RoboCup 2005 Publications

From RoboCup Wiki
Jump to: navigation, search
Author Title Year Journal/Proceedings Reftype DOI/URL
Andreas D. Lattner, Andrea Miene, U.V. & Herzog, O.

Sequential Pattern Mining for Situation and Behavior Prediction in Simulated Robotic Soccer


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 118-129 inproceedings

DOI URL

Abstract: Agents in dynamic environments have to deal with world representations that change over time. In order to allow agents to act autonomously and to make their decisions on a solid basis an interpretation of the current scene is necessary. If intentions of other agents or events that are likely to happen in the future can be recognized the agent’s performance can be improved as it can adapt the behavior to the situation. In this work we present an approach which applies unsupervised symbolic learning off-line to a qualitative abstraction in order to create frequent patterns in dynamic scenes. These patterns can be later applied during runtime in order to predict future situations and behaviors. The pattern mining approach was applied to two games of the 2D RoboCup simulation league.
Anzani, F., Bosisio, D., Matteucci, M. & Sorrenti, D.G.

On-Line Color Calibration in Non-Stationary Environments


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 396-407 inproceedings

DOI URL

Abstract: In this paper we propose an approach to color classification and image segmentation in non-stationary environments. Our goal is to cope with changing illumination condition by on-line adapting both the parametric color model and its structure/complexity. Other authors used parametric statistics to model color distribution in segmentation and tracking problems, but with a fixed complexity model. Our approach is able to on-line adapt also the complexity of the model, to cope with large variations in the scene illumination and color temperature.
This work has been partially supported by the MADSys project P.R.I.N. 2003 “Software technologies and knowledge models for design, implementation and testing of multi-agent robotic system in real environments”, funded by M.I.U.R.
Asplund, L., Bergkvist, H. & Nordh, T.

ChipVision2 - A Stereo Vision System for Robots Based on Reconfigurable Hardware


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 359-370 inproceedings

DOI URL

Abstract: A system utilizing reconfigurable hardware of 1 million gates and two CMOS cameras is used in an image analysis system. The system is a part of a sensor system for a robot, and can deliver data about the robots position as well as relative distances of other objects in real time.
Behnke, S., Müller, J. & Schreiber, M.

Playing Soccer with RoboSapien


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 36-48 inproceedings

DOI URL

Abstract: Due to limited availability of humanoid robots and the high costs involved, multi-agent experiments with humanoid robots have been at least difficult so far. With the introduction of RoboSapien, a low-cost humanoid robot developed for the toy market, this situation has changed.
This paper describes how we augmented multiple RoboSapiens to obtain a team of soccer playing humanoid robots. We added a Pocket PC and a color camera to the robot base to make it autonomous.
For a team of these augmented RoboSapiens, we implemented computer vision and self localization. We designed basic soccer skills, such as approaching the ball, dribbling the ball towards the goal, and defending the goal. We set up a soccer field and played test games in our lab to evaluate the system.
The paper reports experiences made during these soccer matches as well as results on a scoring task. We also tested this system at RoboCup German Open 2005, where we played soccer matches against the Brainstormers Osnabrück, who also used augmented RoboSapiens.
Behnke, S., Müller, J. & Schreiber, M.

Toni: A Soccer Playing Humanoid Robot


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 59-70 inproceedings

DOI URL

Abstract: This paper describes the humanoid robot Toni that has been designed to play soccer in the RoboCup Humanoid League. The paper details Toni’s mechanical and electrical design, perception, self localization, behavior control, and infrastructure.
Toni is fully autonomous, has a low weight (2.2kg), and is much taller (74cm) than most servo-driven humanoid robots. It has a wide field of view camera, ample computing power, and wireless communication.
Toni possesses basic soccer skills. It walks dynamically in all directions (up to 20cm/s in forward direction) and turns on the spot. It perceives the ball and the goals and localizes itself on the field. Toni is able to approach the ball and to dribble it. It can kick the ball without falling.
We performed tests in our lab and penalty kick demonstrations at RoboCup German Open 2005. Toni’s successors Jupp, Sepp, and Max performed well at the RoboCup 2005 Humanoid League competitions.
Birk, A. & Condea, C.

Mobile Robot Communication Without the Drawbacks of Wireless Networking


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 585-592 inproceedings

DOI URL

Abstract: The default solution for mobile robot communication is RF-networking, typically based on one of the IEEE 802.11 standards also known as WLAN technology. Radio communication frees the robots from umbilical cords. But it suffers from several significant drawbacks, especially limited bandwidth and range. The limitations of both aspects are in addition hard to predict as they are strongly dependent on environment conditions. An outdoor RF-link may easily cover 100m over a line-of-sight with full bandwidth. In an indoor environment, the range often drops to a few rooms. Walls made of hardened concrete even completely block the communication. Driven by a concrete application scenario where communication is vital, namely robot rescue, we developed a communication system based on glassfibre links. The system provides 100MBit ethernet connections over up to 100m in its default configuration. The glassfibres provide high bandwidth, they are very lightweight and thin, and they can take a lot of stress, much more than normal copper cable. The glassfiber links are deployed from the mobile robot via a cable drum. The system is based on media converters at both ends. One of them is integrated on the drum, thus allowing the usage of inexpensive wired sliprings. The glassfibre system turned out to be very performant and reliable, both in operation in the challenging environment of rescue robotics as well as in concrete experiments.
Birk, A. & Pfingsthorn, M.

A HMI Supporting Adjustable Autonomy of Rescue Robots


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 255-266 inproceedings

DOI URL

Abstract: Human rescue workers are a scarce resource at disaster sites. But it is still a long way to go before fully autonomous rescue robots will be fieldable. The usefulness of rescue robots will hence strongly depend on the availability of user interfaces that enable a single first responder to operate a whole set of robots. For this challenge, it is important to preprocess and streamline the immense data flow from the robots and to assist the operator as much as possible in the processes of controlling the robots. This paper introduces an adaptive graphical interface supporting adjustable autonomy of rescue robots. The design is based on insights from the literature in this field where intensive surveys of the actual needs in this domain were compiled.
Bonarini, A., Lavatelli, D. & Matteucci, M.

A Composite System for Real-Time Robust Whistle Recognition


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 130-141 inproceedings

DOI URL

Abstract: In Robocup Middle-Size League (MSL) the challenge to recognize signals given by the referee by whistling has been introduced from this year as a way to reduce the interaction via radio-link. We present Whistle Recognizer (WR), a system able to recognize different whistling patterns, after a relatively short training done in advance. This composite system encompasses neural networks and more traditional information processing techniques. It demonstrated to be quite effective and can be easily integrated in a multi-thread control architecture, as the vast majority of those used in the league; thus, it candidates itself as a potential off-the-shelf module to be used by MSL teams not interested in research about signal processing and analysis.
de Cabrol, A., Bonnin, P., Costis, T., Hugel, V., Blazevic, P. & Bouchefra, K.

A New Video Rate Region Color Segmentation and Classification for Sony Legged RoboCup Application


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 436-443 inproceedings

DOI URL

Abstract: Whereas numerous methods are used for vision systems embedded on robots, only a few use colored region segmentation mainly because of the processing time. In this paper, we propose a real-time (i.e. video rate) color region segmentation followed by a robust color classification and region merging dedicated to various applications such as RoboCup four-legged league or an industrial conveyor wheeled robot. Performances of this algorithm and confrontation with other existing methods are provided.
Carpin, S., Wang, J., Lewis, M., Birk, A. & Jacoff, A.

High Fidelity Tools for Rescue Robotics: Results and Perspectives


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 301-311 inproceedings

DOI URL

Abstract: USARSim is a high fidelity robot simulation tool based on a commercial game engine. We illustrate the overall structure of the simulator and we argue about its use as a bridging tool between the RoboCupRescue Real Robot League and the RoboCupRescue Simulation League. In particular we show some results concerning the validation of the system. Algorithms useful for the search and rescue task have been developed in the simulator and then executed on real robots providing encouraging results.
Chandrasekharan, S., Esfandiari, B. & Hassan, T.

Advantages of the Signaling Strategy in a Dynamic Environment: Cognitive Modeling Using RoboCup, The


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 665-672 inproceedings

DOI URL

Abstract: We report a cognitive modeling experiment where the RoboCup simulation environment was used to study the advantages provided by signals. We used the passing problem in RoboCup as our test problem and soccer-players’ ’yells’ of their ’passability’ values as the task-specific signals. We found that yells improve pass completion – using yells to decide the best player (to pass the ball) led to a 8-17 percentage points increase in performance compared to a centralized calculation of best pass. However, the passability values themselves did not make a difference, indicating that the advantage of signals come from their different perspective in identifying a pass, the actual content of signals do not matter. We present some problems we faced in using Robocup as a modeling environment, and suggest features that would help promote the use of RoboCup in cognitive modeling.
Dashti, H.T., Aghaeepour, N., Asadi, S., Bastani, M., Delafkar, Z., Disfani, F.M., Ghaderi, S.M., Kamali, S., Pashami, S. & Siahpirani, A.F.

Dynamic Positioning Based on Voronoi Cells (DPVC)


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 219-229 inproceedings

DOI URL

Abstract: In this paper we are proposing an approach for flexible positioning of players in Soccer Simulation in a Multi-Agent environment. We introduce Dynamic Positioning based on Voronoi Cells (DPVC) as a new method for players’ positioning which uses Voronoi Diagram for distributing agents in the field. This method also uses Attraction Vectors that indicate agents’ tendency to specific objects in the field with regard to the game situation and players’ roles. Finally DPVC is compared with SBSP as the conventional method of positioning.
Düffert, U. & Hoffmann, J.

Reliable and Precise Gait Modeling for a Quadruped Robot


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 49-58 inproceedings

DOI URL

Abstract: We present a parametric walk model for a four-legged robot. The walk model is improved using a genetic algorithm, but unlike previous approaches, the fitness is determined in a run that closely resembles the later application. We thus not only achieve high speeds, but also a high degree of flexibility. In addition to the walking model being flexible, we present a means of automatically calibrating the walking engine. This allows for highly precise robot control and greatly improved odometry accuracy. Lastly, we show how the motion model can be extended to integrate specialized motions to further increase locomotion speed without compromising flexibility.
Fathzadeh, R., Mokhtari, V., Mousakhani, M. & Shahri, A.M.

Coaching with Expert System Towards RoboCup Soccer Coach Simulation


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 488-495 inproceedings

DOI URL

Abstract: In this paper we will describe our research in case of using Expert System as a decision-making system. We made our attempt to expose a base strategy from past log files and implement an online learning system which receives information from the environment. In developing the coach, the main research effort comprises two complementary parts: (a) Design a rule-based expert system in which its task is to analyze the game (b) Employing the decision-making trees for generating advice. Considering these two methods, coach learns to predict agent behavior and automatically generates advice to improve team’s performance. This structure is tested previously in RoboCup Soccer Coach Simulation League. Using this approach, the MRLCoach2004 took first place in the competition held in 2004.
Ferrein, A., Hermanns, L. & Lakemeyer, G.

Comparing Sensor Fusion Techniques for Ball Position Estimation


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 154-165 inproceedings

DOI URL

Abstract: In robotic soccer a good ball position estimate is essential for successful play. Given the uncertainties in the perception of each individual robot, merging the local perceptions of the robots into a global ball estimate often results in a more reliable estimate and helps to increase team performance. Robots can use the global ball position even if they themselves do not see the ball or they can use it to adjust their own perception faults. In this paper we report on our results of comparing state-of-the-art sensor fusion techniques like Kalman filters or the Monte Carlo approach in RoboCup’s Middle-size league. We compare our results to previously published work from other Middle-size league teams and show how the quality of perceiving the ball position is increased.
Flentge, F.

Learning to Approach a Moving Ball with a Simulated Two-Wheeled Robot


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 106-117 inproceedings

DOI URL

Abstract: We show how a two-wheeled robot can learn to approach a moving ball using Reinforcement Learning. The robot is controlled by setting the velocities of its two wheels. It has to reach the ball under certain conditions to be able to kick it towards a given target. In order to kick, the ball has to be in front of the robot. The robot also has to reach the ball at a certain angle in relation to the target, because the ball is always kicked in the direction from the center of the robot to the ball. The robot learns which velocity differences should be applied to the wheels: one of the wheels is set to the maximum velocity, the other one according to this difference. We apply a REINFORCE algorithm [1] in combination with some kind of extended Growing Neural Gas (GNG) [2] to learn these continuous actions. The resulting algorithm, called ReinforceGNG, is tested in a simulated environment with and without noise.
Fujii, H., Kato, M. & Yoshida, K.

Cooperative Action Control Based on Evaluating Objective Achievements


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 208-218 inproceedings

DOI URL

Abstract: This paper deals with a cooperative control method for a multi-agent system in dynamic environment. This method enables a robot to perform flexi-ble cooperation based on the global evaluation of the achievement of objectives. Each robot communicates qualitative evaluation on the achievement level of each objective. Each robot calculates the global evaluation on the achievement of the team objective from the individual evaluation. The evaluation on the ob-jective achievement is abstracted in order to reduce the influence of variation of the evaluation value and the communication load. As an example, the method is applied to the EIGEN team robots for the Middle Size League of RoboCup, since it is necessary for the soccer robots to cooperate each other in dynamic environment. Its effectiveness was demonstrated through the RoboCup 2004 competition.
Hamraz, S.H. & Feyzabadi, S.S.

General-Purpose Learning Machine Using K-Nearest Neighbors Algorithm


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 529-536 inproceedings

DOI URL

Abstract: The field of machine learning is concerned with the question of how to construct computer programs that automatically improve with experience. The aim of this paper is to describe a learner machine which can be used in different learning problems without any change in the system. We designed such a machine using k-nearest neighbors algorithm. How to optimize k-nearest neighbors algorithm to be effectively used in the machine is also discussed. Experimental results are also demonstrated at the end.
Hayashi, Y., Tohyama, S. & Fujiyoshi, H.

Mosaic-Based Global Vision System for Small Size Robot League


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 593-601 inproceedings

DOI URL

Abstract: In the RoboCup F180 Small Size League, a global vision system using multiple cameras has been used to capture the whole field view. In the overlapping area of two cameras’ views, a process to merge information from both cameras is needed. To avoid this complex process and rule-based approach, we propose a mosaic-based global vision system which produces high resolution images from multiple cameras. Three mosaic images, which take into account the height of each object such as our robots, opponent robots, and the ball on the field, are generated by pseudo corresponding points. Our system archives a position accuracy of better than 14.2 mm(mean: 4 mm) over a 4 × 5.5 m field.
Hoffmann, J., Spranger, M., Göhring, D. & Jüngel, M.

Exploiting the Unexpected: Negative Evidence Modeling and Proprioceptive Motion Modeling for Improved Markov Localization


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 24-35 inproceedings

DOI URL

Abstract: This paper explores how sensor and motion modeling can be improved to better Markov localization by exploiting deviations from expected sensor readings. Proprioception is achieved by monitoring target and actual motions of robot joints. This provides information about whether or not an action was executed as desired, yielding a quality measure of the current odometry. Odometry is usually extremely prone to errors for legged robots, especially in dynamic environments where collisions are often unavoidable, due to the many degrees of freedom of the robot and the numerous possibilities of motion hindrance. A quality measure helps differentiate the periods of unhindered motion from periods where robot motion was impaired for whatever reason. Negative evidence is collected when a robot fails to detect a landmark that it expects to see. Therefore the gaze direction of the camera has to be modeled accordingly. This enables the robot to localize where it could not when only using landmarks. In the general localization task, the probability distribution converges more quickly when negative information is taken into account.
Hugel, V., Amouroux, G., Costis, T., Bonnin, P. & Blazevic, P.

Specifications and Design of Graphical Interface for Hierarchical Finite State Machines


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 648-655 inproceedings

DOI URL

Abstract: This paper presents the specifications and the design of a simple graphical interface for building hierarchical finite state machines. This kind of tool can prove very useful for quickly designing hierarchical behaviors. It can be used in the frame of RoboCup to develop deterministic complex behaviors without focusing on C++ coding because source code can be generated from the interface. It is also possible to use it to generate hierarchical finite state machines for whatever purpose needed. The user can create state diagrams by drawing boxes for states and specifying transitions between states. A state diagram can represent a behavior and be considered as a metastate. Diagrams of metastates are possible to constitute several levels.
Iravani, P.

Discovering Relevant Sensor Data by Q-Analysis


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 81-92 inproceedings

DOI URL

Abstract: This paper proposes a novel method for supervised classification based on the methodology of Q-analysis. The classification is based on finding ‘relevant’ structures in the features describing the data, and using them to define each of the classes. The features not included in the structural definition of a class are considered as ‘irrelevant’. The paper uses three different data-sets to experimentally validate the method.
Jahshan, D.

Very High Speed, Close Field, Object Positioning Using Tri-Linear CCDs


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 702-707 inproceedings

DOI URL

Abstract: To be able to effectively intercept and control a soccer ball travelling at high speed, it is useful to be able to accurately track the position of the ball as it approaches the robot. In this paper we present a method that can calculate the position in two dimensions at thousands of frames per second using a pair of inexpensive tri-linear CCDs. Each CCD gathers RGB information, which is then colour segmented. This data is then fused to calculate the location of the object in 2D. Further, the amount of processing required to detect these objects is low, and can be accomplished using inexpensive electronic components.
Kaplan, K., Çelik, B., Meriçli, T., Meriçli, Ç. & Akın, H.L.

Practical Extensions to Vision-Based Monte Carlo Localization Methods for Robot Soccer Domain


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 624-631 inproceedings

DOI URL

Abstract: This paper proposes a set of practical extensions to the vision-based Monte Carlo localization (MCL) for RoboCup Sony AIBO legged robot soccer domain. The main disadvantage of AIBO robots is that they have a narrow field of view so the number of landmarks seen in one frame is usually not enough for geometric calculation. MCL methods have been shown to be accurate and robust in legged robot soccer domain but there are some practical issues that should be handled in order to maintain stability/elasticity ratio in a reasonable level. In this work, we presented four practical extensions in which two of them are novel approaches and the remaining ones are different from the previous implementations.
Karami, M.A.

Multi Power Multi Direction Kicking System


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 602-607 inproceedings

DOI URL

Abstract: In this paper a multi power multi direction kicking system is developed. This versatility is achieved without a need of either changing the power supply of robot or the direction of whole chassis of robot .The main factors for designing a suitable solenoid were studied and the casual agents to obtain maximum velocity of ball were introduced. A main circuit is used for getting different currents for solenoid to have different power of shooting. An arrangement of solenoids is introduced to have multi direction kicking system.
Karol, A. & Williams, M.-A.

Distributed Sensor Fusion for Object Tracking


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 504-511 inproceedings

DOI URL

Abstract: In a dynamic situation like robot soccer any individual player can only observe a limited portion of their environment at any given time. As such to develop strategies based upon planning and cooperation between different players it is imperative that they be able to share information which may or may not be in any individual player’s field of vision. In this paper we propose a method for multi-agent cooperation for perception based upon the Extended Kalman Filter (EKF) which enables players to track objects absent from their field of vision and also to improve the accuracy of position and velocity estimates of objects in their field of vision.
Kenn, H. & Pfeil, A.

Localizing Victims Through Sound and Probabilistic Grid Maps in an Urban Search and Rescue Scenario


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 312-322 inproceedings

DOI URL

Abstract: Sound source localization can be used in the Robocup Rescue Robots League as a sensor that is capable to autonomously detect victims that emit sound. Using differential time of flight measurements through energy cross-spectrum evaluation of the sound signals, the angular direction to multiple sound sources can be determined with a pair of microphones for SNRs better than -8dB. Assuming that the robot pose is known, this information is sufficient to create probabilistic occupancy grid map of the sound sources in the environment and thus localize the victims in a global map. This has been demonstrated using example measurements in an urban search and rescue scenario.
Ketill Gunnarsson, F.W. & Rojas, R.

The Color and the Shape: Automatic On-Line Color Calibration for Autonomous Robots, The


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 347-358 inproceedings

DOI URL

Abstract: This paper presents a method for automatic on-line color calibration of soccer-playing robots. Our method requires a geometrical model of the field-lines in world coordinates, and one of the ball in image coordinates. No specific assumptions are made about the color of the field, ball, or goals except that they are of roughly homogeneous distinct colors, and that the field-lines are bright relative to the field. The classification works by localizing the robot(without using color information), then growing homogeneously colored regions and matching their size and shape with those of the expected regions. If a region matches the expected one, its color is added to the respective color class. This method can be run in a background thread thus enabling the robot to quickly recalibrate in response to changes in illumination.
Kikuchi, T., Umeda, K., Ueda, R., Jitsukawa, Y., Osumi, H. & Arai, T.

Improvement of Color Recognition Using Colored Objects


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 537-544 inproceedings

DOI URL

Abstract: Recognition of colored objects is important and practical for robot vision. This paper proposes a color recognition method which is robust to illumination change. A color space of Cb/Y and Cr/Y is introduced where Cb and Cr are color difference and Y is intensity of YCbCr color space. This color space is not affected by the change of the brightness of illumination. And a method to update clusters of color table is proposed. The method can cope with the change of the color of illumination. Experiments show that the proposed method can recognize color more robustly for illumination change. RoboCup four legged robot league is chosen as the research platform.
Kleiner, A., Brenner, M., Bräuer, T., Dornhege, C., Göbelbecker, M., Luber, M., Prediger, J., Stückler, J. & Nebel, B.

Successful Search and Rescue in Simulated Disaster Areas


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 323-334 inproceedings

DOI URL

Abstract: RoboCupRescue Simulation is a large-scale multi-agent simulation of urban disasters where, in order to save lives and minimize damage, rescue teams must effectively cooperate despite sensing and communication limitations. This paper presents the comprehensive search and rescue approach of the ResQ Freiburg team, the winner in the RoboCupRescue Simulation league at RoboCup 2004.
Specific contributions include the predictions of travel costs and civilian life-time, the efficient coordination of an active disaster space exploration, as well as an any-time rescue sequence optimization based on a genetic algorithm.
We compare the performances of our team and others in terms of their capability of extinguishing fires, freeing roads from debris, disaster space exploration, and civilian rescue. The evaluation is carried out with information extracted from simulation log files gathered during RoboCup 2004. Our results clearly explain the success of our team, and also confirm the scientific approaches proposed in this paper.
Kok, J.R. & Vlassis, N.

Using the Max-Plus Algorithm for Multiagent Decision Making in Coordination Graphs


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 1-12 inproceedings

DOI URL

Abstract: Coordination graphs offer a tractable framework for cooperative multiagent decision making by decomposing the global payoff function into a sum of local terms. Each agent can in principle select an optimal individual action based on a variable elimination algorithm performed on this graph. This results in optimal behavior for the group, but its worst-case time complexity is exponential in the number of agents, and it can be slow in densely connected graphs. Moreover, variable elimination is not appropriate for real-time systems as it requires that the complete algorithm terminates before a solution can be reported. In this paper, we investigate the max-plus algorithm, an instance of the belief propagation algorithm in Bayesian networks, as an approximate alternative to variable elimination. In this method the agents exchange appropriate payoff messages over the coordination graph, and based on these messages compute their individual actions. We provide empirical evidence that this method converges to the optimal solution for tree-structured graphs (as shown by theory), and that it finds near optimal solutions in graphs with cycles, while being much faster than variable elimination.
Köse, H. & Akın, H.L.

A Fuzzy Touch to R-MCL Localization Algorithm


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 420-427 inproceedings

DOI URL

Abstract: In this work, a novel method called Fuzzy Reverse Monte Carlo Localization (Fuzzy R-MCL) for global localization of autonomous mobile agents in the robotic soccer domain is proposed to overcome the uncertainty in the sensors, environment and the motion model. R-MCL is a hybrid method based on both Markov Localization(ML) and Monte Carlo Localization(MCL) where the ML module finds the region where the robot should be and MCL predicts the geometrical location with high precision by selecting samples in this region. In this work, a fuzzy approach is embedded in this method, to improve flexibility, accuracy and robustness. In addition to using Fuzzy membership functions in modeling the uncertainty of the grid cells and samples, different heuristics are used to enable the adaptation of the method to different levels of noise and sparsity. The method is very robust and fast and requires less computational power and memory compared to similar approaches and is accurate enough for high level decision making which is vital for robot soccer.
Lauer, M., Lange, S. & Riedmiller, M.

Calculating the Perfect Match: An Efficient and Accurate Approach for Robot Self-Localization


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 142-153 inproceedings

DOI URL

Abstract: The paper develops a new approach for robot self-localization in the Robocup Midsize league. The approach is based on modeling the quality of an estimate using an error term and numerically minimizing it. Furthermore, we derive the reliability of the estimate analyzing the error function and apply the derived uncertainty value to a sensor integration process. The approach is characterized by high precision, robustness and computational efficiency.
Li, T.-H.S., Chang, C.-C., Ye, Y.-J. & Tasi, G.-R.

Autonomous Parking Control Design for Car-Like Mobile Robot by Using Ultrasonic and Infrared Sensors


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 472-479 inproceedings

DOI URL

Abstract: This research presents the design and implementation of the intelligent autonomous parking controller (APC) and accomplishes it in a car-like mobile robot (CLMR). This car possesses the function to accept and estimate the environment by integration of infrared and ultrasonic sensors. We propose five parking modes including parallel-parking mode, a narrow path parallel-parking mode, garage-parking mode, a narrow path garage-parking mode, and none parking mode. And the CLMR can autonomously determine which mode to use and park itself into the parking lot. Finally, it is perceived that our intelligent APC is feasible from the practical experiments.
Loncomilla, P. & del Solar, J.R.

Gaze Direction Determination of Opponents and Teammates in Robot Soccer


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 230-242 inproceedings

DOI URL

Abstract: Gaze direction determination of opponents and teammates is a very important ability for any soccer player, human or robot. However, this ability is still not developed in any of the RoboCup soccer leagues. We aim at reverting this situation by proposing a gaze direction determination system for robot soccer; the system is designed primarily for the four-legged league, but it could be extended to other leagues. This system is based on a robot-head pose detection system, consisting on two main processing stages: (i) computation of scale-invariant local descriptors of the observed scene, and (ii) matching of these descriptors against descriptors of robot-head prototypes already stored in a model database. After the robot-head pose is detected, the robot gaze direction is determined using a head model of the observed robot, and the current 3D position of the observing robot camera. Experimental results of the proposed approach are presented.
This research was funded by Millenium Nucleus Center for Web Research, Grant P04-067-F, Chile.
Lovell, N.

Illumination Independent Object Recognition


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 384-395 inproceedings

DOI URL

Abstract: Object recognition under uncontrolled illumination conditions remains one of hardest problems in machine vision. Under known lighting parameters, it is a simple task to calculate a transformation that maps sensed values to the expected colors in objects (and minimize the problems of reflections and/or texture). However, RoboCup aims to develop vision systems for natural lighting conditions in which the conditions are not only unknown but also dynamic. This makes fixed color-based image segmentation infeasible. We present a method for color determination under varying illumination conditions that succeeds in tracking the objects of interest in the RoboCup legged league.
Mantz, F., Jonker, P. & Caarls, W.

Behavior-Based Vision on a 4 Legged Soccer Robot


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 480-487 inproceedings

DOI URL

Abstract: In this paper the architecture of a 4 legged soccer robot is divided into a hierarchy of behaviors, where each behavior represents an independent sense-think-act loop. Based on this view we have implemented a behavior-based vision system, improving performance due to object-specific image processing,behavior-specific image processing and behavior-specific self localization. The system was tested under various lighting conditions, off-line using sets of images, and on-line in real tests for a robot in the role of goalkeeper. It appeared hat the performance of the goalie doubled, that it could play under a wider range of lighting and environmental conditions and used less CPU power.
Martín, F., Matellán, V., Cañas, J.M. & Barrera, P.

Visual Based Localization for a Legged Robot


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 708-715 inproceedings

DOI URL

Abstract: This paper presents a visual based localization mechanism for a legged robot in indoor office environments. Our proposal is a probabilistic approach which uses partially observable Markov decision processes. We use a precompiled topological map where natural landmarks like doors or ceiling lights are recognized by the robot using its on-board camera. Experiments have been conducted using the AIBO Sony robotic dog showing that it is able to deal with noisy sensors like vision and to approximate world models representing indoor office environments. The major contributions of this work is the use of an active vision as the main input and localization in not-engineered environments.
This work has been supported by grant DPI2004-07993-C03-01 of Spanish Government.
Mayer, N.M., Asada, M. & da Silva Guerra, R.

Using a Symmetric Rotor as a Tool for Balancing


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 71-80 inproceedings

DOI URL

Abstract: In the Humanoid Leagues balancing during walking and running is still the biggest challenge for most of the teams. We present here some work in which a dynamic walker is stabilised by using a fast heavy rotor, a gyro. The dynamics of a symmetric, fast rotating gyro is different from that of a non-rotating solid body, e.g. in the case of small disturbances it tends to keep the axes the same. Results show that the rotor enhances the stability of the walking in the simulations. In a model for an actuated robot the rotor is used as a reaction wheel, i.e. the pitch of the robot is stabilised by accelerating and decelerating it. We see this method –though it is not biologically inspired – as an intermediate step for learning balancing in biped robots. The control algorithm responsible for balancing the pitch is discussed in detail. The simulations show that, by using this kind of stabilisation, movements like stand up, walk and jump are easily possible by using open loop control for the legs, however high torques for the rotor are necessary. Finally, a robot design that consists just of a trunk is presented.
McKinnon, B., Baltes, J. & Anderson, J.

A Region-Based Approach to Stereo Matching for USAR


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 452-463 inproceedings

DOI URL

Abstract: Stereo vision for mobile robots is challenging, particularly when employing embedded systems with limited processing power. Objects in the field of vision must be extracted and represented in a fashion useful to the observer, while at the same time, methods must be in place for dealing with the large volume of data that stereo vision necessitates, in order that a practical frame rate may be obtained. We are working with stereo vision as the sole form of perception for Urban Search and Rescue (USAR) vehicles. This paper describes our procedure for extracting and matching object data using a stereo vision system. Initial results are provided to demonstrate the potential of this system for USAR and other challenging domains.
Nagasaka, Y., Saeki, M., Shibata, S., Fujiyoshi, H., Fujii, T. & Sakata, T.

A New Practice Course for Freshmen Using RoboCup Based Small Robots


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 428-435 inproceedings

DOI URL

Abstract: Contemporary engineers need to have the ability not only to freely make use of their professional knowledge and skills, but also to integrate and combine a wide range of knowledge and skills and to build a complex system to solve a problem. But the current educational programs of individual departments (mechanical engineering, electrical engineering, electronic engineering, computer science) are usually designed and performed independently. Therefore it is hard for students to understand how knowledge and technologies of each field are integrated and combined in the objects of the real world. In order to increase student understanding in this area, we propose a new practice course dealing with a completely functional object: a robot.
Nakashima, T., Takatani, M., Udo, M., Ishibuchi, H. & Nii, M.

Performance Evaluation of an Evolutionary Method for RoboCup Soccer Strategies


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 616-623 inproceedings

DOI URL

Abstract: This paper proposes an evolutionary method for acquiring team strategies of RoboCup soccer agents. The action of an agent in a subspace is specified by a set of action rules. The antecedent part of action rules includes the position of the agent and the distance to the nearest opponent. The consequent part indicates the action that the agent takes when the antecedent part of the action rule is satisfied. The action of each agent is encoded into an integer string that represents the action rules. A chromosome is the concatenated string of integer strings for all agents. We employ an ES-type generation update scheme after producing new integer strings by using crossover and mutation. Through computer simulations, we show the effectiveness of the proposed method.
Nia, H.T., Alemohammad, S.H., Bagheri, S., Khiabani, R.H. & Meghdari, A.

Design, Dynamic Analysis and Optimization of a Rover for Rescue Operations


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 290-300 inproceedings

DOI URL

Abstract: In this paper a new approach to dynamic optimization of a rough terrain rover is introduced. Since rover wheels traction has a significant role in rover mobility, optimization is based on the minimization of traction at rover wheel-ground interfaces. The method of optimization chosen is Genetic Algorithm (GA) which is a directed random search technique along with the usual optimization based on directional derivatives. GA is a suitable and efficient method of optimization for nonlinear problems. The procedure is applied on a specific rough terrain rover called CEDRA-I Shrimp Rover. The present work resulted in design and manufacturing of the optimized rover called CEDRA-II Shrimp Rover.
Nisticò, W. & Röfer, T.

Improving Percept Reliability in the Sony Four-Legged Robot League


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 545-552 inproceedings

DOI URL

Abstract: This paper presents selected methods used by the vision system of the GermanTeam, the World Champion in the Sony Four-Legged League in 2004. Color table generalization is introduced as a means to achieve a larger independence of the lighting situation. Camera calibration is necessary to deal with the weaknesses of the new platform used in the league, the Sony Aibo ERS-7. Since the robot camera uses a rolling shutter, motion compensation is required to improve the information extracted from the camera images.
The Deutsche Forschungsgemeinschaft supports this work through the priority program “Cooperating teams of mobile robots in dynamic environments”.
Nitschke, G.

Emergent Cooperation in RoboCup: A Review


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 512-520 inproceedings

DOI URL

Abstract: This article presents a survey of prevalent research results pertaining to emergent cooperation in RoboCup soccer. Results reviewed maintain particular reference to research that uses biologically inspired design principles and concepts, such as emergence and evolution, as a means of attaining cooperative behavior. The core of this article argues that even though emergent cooperative behavior derived within RoboCup (and the larger field of multi-robot systems) is still in its infancy, it holds considerable future potential, as a problem solver in domains where systems comprising many interacting components must cooperatively solve a global task.
Noda, I.

Consistency Management Framework for Database Used in Integrated Simulations


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 279-289 inproceedings

DOI URL

Abstract: In this article, I propose a mathematical framework of consistency management that guarantees validity among data that are used in integrated simulation systems. When we apply integrated simulations to real-time prediction/evaluation of complex social phenomena like disaster and rescue, checking and keeping consistency of data is an important issue to validate simulation results, because multiple and delayed information are reported to the database continuously in real-time applications. The proposed formalization gives fundamental background of consistency and validity of database and simulation. I also investigate about cost of the management in two major implementation styles.
Nüchter, A., Wulf, O., Lingemann, K., Hertzberg, J., Wagner, B. & Surmann, H.

3D Mapping with Semantic Knowledge


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 335-346 inproceedings

DOI URL

Abstract: A basic task of rescue robot systems is mapping of the environment. Localizing injured persons, guiding rescue workers and excavation equipment requires a precise 3D map of the environment. This paper presents a new 3D laser range finder and novel scan matching method for the robot Kurt3D [9]. Compared to previous machinery [12], the apex angle is enlarged to 360°. The matching is based on semantic information. Surface attributes are extracted and incorporated in a forest of search trees in order to associate the data, i.e., to establish correspondences. The new approach results in advances in speed and reliability.
Obst, O. & Boedecker, J.

Flexible Coordination of Multiagent Team Behavior Using HTN Planning


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 521-528 inproceedings

DOI URL

Abstract: The domain of robotic soccer is known as a highly dynamic and non-deterministic environment for multiagent research. We introduce an approach using Hierarchical Task Network planning in each of the agents for high-level coordination and description of team strategies. Our approach facilitates the maintenance of expert knowledge specified as team strategies separated from the agent implementation. By combining high level plans with reactive basic operators, agents can pursue a grand strategy while staying reactive to changes in the environment. Our results show that the use of a planner in a multiagent system is both possible and useful despite the constraints in dynamic environments.
Oubbati, M., Schanz, M., Buchheim, T. & Levi, P.

Velocity Control of an Omnidirectional RoboCup Player with Recurrent Neural Networks


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 691-701 inproceedings

DOI URL

Abstract: In this paper, a recurrent neural network is used to develop a dynamic controller for mobile robots. The advantage of the control approach is that no knowledge about the robot model is required. This property is very useful in practical situations, where the exact knowledge about the robot parameters is almost unattainable. The proposed approach has been experimentally tested on an Omnidirectional RoboCup Player available at the Robotics Lab of the University of Stuttgart.
This work was partially supported by the German Academic Exchange Service (DAAD).
Paquet, S., Tobin, L. & draa, B.C.

An Online POMDP Algorithm Used by the PoliceForce Agents in the RoboCupRescue Simulation


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 196-207 inproceedings

DOI URL

Abstract: In the RoboCupRescue simulation, the PoliceForce agents have to decide which roads to clear to help other agents to navigate in the city. In this article, we present how we have modelled their environment as a POMDP and more importantly we present our new online POMDP algorithm enabling them to make good decisions in real-time during the simulation. Our algorithm is based on a look-ahead search to find the best action to execute at each cycle. We thus avoid the overwhelming complexity of computing a policy for each possible situation. To show the efficiency of our algorithm, we present some results on standard POMDPs and in the RoboCupRescue simulation environment.
Pedreiras, P., Teixeira, F., Ferreira, N., Almeida, L., Pinho, A. & Santos, F.

Enhancing the Reactivity of the Vision Subsystem in Autonomous Mobile Robots Using Real-Time Techniques


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 371-383 inproceedings

DOI URL

Abstract: Interest on using mobile autonomous agents has been growing, recently, due to their capacity to cooperate for diverse purposes, from rescue to demining and security. In many of these applications the environments are inherently unstructured and dynamic, requiring substantial computation resources for gathering enough sensory input data to allow a safe navigation and interaction with the environment. As with humans, who depend heavily on vision for these purposes, mobile robots employ vision frequently as the primary source of input data when operating in such environments. However, vision-based algorithms are seldom developed with reactive and real-time concerns, exhibiting large variations in the execution time and leading to occasional periods of black-out or vacant input data. This paper addresses this problem in the scope of the CAMBADA robotic soccer team developed at the University of Aveiro, Portugal. It presents an evolution from a monolithic to a modular architecture for the vision system that improves its reactivity. With the proposed architecture it is possible to track different objects with different rates without losing any frames.
This work was partially supported by the European Community through the ARTIST2 NoE (IST-004527) and by the Portuguese Government through the project FCT-POSI/ROBO/43908/2002, also partially funded by FEDER.
Penta, R.S. & Karlapalem, K.

Agent Community Extraction for 2D-RoboSoccer


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 184-195 inproceedings

DOI URL

Abstract: Agents perform tasks to maximize their benefits. There are several instances where the agent can not perform a task individually. In these situations, agents need to cooperate and coordinate with other agents effectively and efficiently to maximize their benefits in a limited time. In several domains, we can analyze the behavior of successful agents and the way they interact with other agents forming strong communities or coalitions. This knowledge can be used by a new or unsuccessful agent to collaborate with other agents that gives maximum benefit under strict time constraints. This paper proposes a generic procedure for extracting these hidden communities that can be used by the agents in a productive manner. We tested the framework on robosoccer simulation environment and our experiments indeed show drastic increase in both agent and team performance.
Qiu, X.-J., Wang, Z.-Q., Xia, S.-H. & Sun, Y.-C.

Inferring 3D Body Pose from Uncalibrated Video


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 553-560 inproceedings

DOI URL

Abstract: Recovery of 3D body pose is a fundamental problem for human motion analysis in many applications such as motion capture, vision interface, visual surveillance, and gesture recognition. In this paper, we present a new image-based approach to infer 3D human structure parameters from uncalibrated video. The estimation is example based. First, we acquire a special motion database through an off-line motion capture process. Second, given uncalibrated motion video, we abstract the extrinsic parameters and then silhouettes database associated with 3D poses is built by projecting each data of the 3D motion database into 2D plane with the extrinsic parameters. Next, with the image silhouettes abstracted from video, the unknown structure parameters are inferred by performs a similarity search in the database of silhouettes using approach based on shape matching. That is, the 3D structure parameters whose 2D projective silhouette is the most similar to the 2D image silhouette are took as the 3D reconstruction structure. We use trampoline sport motion, an example of complex human motion, to demonstrate the effectiveness of our approach.
Rooker, M.N. & Birk, A.

Communicative Exploration with Robot Packs


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 267-278 inproceedings

DOI URL

Abstract: Exploration is a core challenge for RoboCup Rescue. So-called communicative exploration is a novel strategy for multi-robot exploration that unlike other approaches takes the limits of wireless communication systems into account. Here, previous results that where achieved for a team of robots linked to a basestation are significantly extended to also cover robot packs, i.e., multi-robot teams that are not permanently tied to an operator’s station. Unlike teams that are constrained by the immobility of a basestation, packs can explore arbitrarily large regions. Results from experiments with packs of 4, 5 and 6 robots are presented. The new strategy constantly maintains the communication between the robots while exploring, whereas the commonly used frontier-based exploration strategy, which is used in the experiments as comparison to our approach, leads to a rapid loss of communication.
Röfer, T., Laue, T. & Thomas, D.

Particle-Filter-Based Self-Localization Using Landmarks and Directed Lines


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 608-615 inproceedings

DOI URL

Abstract: The paper presents the self-localization approach used by the World Champion in the Sony Four-Legged Robot League 2004. The method is based on a particle filter that makes use of different features from the environment (beacons, goals, field lines, field wall) that provide different kinds of localization information and that are recognized with different frequencies. In addition, it is discussed how the vision system acquires these features, especially, how the orientation of field lines is determined at low computational costs.
The Deutsche Forschungsgemeinschaft supports this work through the priority program “Cooperating teams of mobile robots in dynamic environments”.
Sedaghat, M.N., Nejad, L.P., Iravanian, S. & Rafiee, E.

Task Allocation for the Police Force Agents in RoboCupRescue Simulation


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 656-664 inproceedings

DOI URL

Abstract: In this paper, three mechanisms for task allocation among police force agents in the rescue simulation environment are presented. Three different approaches namely full auction-based, partitioning-based and hybrid approaches are briefly described. The empirical results of using the hybrid approach show a significant improvement in performance over the other two approaches. By using the hybrid mechanism for the police forces, our agents together with other types agents ranked third in the RoboCupRescue 2004 simulation competitions.
Shimizu, S., Nagahashi, T. & Fujiyoshi, H.

Robust and Accurate Detection of Object Orientation and ID Without Color Segmentation


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 408-419 inproceedings

DOI URL

Abstract: This paper describes a novel approach to detecting orientation and identity of robots without color segmentation. The continuous DP matching calculates the similarity between the reference pattern and the input pattern by matching the intensity changes of the robot markers. After the continuous DP matching, a similarity value is used for object identification. Correspondences of the optimal route obtained by back tracing are used for estimating the robot’s orientation. This method archives orientation estimations of less than 1 degree and robustness with respect to varying light conditions.
Siebra, C. & Tate, A.

Integrating Collaboration and Activity-Oriented Planning for Coalition Operations Support


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 561-568 inproceedings

DOI URL

Abstract: The use of planning assistant agents is an appropriate option to provide support for members of a coalition. Planning agents can extend the human abilities and be customised to attend different kinds of activities. However, the implementation of a planning framework must also consider other important requirements for coalitions, such as the performance of collaborative activities and human-agent interaction (HAI). This paper discusses the integration of an activity-oriented planning with collaborative concepts using a constraint-based ontology for that. While the use of collaborative concepts provides a better performance to systems as a whole, a unified representation of planning and collaboration enables an easy customisation of activity handlers and the basis for a future incorporation of HAI mechanisms.
Skinner, C. & Barley, M.

Robocup Rescue Simulation Competition: Status Report


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 632-639 inproceedings

DOI URL

Abstract: This is the fifth anniversary of the Robocup Rescue Simulation Competitions and the tenth anniversary of the disaster that inspired the Competitions. This is a good time to take stock of what milestones have been achieved and what milestones we should be aiming for. Specifically, this paper looks at the goals that led to the establishment of the competition, the current status of the simulation platform and infrastructure, and finally suggests areas of the current simulation platform which should be improved and parts of the Robocup Rescue technical and social infrastructure which should be extended.
Sridharan, M. & Stone, P.

Towards Eliminating Manual Color Calibration at RoboCup


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 673-681 inproceedings

DOI URL

Abstract: Color calibration is a time-consuming, and therefore costly requirement for most robot teams at RoboCup. This paper presents an approach for autonomous color learning on-board a mobile robot with limited computational and memory resources. It works without any labeled training data and trains autonomously from a color-coded map of its environment. The process is fully implemented, completely autonomous, and provides high degree of segmentation accuracy. Most importantly, it dramatically reduces the time needed to train a color map in a new environment.
Stanton, C. & Williams, M.-A.

A Novel and Practical Approach Towards Color Constancy for Mobile Robots Using Overlapping Color Space Signatures


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 444-451 inproceedings

DOI URL

Abstract: Color constancy is the ability to correctly perceive an object’s color regardless of illumination. Within the controlled, color-coded environments in which many robots operate (such as RoboCup), engineers have been able to avoid the color constancy problem by using straightforward mappings of pixel values to symbolic colors. However, for robots to perform color vision tasks under natural light the color constancy problem must be addressed. We have developed a color vision system which allows for the color space signatures of different symbolic colors to overlap. This raises the question: if a specific pixel value can be mapped to multiple symbolic colors, how does the robot determine which color is the “correct” one? Context plays an important role. We adopt a knowledge driven approach which allows the robot to reason about uncertain color values. The system is fully implemented on a Sony AIBO.
Steinbauer, G., Mörth, M. & Wotawa, F.

Real-Time Diagnosis and Repair of Faults of Robot Control Software


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 13-23 inproceedings

DOI URL

Abstract: Faults in hardware and software are not totally avoidable not even if the components are carefully designed, implemented and tested. In this paper we present a solution for detection, localization and repair of faults in the control software for autonomous mobile robots. The presented diagnosis system uses model-based diagnosis for fault detection and localization. Furthermore, we present a method which enables the robot control software to recover from located faults. The novelty of our approach is that fault localization and repair takes place at runtime. Moreover, we present experimental results of the proposed diagnosis system obtained in the RoboCup Middle-Size scenario.
This research has been funded in part by the Austrian Science Fund (FWF) under grant P17963-N04 and Land Steiermark under grant 40Ro03-PE ”RoboCup 2004”.
Stolarski, D.

Conceptual Representation for a Soccer Commentary Generator


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 496-503 inproceedings

DOI URL

Abstract: The issue of this paper is bound with the project of building a commentary system for a soccer game. We surveyed an example of natural commentary in order to examine various sorts of comments used by a human reporter to cover the soccer match. The content and function of the messages was taken into account. This survey enabled us to create a conceptual representation for the RoboCup soccer game, which constitutes the input for our commentary generation system. We also give some details about how the representation is created from the raw quantitative data. The presented results may be interesting for human-machine communication researchers and software engineers involved in a similar project.
Stone, P., Kuhlmann, G., Taylor, M.E. & Liu, Y.

Keepaway Soccer: From Machine Learning Testbed to Benchmark


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 93-105 inproceedings

DOI URL

Abstract: Keepaway soccer has been previously put forth as a testbed for machine learning. Although multiple researchers have used it successfully for machine learning experiments, doing so has required a good deal of domain expertise. This paper introduces a set of programs, tools, and resources designed to make the domain easily usable for experimentation without any prior knowledge of RoboCup or the Soccer Server. In addition, we report on new experiments in the Keepaway domain, along with performance results designed to be directly comparable with future experimental results. Combined, the new infrastructure and our concrete demonstration of its use in comparative experiments elevate the domain to a machine learning benchmark, suitable for use by researchers across the field.
Strack, A., Ferrein, A. & Lakemeyer, G.

Laser-Based Localization with Sparse Landmarks


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 569-576 inproceedings

DOI URL

Abstract: Self-localization in dynamic environments is a central problem in mobile robotics and is well studied in the literature. One of the most popular methods is the Monte Carlo Localization algorithm (MCL). Many deployed systems use MCL together with a laser range finder in well structured indoor environments like office buildings with a rather rich collection of landmarks. In symmetric environments like robotic soccer with sparse landmarks which are occluded by other robots, most of the time the standard method does not yield satisfying results. In this paper we propose a new heuristic weight function to integrate a full 360° sweep from a laser range finder and introduce so-called don’t-care regions which allow to ignore some parts of the environment. The proposed method is not specific to robotic soccer and scales very well outside the soccer domain.
This work was supported by the German National Science Foundation (DFG) in the Priority Program 1125, Cooperating Teams of Mobile Robots in Dynamic Environments and by a grant of the Ministry of Science and Research of North Rhine-Westphalia, Germany.
Takahashi, Y., Edazawa, K., Noma, K. & Asada, M.

Simultaneous Learning to Acquire Competitive Behaviors in Multi-Agent System Based on Modular Learning System


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 243-254 inproceedings

DOI URL

Abstract: The existing reinforcement learning approaches have been suffering from the policy alternation of others in multiagent dynamic environments. A typical example is a case of RoboCup competitions since other agent behaviors may cause sudden changes in state transition probabilities of which constancy is needed for the learning to converge. The keys for simultaneous learning to acquire competitive behaviors in such an environment are
– a modular learning system for adaptation to the policy alternation of others, and
– an introduction of macro actions for simultaneous learning to reduce the search space.
This paper presents a method of modular learning in a multiagent environment, by which the learning agents can simultaneously learn their behaviors and adapt themselves to the situations as consequences of the others’ behaviors.
Takahashi, Y., Nishi, T. & Asada, M.

Self Task Decomposition for Modular Learning System Through Interpretation of Instruction by Coach


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 640-647 inproceedings

DOI URL

Abstract: One of the most formidable issues of RL application to real robot tasks is how to find a suitable state space, and this has been much more serious since recent robots tends to have more sensors and the environment including other robots becomes more complicated. In order to cope with the issue, this paper presents a method of self task decomposition for modular learning system based on self-interpretation of instructions given by a coach. The proposed method is applied to a simple soccer situation in the context of RoboCup.
Thianwiboon, M. & Sangveraphunsiri, V.

Traction Control for a Rocker-Bogie Robot with Wheel-Ground Contact Angle Estimation


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 682-690 inproceedings

DOI URL

Abstract: A method for kinematics modeling of a six-wheel Rocker-Bogie mobile robot is described in detail. The forward kinematics is derived by using wheel Jacobian matrices in conjunction with wheel-ground contact angle estimation. The inverse kinematics is to obtain the wheel velocities and steering angles from the desired forward velocity and turning rate of the robot. Traction Control is also developed to improve traction by comparing information from onboard sensors and wheel velocities to minimize wheel slip. Finally, a simulation of a small robot using rocker-bogie suspension has been performed and simulate in two conditions of surfaces including climbing slope and travel over a ditch.
Tim Laue, K.S. & Röfer, T.

SimRobot - A General Physical Robot Simulator and Its Application in RoboCup


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 173-183 inproceedings

DOI URL

Abstract: This paper describes SimRobot, a robot simulator which is able to simulate arbitrary user-defined robots in three-dimensional space. It includes a physical model which is based on rigid body dynamics. To allow an extensive flexibility in building accurate models, a variety of different generic bodies, sensors and actuators has been implemented. Furthermore, the simulator follows an user-oriented approach by including several mechanisms for visualization, direct actuator manipulation, and interaction with the simulated world. To demonstrate the general approach, this paper presents multiple examples of different robots which have been simulated so far.
The Deutsche Forschungsgemeinschaft supports this work through the priority program “Cooperating teams of mobile robots in dynamic environments”.
Wisspeintner, T., Nowak, W. & Bredenfeld, A.

VolksBot - A Flexible Component-Based Mobile Robot System


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 716-723 inproceedings

DOI URL

Abstract: In this paper we present a component-based framework for rapid prototyping of mobile robots for research, education and application. The VolksBot construction kit addresses the rising demand for reusability in software, electronic hardware and mechanics by offering open and clearly defined interfaces as well as standardized components in all three fields. We show the versatility of the concept by applying it to different domains, particularly RoboCup Middle Size League as well as the Rescue scenario.
Zagal, J.C., Sarmiento, I. & del Solar, J.R.

An Application Interface for UCHILSIM and the Arrival of New Challenges


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 464-471 inproceedings

DOI URL

Abstract: UCHILSIM is a robot simulator recently introduced in the RoboCup Four Legged League. A main attractive of the simulator is the possibility of reproducing with accuracy the dynamical behavior of AIBO robots as well as providing good graphical representations of their surroundings on a soccer scenario. Learning over virtual environments can be performed with successful transfers of resulting behaviors to real environments. Previous version of the simulator had a major drawback: Only the UChile1 team could make use of it since the developed system had high dependency on the team code. In this paper we present results of a development work which was envisioned on the first presentation of UCHILSIM; an application interface for allowing any OPEN-R software code to be directly used over the UCHILSIM simulator. The possibility of having this kind of tool opens a great field of developments and challenges since more people will develop OPEN-R software, even without having the robotic hardware but the simulator. Other recent improvements on our simulator are briefly presented here as well.
van der Zant, T. & Plöger, P.G.

Lightweight Management - Taming the RoboCup Development Process


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 577-584 inproceedings

DOI URL

Abstract: RoboCup projects can face a lack of progress and continuity. The teams change continuously and knowledge gets lost. The approach used in previous years is no longer valid due to rule changes and specialists leaving the team leave black boxes that no-one understands. This article presents the application of a recent software development technique called eXtreme Programming to the realm of RoboCup. Many common problems typical for teams of students seem to be solvable with this technique. It also gradually spreads out in professional software production companies. Students mastering it are of high use for their further career after having left the university. The strategy is being tested on a real RoboCup Mid-Size and an Aibo league project and produces very promising results. The approach makes it possible to modularize scientific knowledge into software that can be re-used. Both the scientist/expert, who has the knowledge, and the software development team benefit from this approach without much overhead on the project.
van der Zant, T. & Wisspeintner, T.

RoboCup X: A Proposal for a New League Where RoboCup Goes Real World


2006 RoboCup 2005: Robot Soccer World Cup IX, pp. 166-172 inproceedings

DOI URL

Abstract: To put more emphasis on real-world problems, the authors propose to extend the RoboCup competitions. In order to foster progress in the desired abilities the authors propose to expand the existing challenges by a set of simple tests. The passing of the entire set should lead to robots that are capable of working both autonomously and in cooperation with humans in different realistic scenarios. Robots from all RoboCup leagues but also from outside of RoboCup should be allowed to participate. The new league especially aims at fostering the development of practical solutions and applications for supporting humans in everyday life.