric frame of reference. But when transforming the egocentric object model into a global one, it inherits the localization error of the robot in addition to the error associated with the egocentric model. We propose using the relation of objects detected in camera images to other objects in the same camera image as a basis for estimating the position of the object in a global coordinate system. The spacial relation of objects with respect to stationary objects (e.g., landmarks) offers several advantages: The information is independent of robot localization and odometry. It can be easily communicated. We present experimental evidence that shows how two robots are able to infer the position of an object within a global frame of reference, even though they are not localized themselves. We will also show, how to use this object information for self localization. A third aspect of this work will cope with the communication delay therefore we will show, how the Markov Model can be extended for distributed object tracking.
    @InProceedings{goehring06csp,
    author = {Daniel Göhring},
    title = {Distributed Object Modeling Using Object Relations in Dynamic Environments},
    booktitle = {Proceedings of Concurrency, Specification and Programming (CS\&P 2006)},
    year = {2006},
    abstract = {In this paper we describe how a group of agents can commonly estimate the position of objects. Furthermore we will show how
    these modeled object positions can be used for an improved self localization. Modeling of moving objects is commonly done by a single agent and in a robo-centric coordinate frame because this information is sufficient for most low level robot control and it is independent of the quality of the current robot localization. Especially when many robots cooperate
    with each other in a partially observable environment they have to share and to communicate information. For multiple robots to cooperate and to share information, though, they need to agree on a global, allocentric frame of reference. But when transforming the egocentric object model into a global one, it inherits the localization error of the robot in addition
    to the error associated with the egocentric model. We propose using the relation of objects detected in camera images to
    other objects in the same camera image as a basis for estimating the position of the object in a global coordinate system. The spacial relation of objects with respect to stationary objects (e.g., landmarks) offers several advantages: The information is independent of robot localization and odometry. It can be easily communicated. We present experimental
    evidence that shows how two robots are able to infer the position of an object within a global frame of reference, even though they are not localized themselves. We will also show, how to use this object information for self localization. A third aspect of this work will cope with the communication delay therefore we will show, how the Markov Model can be
    extended for distributed object tracking.},
    timestamp = {2007.05.04},
    }
  • Multi robot object tracking and self localization using visual percept relations
    D. Gohring and H. d. Burkhard
    2006 ieee/rsj international conference on intelligent robots and systems , 2006 , pp. 31-36.
    [BibTeX] [Abstract] [PDF]
    In this paper we present a novel approach to estimating the position of objects tracked by a team of mobile robots and to use these objects for a better self localization. Modeling of moving objects is commonly done in a robo-centric coordinate frame because this information is sufficient for most low level robot control and it is independent of the quality of the current robot localization. For multiple robots to cooperate and share information, though, they need to agree on a global, allocentric frame of reference. When transforming the egocentric object model into a global one, it inherits the localization error of the robot in addition to the error associated with the egocentric model. We propose using the relation of objects detected in camera images to other objects in the same camera image as a basis for estimating the position of the object in a global coordinate system. The spatial relation of objects with respect to stationary objects (e.g., landmarks) offers several advantages: a) Errors in feature detection are correlated and not assumed independent. Furthermore, the error of relative positions of objects within a single camera frame is comparably small, b) The information is independent of robot localization and odometry. c) As a consequence of the above, it provides a highly efficient method for communicating information about a tracked object and communication can be asynchronous, d) As the modeled object is independent from robo-centric coordinates, its position can be used for self localization of the observing robot. We present experimental evidence that shows how two robots are able to infer the position of an object within a global frame of reference, even though they are not localized themselves and then use this object information for self- localization
    @InProceedings{4058894,
    author = {D. Gohring and H. d. Burkhard},
    title = {Multi Robot Object Tracking and Self Localization Using Visual Percept Relations},
    booktitle = {2006 IEEE/RSJ International Conference on Intelligent Robots and Systems},
    year = {2006},
    pages = {31-36},
    month = {Oct},
    abstract = {In this paper we present a novel approach to estimating the position of objects tracked by a team of mobile robots and to use these objects for a better self localization. Modeling of moving objects is commonly done in a robo-centric coordinate frame because this information is sufficient for most low level robot control and it is independent of the quality of the current robot localization. For multiple robots to cooperate and share information, though, they need to agree on a global, allocentric frame of reference. When transforming the egocentric object model into a global one, it inherits the localization error of the robot in addition to the error associated with the egocentric model. We propose using the relation of objects detected in camera images to other objects in the same camera image as a basis for estimating the position of the object in a global coordinate system. The spatial relation of objects with respect to stationary objects (e.g., landmarks) offers several advantages: a) Errors in feature detection are correlated and not assumed independent. Furthermore, the error of relative positions of objects within a single camera frame is comparably small, b) The information is independent of robot localization and odometry. c) As a consequence of the above, it provides a highly efficient method for communicating information about a tracked object and communication can be asynchronous, d) As the modeled object is independent from robo-centric coordinates, its position can be used for self localization of the observing robot. We present experimental evidence that shows how two robots are able to infer the position of an object within a global frame of reference, even though they are not localized themselves and then use this object information for self- localization},
    doi = {10.1109/IROS.2006.282427},
    issn = {2153-0858},
    keywords = {feature extraction;mobile robots;multi-robot systems;path planning;position control;egocentric object model;feature detection;mobile robots;multi robot object tracking;self localization;visual percept relations;Cameras;Data mining;Intelligent robots;Mobile robots;Object detection;Orbital robotics;Robot kinematics;Robot localization;Robot sensing systems;Robot vision systems;Sensor Fusion;Sensor Networks},
    url = {http://www.naoteamhumboldt.de/wp-content/papercite-data/pdf/2006-Goehring-MultiRobotObjectTrackingandSelfLocalizationUsingVisualPerceptRelations.pdf},
    }
  • Further studies on the use of negative information in mobile robot localization
    J. Hoffmann, M. Spranger, D. Gohring, M. Jungel, and H. Burkhard
    Proceedings 2006 ieee international conference on robotics and automation, 2006. icra 2006. , 2006 , pp. 62-67.
    [BibTeX] [Abstract] [PDF]
    This paper deals with how the absence of an expected sensor reading can be used to improve Markov localization. Negative information has not been used for robot localization for various reasons like sensor imperfections, and occlusions that make it hard to determine if a missing sensor reading is really caused by the absence of a feature. We address these difficulties by carefully modeling the robot’s main sensor, its camera. Taking into account the viewing frustum and detected obstacles, the absence of a sensor reading can be associated with the absence of that particular feature. This information can then be integrated into the localization process. We show the positive effect on robot localization in various experiments. (a) In a specific setup, the robot is able to localize using negative information where without it, it is unable to localize. (b) We demonstrate the importance of modeling occlusions and the impact of false negatives on localization. (c) We show the positive impact in a typical run
    @InProceedings{1641162,
    author = {J. Hoffmann and M. Spranger and D. Gohring and M. Jungel and H. Burkhard},
    title = {Further studies on the use of negative information in mobile robot localization},
    booktitle = {Proceedings 2006 IEEE International Conference on Robotics and Automation, 2006. ICRA 2006.},
    year = {2006},
    pages = {62-67},
    month = {May},
    abstract = {This paper deals with how the absence of an expected sensor reading can be used to improve Markov localization. Negative information has not been used for robot localization for various reasons like sensor imperfections, and occlusions that make it hard to determine if a missing sensor reading is really caused by the absence of a feature. We address these difficulties by carefully modeling the robot's main sensor, its camera. Taking into account the viewing frustum and detected obstacles, the absence of a sensor reading can be associated with the absence of that particular feature. This information can then be integrated into the localization process. We show the positive effect on robot localization in various experiments. (a) In a specific setup, the robot is able to localize using negative information where without it, it is unable to localize. (b) We demonstrate the importance of modeling occlusions and the impact of false negatives on localization. (c) We show the positive impact in a typical run},
    doi = {10.1109/ROBOT.2006.1641162},
    issn = {1050-4729},
    keywords = {Markov processes;mobile robots;path planning;Markov localization;mobile robot localization;negative information;Artificial intelligence;Cameras;Computer science;Intelligent sensors;Laboratories;Mobile robots;Monte Carlo methods;Robot localization;Robot sensing systems;Robot vision systems},
    url = {http://www.naoteamhumboldt.de/wp-content/papercite-data/pdf/2006-Hoffmann-FurtherStudiesOnTheUseOfNegativeInformationInMobileRobotLocalization.pdf},
    }
  • Xabsl – a pragmatic approach to behavior engineering
    M. Lötzsch, M. Risler, and M. Jüngel
    Proceedings of ieee/rsj international conference of intelligent robots and systems (iros) , Beijing, China , 2006 , pp. 5124-5129.
    [BibTeX] [Abstract] [PDF]
    This paper introduces the Extensible Agent Behavior Specification Language (XABSL) as a pragmatic tool for engineering the behavior of autonomous agents in complex and dynamic environments. It is based on hierarchies of finite state machines (FSM) for action selection and supports the design of longterm and deliberative decision processes as well as of short-term and reactive behaviors. A platform-independent execution engine makes the language applicable on any robotic platform and together with a variety of visualization, editing and debugging tools, XABSL is a convenient and powerful system for the development of complex behaviors. The complete source code can be freely downloaded from the XABSL website (http://www.informatik.huberlin.de/ki/XABSL/). The language has been successfully applied on many robotic platforms, mainly in the domain of RoboCup robot soccer. It gave the GermanTeam the crucial advantage over other teams to become the 2004 and 2005 world champion in the Four-Legged League and helped the team CoPS Stuttgart to become third in the Middle Size League in 2004.
    @InProceedings{2006:IROSLoetzschRislerJuengel,
    author = {M. Lötzsch and M. Risler and M. Jüngel},
    title = {XABSL - A Pragmatic Approach to Behavior Engineering},
    booktitle = {Proceedings of IEEE/RSJ International Conference of Intelligent Robots and Systems (IROS)},
    year = {2006},
    pages = {5124-5129},
    address = {Beijing, China},
    month = {October 9-15},
    abstract = {This paper introduces the Extensible Agent Behavior Specification Language (XABSL) as a pragmatic tool for engineering the behavior of autonomous agents in complex and dynamic environments. It is based on hierarchies of finite state machines
    (FSM) for action selection and supports the design of longterm and deliberative decision processes as well as of short-term
    and reactive behaviors. A platform-independent execution engine makes the language applicable on any robotic platform and together with a variety of visualization, editing and debugging tools, XABSL is a convenient and powerful system for the development of complex behaviors. The complete source code can be freely downloaded from the XABSL website (http://www.informatik.huberlin.de/ki/XABSL/). The language has been successfully applied on many robotic platforms, mainly in the domain of RoboCup robot soccer. It gave the GermanTeam the crucial advantage over other teams to become the 2004 and 2005 world champion in the Four-Legged League and helped the team CoPS Stuttgart to become third in the Middle Size League in 2004.},
    timestamp = {2014.01.27},
    }

2005

  • Programming bounded rationality
    H. Burkhard
    in Monitoring, security, and rescue techniques in multiagent systems , Berlin, Heidelberg: Springer Berlin Heidelberg , 2005 , pp. 347-362 .
    [BibTeX] [Abstract] [PDF]
    Research on Artificial Intelligence and Robotics helps to understand problems of rationality. Autonomous robots have to act and react in complex environments under constraints of their bounded resources. Simple daily tasks are much more difficult to implement than playing chess. Soccer playing robots are considered as test field for rational behavior. Our implementations are inspired by the belief-desire-intention model.
    @InBook{Burkhard2005,
    pages = {347--362},
    title = {Programming Bounded Rationality},
    publisher = {Springer Berlin Heidelberg},
    year = {2005},
    author = {Burkhard, Hans-Dieter},
    address = {Berlin, Heidelberg},
    isbn = {978-3-540-32370-9},
    abstract = {Research on Artificial Intelligence and Robotics helps to understand problems of rationality. Autonomous robots have to act and react in complex environments under constraints of their bounded resources. Simple daily tasks are much more difficult to implement than playing chess. Soccer playing robots are considered as test field for rational behavior. Our implementations are inspired by the belief-desire-intention model.},
    booktitle = {Monitoring, Security, and Rescue Techniques in Multiagent Systems},
    doi = {10.1007/3-540-32370-8_28},
    url = {https://doi.org/10.1007/3-540-32370-8_28},
    }

2004

  • Vierbeiniges Laufen: Modellierung und Optimierung von Roboterbewegungen
    U. Düffert , Diploma Thesis , Humboldt-Universität zu Berlin , 2004 , (in German).
    [BibTeX]
    @MastersThesis{DiplomaThesis-Dueffert-04,
    author = {Uwe Düffert},
    title = {{V}ierbeiniges {L}aufen: {M}odellierung und {O}ptimierung von {R}oboterbewegungen},
    school = {Humboldt-Universität zu Berlin},
    year = {2004},
    type = {Diploma Thesis},
    note = {(in German)},
    timestamp = {2014.01.27},
    }

2000

  • Humboldt hereos in robocup-99 (team description)
    H. Burkhard, M. Werner, M. Ritzschke, F. Winkler, J. Wendler, A. Georgi, U. Düffert, and H. Myritz
    in Robocup-99: robot soccer world cup iii , M. Veloso, E. Pagello, and H. Kitano, Eds. , Berlin, Heidelberg: Springer Berlin Heidelberg , 2000 , pp. 770-773 .
    [BibTeX] [Abstract] [PDF]
    The team members include students as well as members of the teaching stuff from the Department of Computer Science at the Humboldt University. They represent the groups of Artificial Intelligence, Responsive Computing, and Signal Processing, respectively. It was the aim of the project to combine the skills of these disciplines to program soccer playing legged robots.
    @InBook{HH-Burkhard2000,
    pages = {770--773},
    title = {Humboldt Hereos in RoboCup-99 (Team description)},
    publisher = {Springer Berlin Heidelberg},
    year = {2000},
    author = {Burkhard, Hans-Dieter and Werner, Matthias and Ritzschke, Michael and Winkler, Frank and Wendler, Jan and Georgi, Andrej and D{\"u}ffert, Uwe and Myritz, Helmut},
    editor = {Veloso, Manuela and Pagello, Enrico and Kitano, Hiroaki},
    address = {Berlin, Heidelberg},
    isbn = {978-3-540-45327-7},
    abstract = {The team members include students as well as members of the teaching stuff from the Department of Computer Science at the Humboldt University. They represent the groups of Artificial Intelligence, Responsive Computing, and Signal Processing, respectively. It was the aim of the project to combine the skills of these disciplines to program soccer playing legged robots.},
    booktitle = {RoboCup-99: Robot Soccer World Cup III},
    doi = {10.1007/3-540-45327-X_102},
    url = {http://www.naoteamhumboldt.de/wp-content/papercite-data/pdf/1999-HumboldtHereosTeamDescription.pdf},
    }
  • At humboldt in robocup-99 (team description)
    H. Burkhard, J. Wendler, T. Meinert, H. Myritz, and G. Sander
    in Robocup-99: robot soccer world cup iii , M. Veloso, E. Pagello, and H. Kitano, Eds. , Berlin, Heidelberg: Springer Berlin Heidelberg , 2000 , pp. 542-545 .
    [BibTeX] [Abstract] [PDF]
    Our agent team AT Humboldt 99 (AT stands for “Agent Team”) was developed as extension of our former team AT Humboldt 98, which became vice champion at RoboCup-98. We started to extend it by improved skills, new options and a larger planning horizon, respectively. So the most features of our current team were already part of AT Humboldt 98 which has been briefly described in [3] and extensive described in [5]. A description of our first soccer team AT Humboldt 97, which became world champion at RoboCup-97, can be found in [1].
    @InBook{ATH-Burkhard2000,
    pages = {542--545},
    title = {AT Humboldt in RoboCup-99 (Team description)},
    publisher = {Springer Berlin Heidelberg},
    year = {2000},
    author = {Burkhard, Hans-Dieter and Wendler, Jan and Meinert, Thomas and Myritz, Helmut and Sander, Gerd},
    editor = {Veloso, Manuela and Pagello, Enrico and Kitano, Hiroaki},
    address = {Berlin, Heidelberg},
    isbn = {978-3-540-45327-7},
    abstract = {Our agent team AT Humboldt 99 (AT stands for ``Agent Team'') was developed as extension of our former team AT Humboldt 98, which became vice champion at RoboCup-98. We started to extend it by improved skills, new options and a larger planning horizon, respectively. So the most features of our current team were already part of AT Humboldt 98 which has been briefly described in [3] and extensive described in [5]. A description of our first soccer team AT Humboldt 97, which became world champion at RoboCup-97, can be found in [1].},
    booktitle = {RoboCup-99: Robot Soccer World Cup III},
    doi = {10.1007/3-540-45327-X_46},
    url = {http://www.naoteamhumboldt.de/wp-content/papercite-data/pdf/1999-ATH-TeamDescription.pdf},
    }

1999

  • At humboldt in robocup-98 (team description)
    P. Gugenberger, J. Wendler, K. Schröter, and H. Burkhard
    in Robocup-98: robot soccer world cup ii , M. Asada and H. Kitano, Eds. , Berlin, Heidelberg: Springer Berlin Heidelberg , 1999 , pp. 358-363 .
    [BibTeX] [Abstract] [PDF]
    The paper describes the scientific goals of the virtual soccer team “AT Humboldt 98”, which became vice champion in RoboCup-98 in Paris. It is the successor of the world champion “AT Humboldt” from RoboCup-97 in Nagoya.
    @InBook{Gugenberger1999,
    pages = {358--363},
    title = {AT Humboldt in RoboCup-98 (Team description)},
    publisher = {Springer Berlin Heidelberg},
    year = {1999},
    author = {Gugenberger, Pascal and Wendler, Jan and Schr{\"o}ter, Kay and Burkhard, Hans-Dieter},
    editor = {Asada, Minoru and Kitano, Hiroaki},
    address = {Berlin, Heidelberg},
    isbn = {978-3-540-48422-6},
    abstract = {The paper describes the scientific goals of the virtual soccer team ``AT Humboldt 98'', which became vice champion in RoboCup-98 in Paris. It is the successor of the world champion ``AT Humboldt'' from RoboCup-97 in Nagoya.},
    booktitle = {RoboCup-98: Robot Soccer World Cup II},
    doi = {10.1007/3-540-48422-1_30},
    url = {http://www.naoteamhumboldt.de/wp-content/papercite-data/pdf/1998-ATH-TeamDescription.pdf},
    }

1998

  • At humboldt — development, practice and theory
    H. Burkhard, M. Hannebauer, and J. Wendler
    in Robocup-97: robot soccer world cup i , H. Kitano, Ed. , Berlin, Heidelberg: Springer Berlin Heidelberg , 1998 , pp. 357-372 .
    [BibTeX] [Abstract] [PDF]
    This article covers three basics of our virtual soccer team AT Humboldt: We describe our development process in the frame of a practical exercise for students. The resulting efficient agent-oriented realization is explained, and we give a theoretical embedding of our planning component based on BDI.
    @InBook{Burkhard1998,
    pages = {357--372},
    title = {AT Humboldt --- Development, practice and theory},
    publisher = {Springer Berlin Heidelberg},
    year = {1998},
    author = {Burkhard, Hans-Dieter and Hannebauer, Markus and Wendler, Jan},
    editor = {Kitano, Hiroaki},
    address = {Berlin, Heidelberg},
    isbn = {978-3-540-69789-3},
    abstract = {This article covers three basics of our virtual soccer team AT Humboldt: We describe our development process in the frame of a practical exercise for students. The resulting efficient agent-oriented realization is explained, and we give a theoretical embedding of our planning component based on BDI.},
    booktitle = {RoboCup-97: Robot Soccer World Cup I},
    doi = {10.1007/3-540-64473-3_73},
    url = {http://www.naoteamhumboldt.de/wp-content/papercite-data/pdf/1998-ATH-DevelopmentPracticeTheory.pdf},
    }