RoboCup 2010 Publications

From RoboCup Wiki
Jump to: navigation, search
Author Title Year Journal/Proceedings Reftype DOI/URL
de Assis Moura Pimentel, F., Frias, D., Simões, M. & de Souza, J.R.

MR-Simulator: A Simulator for the Mixed Reality Competitions of RoboCup

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 82-96 inproceedings


Abstract: Simulations play an important roll in the development of intelligent and collaborative robots. In this paper we describe the development of a simulator for the RoboCup Mixed Reality (MR) competition. MR uses a mix of simulated and real world to foster intelligent robotics development. Our simulator ”virtualizes” the players within the MR hardware and software framework, providing the game server with the player-positional information usually supplied by the central vision system. To do that the simulator, after receiving the action commands from each agent (player) must track its expected trajectory and behavior in collision events for all the players. We developed fast algorithms for simulating the robot motion in most usual cases. The simulator was validated comparing its results against the real environment and proved to be realistic. This tool is important for setting-up, training and testing MR competition teams and could help to overcome current difficulties with robot hardware acquisition and maintenance.
Burchardt, A., Laue, T. & Röfer, T.

Optimizing Particle Filter Parameters for Self-Localization

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 145-156 inproceedings


Abstract: Particle filter-based approaches have proven to be capable of efficiently solving the self-localization problem in RoboCup scenarios and are therefore applied by many participating teams. Nevertheless, they require a proper parametrization - for sensor models and dynamic models as well as for the configuration of the algorithm - to operate reliably. In this paper, we present an approach for optimizing all relevant parameters by using the Particle Swarm Optimization algorithm. The approach has been applied to the self-localization component of a Standard Platform League team and shown to be capable of finding a parameter set that leads to more precise position estimates than the previously used hand-tuned parametrization.
Correa, M., del Solar, J.R., Parra-Tsunekawa, I. & Verschae, R.

A Realistic Simulation Tool for Testing Face Recognition Systems under Real-World Conditions

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 13-24 inproceedings


Abstract: In this article, a tool for testing face recognition systems under uncontrolled conditions is proposed. The key elements of this tool are a simulator and real face and background images taken under real-world conditions with different acquisition angles. Inside the simulated environment, an observing agent, the one with the ability to recognize faces, can navigate and observe the real face images, at different distances, angles and with indoor or outdoor illumination. During the face recognition process, the agent can actively change its viewpoint and relative distance to the faces in order to improve the recognition results. The simulation tool provides all functionalities to the agent (navigation, positioning, face's image composing under different angles, etc.), except the ones related with the recognition of faces. This tool could be of high interest for HRI applications related with the visual recognition of humans, as the ones included in the RoboCup @Home league. It allows comparing and quantifying the face recognition capabilities of service robots under exactly equal working conditions. It could be a complement to existing tests in the RoboCup @Home league. The applicability of the proposed tool is validated in the comparison of three state of the art face recognition methods.
Czarnetzki, S., Kerner, S. & Kruse, M.

Real-time Active Vision by Entropy Minimization Applied to Localization

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 266-277 inproceedings


Abstract: This paper presents an active vision approach to enhance mobile robot localization. A particle filter localization is extended with a module to find active vision decisions that are optimal based on the current localization and its uncertainty. Optimality is expressed as a criterion of entropy minimization. Further approximations are introduced to enable real-time computation. Both the usefulness of the presented approach in a RoboCup scenario and the performance and quality of the approximations are evaluated in different static and dynamic situations.
Danis, F.S., Meriçli, T., Meriçli, Ç. & Akın, H.L.

Robot Detection with a Cascade of Boosted Classifiers Based on Haar-like Features

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 409-417 inproceedings


Abstract: Accurate world modeling is important for efficient multi-robot planning in robot soccer. Visual detection of the robots on the field in addition to all other objects of interest is crucial to achieve this goal. The problem of robot detection gets even harder when robots with only on board sensing capabilities, limited field of view, and restricted processing power are used. This work extends the real-time object detection framework proposed by Viola and Jones, and utilizes the unique chest and head patterns of Nao humanoid robots to detect them in the image. Experiments demonstrate rapid detection with an acceptably low false positive rate, which makes the method applicable for real-time use.
Fabio DallaLibera, Shuhei Ikemoto, T.M.H.I.E.P. & Menegatti, E.

Biologically Inspired Mobile Robot Control Robust to Hardware Failures and Sensor Noise

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 218-229 inproceedings


Abstract: Some bacteria present a movement which can be modeled as a biased random walk. Biased random walk can be used also for artificial creatures as a very simple and robust control policy for tasks like goal reaching. In this paper, we show how a very simple control law based on random walk is able to guide mobile robot equipped with an omnidirectional camera toward a target without any knowledge about the robot's actuators or about the robot's camera parameters. We verified, by several simulation experiments, the robustness of the random biased control law with respect to failures of robot's actuators or sensor damages. These damages are similar to the ones which can occur during a RoboCup match. The tests show that the optimal behavior is obtained using a bias which is roughly proportional to the random walk step, with a coefficient dependent on the physical structure of the robot, on its actuators and on and its sensors after the damage. Finally, we validated the proposed approach with experiments in the real world with a wheeled robot performing a goal reaching task in a Middle-Size RoboCup field without any prior knowledge on the actuators and without any calibration of the very noisy omnidirectional camera mounted on the robot.
Formsma, O., Dijkshoorn, N., van Noort, S. & Visser, A.

Realistic Simulation of Laser Range Finder Behavior in a Smoky Environment

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 336-349 inproceedings


Abstract: The Urban Search and Rescue Simulation used for RoboCup lacks realistic response of laser range finders on smoke. In this paper, the behavior of a Hokuyo and Sick laser range finder in a smoky environment is studied. The behavior of the lasers is among others a function of the visibility level, and in this article this function is quantified into an explicit model. This model is implemented in a simulation environment which is the basis of the Virtual Robot competition of the RoboCup Rescue League. The behavior of both real and virtual laser range finders is compared in a number of validation tests. The validation tests show that the behavior of the laser range finders in the simulation is consistent with the real world.
Gabel, T. & Riedmiller, M.

On Progress in RoboCup: The Simulation League Showcase

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 36-47 inproceedings


Abstract: The annual RoboCup competitions are certainly relevant to present times as they elect the best team of the current year. With respect to RoboCup's well-known 2050 vision, however, it is crucial to assess the general progress being made not just qualitatively, but also in a quantitative manner. Although this is difficult to accomplish in most leagues, the recent development and circumstances in the soccer simulation league led to a unique situation which allowed us to perform an extensive experimental evaluation by means of which we can empirically measure the progress of playing performance made over a number of successive years. The findings we present in this paper, in particular the fact that significant improvements in the level of play can be verified quantitatively in the 2D soccer simulation league, may serve as a showcase for the progress made by the RoboCup initiative in general.
Gokce, B. & Akın, H.L.

Parameter Optimization for a Signal-Based Omni-Directional Biped Locomotion by Using Evolutionary Strategies

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 362-373 inproceedings


Abstract: The ultimate goal of RoboCup depends heavily on the advances in the development of humanoid robots. Flexible walking is a crucial part of playing soccer and bipedal walking has been a very active research topic in RoboCup. In this paper a signal-based omnidirectional walking algorithm for the Aldebaran Nao humanoid robot is presented. Inspired from the existing methods in the literature, the proposed method models the omni-directional motion as the combination of a set of periodic signals. The parameters controlling the characteristics of the signals are encoded into genes and Evolutionary Strategies is used to learn an optimal set of parameters.
Gossow, D., Decker, P. & Paulus, D.

An Evaluation of Open Source SURF Implementations

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 169-179 inproceedings


Abstract: SURF (Speeded Up Robust Features) is a detector and descriptor of local scale- and rotation-invariant image features. By using integral images for image convolutions it is faster to compute than other state-of-the-art algorithms, yet produces comparable or even better results by means of repeatability, distinctiveness and robustness. A library implementing SURF is provided by the authors. However, it is closed-source and thus not suited as a basis for further research.
Several open source implementations of the algorithm exist, yet it is unclear how well they realize the original algorithm. We have evaluated different SURF implementations written in C++ and compared the results to the original implementation.
We have found that some implementations produce up to 33% lower repeatability and up to 44% lower maximum recall than the original implementation, while the implementation provided with the software Pan-o-matic produced almost identical results.
We have extended the Pan-o-matic implementation to use multi-threading, resulting in an up to 5.1 times faster computation on an 8-core machine. We describe our comparison criteria and our ideas that lead to the speed-up. Our software is put into the public domain.
Hausknecht, M. & Stone, P.

Learning Powerful Kicks on the Aibo ERS-7: The Quest for a Striker

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 254-265 inproceedings


Abstract: Coordinating complex motion sequences remains a challenging task for robotics. Machine Learning has aided this process, successfully improving motion sequences such as walking and grasping. However, to the best of our knowledge, outside of simulation, learning has never been applied to the task of kicking the ball. We apply machine learning methods to optimize kick power entirely on a real robot. The resulting learned kick is significantly more powerful than the most powerful hand-coded kick of one of the most successful RoboCup four-legged league teams, and is learned in a principled manner which requires very little engineering of the parameter space. Finally, model inversion is applied to the problem of creating a parameterized kick capable of kicking the ball a specified distance.
Hermosilla, G., Loncomilla, P. & del Solar, J.R.

Thermal Face Recognition using Local Interest Points and Descriptors for HRI Applications

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 25-35 inproceedings


Abstract: In this article a robust thermal face recognition methodology based on the use of local interest points and descriptors, is proposed. The methodology consists of the following stages: face segmentation, vascular network detection, wide baseline matching using local interest points and descriptors, and classification. The main contribution of this work is the use of a standard wide baseline matching methodology for the comparison of vascular networks from thermal face images. The proposed methodology is validated using a database of thermal images. This work could be of high interest for HRI applications related with the visual recognition of humans, as the ones included in the RoboCup @Home league, because the use of thermal images may overcome limitations such as dependency on illumination conditions and facial expressions.
Holz, D., Schnabel, R., Droeschel, D., Stückler, J. & Behnke, S.

Towards Semantic Scene Analysis with 3D Time-of-Flight Cameras

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 121-132 inproceedings


Abstract: For planning grasps and other object manipulation actions in complex environments, 3D semantic information becomes crucial. This paper focuses on the application of recent 3D Time-of-Flight (ToF) cameras in the context of semantic scene analysis. For being able to acquire semantic information from ToF camera data, we a) pre-process the data including outlier removal, filtering and phase unwrapping for correcting erroneous distance measurements, and b) apply a randomized algorithm for detecting shapes such as planes, spheres, and cylinders. We present experimental results that show that the robustness against noise and outliers of the underlying RANSAC paradigm allows for segmenting and classifying objects in 3D ToF camera data captured in natural mobile manipulation setups.
Kerner, S., Czarnetzki, S. & Hegele, M.

Odometry Correction for Humanoid Robots Using Optical Sensors

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 48-59 inproceedings


Abstract: Odometry measurement is an important concept to update localization information, but is prone to error propagation. Still the method is widely applied to wheeled mobile robots since their motion is quite robust to random error such as slipping. While the concept of odometry can also be applied to humanoid robots the dynamically stable walking generation reinforces sliding motions resulting in unpredictable errors. Therefore this paper proposes a novel approach to measure these sliding errors with the help of optical sensors to either correct the odometry update or perform suitable actions to counteract the error.
Leonetti, M. & Iocchi, L.

LearnPNP: A Tool for Learning Agent Behaviors

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 418-429 inproceedings


Abstract: High-level programming of robotic agents requires the use of a representation formalism able to cope with several sources of complexity (e.g. parallel execution, partial observability, exogenous events, etc.) and the ability of the designer to model the domain in a precise way. Reinforcement Learning has proved promising in improving the performance, adaptability and robustness of plans in under-specified domains, although it does not scale well with the complexity of common robotic applications. In this paper we propose to combine an extremely expressive plan representation formalism (Petri Net Plans), with Reinforcement Learning over a stochastic process derived directly from such a plan. The derived process has a significantly reduced search space and thus the framework scales well with the complexity of the domain and allows for actually improving the performance of complex behaviors from experience. To prove the effectiveness of the system, we show how to model and learn the behavior of the robotic agents in the context of Keepaway Soccer (a widely accepted benchmark for RL) and the RoboCup Standard Platform League.
Lu, H., Zhang, H. & Zheng, Z.

A Novel Real-Time Local Visual Feature for Omnidirectional Vision Based on FAST and LBP

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 291-302 inproceedings


Abstract: A novel real-time local visual feature, namely FAST+LBP, is proposed in this paper for omnidirectional vision. It combines the advantages of two computationally simple operators by using Features from Accelerated Segment Test (FAST) as the feature detector and Local Binary Patterns (LBP) operator as the feature descriptor. The matching experiments of the panoramic images from the COLD database are performed to determine its best parameters, and to evaluate and compare its performance with SIFT. The experimental results show that our algorithm performs better, and features can be extracted in real-time.
Meriçli, Ç. & Veloso, M.

Improving Biped Walk Stability Using Real-time Corrective Human Feedback

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 194-205 inproceedings


Abstract: Robust walking is one of the key requirements for soccer playing humanoid robots. Developing such a biped walk algorithm is non-trivial due to the complex dynamics of the walk process. In this paper, we first present a method for learning a corrective closed-loop policy to improve the walk stability for the Aldebaran Nao robot using real-time human feedback combined with an open-loop walk cycle. The open-loop walk cycle is obtained from the recorded joint commands while the robot is walking using an existing walk algorithm as a black-box unit. We capture the corrective feedback signals delivered by a human using a wireless feedback mechanism in the form of corrections to the particular joints and we present experimental results showing that a policy learned from a walk algorithm can be used to improve the stability of another walk algorithm. We then follow up with improving the open-loop walk cycle using advice operators before performing real-time human demonstration. During the demonstration, we then capture the sensory readings and the corrections in the form of displacements of the foot positions while the robot is executing improved open-loop walk cycle. We then translate the feet displacement values into individual correction signals for the leg joints using a simplified inverse kinematics calculation. We use a locally weighted linear regression method to learn a mapping from the recorded sensor values to the correction values. Finally, we use a simple anomaly detection method by modeling the changes in the sensory readings throughout the walk cycle during a stable walk as normal distributions and executing the correction policy only if a sensory reading goes beyond the modeled values. Experimental results demonstrate an improvement in the walk stability.
Meyer, J., Schnitzspan, P., Kohlbrecher, S., Petersen, K., Andriluka, M., Schwahn, O., Klingauf, U., Roth, S., Schiele, B. & von Stryk, O.

A Semantic World Model for Urban Search and Rescue Based on Heterogeneous Sensors

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 180-193 inproceedings


Abstract: In urban search and rescue scenarios, typical applications of robots include autonomous exploration of possibly dangerous sites, and the recognition of victims and other objects of interest. In complex scenarios, relying on only one type of sensor is often misleading, while using complementary sensors frequently helps improving the performance. To that end, we propose a probabilistic world model that leverages information from heterogeneous sensors and integrates semantic attributes. This method of reasoning about complementary information is shown to be advantageous, yielding increased reliability compared to considering all sensors separately. We report results from several experiments with a wheeled USAR robot in a complex indoor scenario. The robot is able to learn an accurate map, and to detect real persons and signs of hazardous materials based on inertial sensing, odometry, a laser range finder, visual detection, and thermal imaging. The results show that combining heterogeneous sensor information increases the detection performance, and that semantic attributes can be successfully integrated into the world model.
Missura, M., Schmitz, A. & Behnke, S.

Learning Footstep Prediction from Motion Capture

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 97-108 inproceedings


Abstract: Central pattern generated walking for bipedal robots has proven to be a versatile and easily implementable solution that is used by several robot soccer teams in the RoboCup Humanoid Soccer League with great success. However, the forward character of generating motor commands from an abstract, rhythmical pattern does not inherently provide the means for controlling the precise location of footsteps. For implementing a footstep planning gait control, we developed a step prediction model that estimates the location of the next footstep in Cartesian coordinates given the same inputs that control the central pattern generator. We used motion capture data recorded from walking robots to estimate the parameters of the prediction model and to verify the accuracy of the predicted footstep locations. We achieved a precision with a mean error of 1.3cm.
Missura, M., Schmitz, A. & Behnke, S.

Designing Effective Humanoid Soccer Goalies

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 374-385 inproceedings


Abstract: Most of the research related to the topic of falling strategies considers falling to be an unavoidable part of bipedal walking and is focused on developing strategies to avoid falls and to minimize mechanical damage. We take an alternative point of view and regard falling as a means to an end. We present our falling strategy for the specific case of a robot soccer goalie that deliberately jumps in front of a moving ball to prevent it from rolling into the goal. The jump decision is based on observed ball position, speed and direction of movement. We show how we implement a targeted falling into the appropriate direction, minimize the time from the jump decision to ground impact, and what solutions we developed to prevent mechanical damage. The presented falling technique was used in RoboCup Humanoid KidSize and TeenSize competitions and proved to be essential for winning.
Mohammad Mahdi Kheirikhah, S.R. & Edalat, M.E.

A Review of Shape Memory Alloy Actuators in Robotics

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 206-217 inproceedings


Abstract: Shape Memory Alloys (SMAs) have been used for a wide variety of applications in various fields such as robotics. If these materials subjected to an appropriate thermomechanical process, they have ability to return to their initial shape. Often, they are used as actuators in robotic applications. The purpose of this paper is to present a brief review of literatures which using SMA in different robots' structure. First an introduction about shape memory effect of these materials will present. Then an assessment of done researches in application of these materials in robots' structure will accomplish and is devoted to the following area of robotics: Crawler, jumper, flower, fish, walker, medical and Bio-mimetic robotic hand.
Müller, J., Laue, T. & Röfer, T.

Kicking a Ball - Modeling Complex Dynamic Motions for Humanoid Robots

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 109-120 inproceedings


Abstract: Complex motions like kicking a ball into the goal are becoming more important in RoboCup leagues such as the Standard Platform League. Thus, there is a need for motion sequences that can be parameterized and changed dynamically. This paper presents a motion engine that translates motions into joint angles by using trajectories. These motions are defined as a set of Bezier curves that can be changed online to allow adjusting, for example, a kicking motion precisely to the actual position of the ball. During the execution, motions are stabilized by the combination of center of mass balancing and a gyro feedback-based closed-loop PID controller.
Naruse, T., Masutani, Y., Mitsunaga, N., Nagasaka, Y., Fujii, T., Watanabe, M., Nakagawa, Y. & Naito, O.

SSL-Humanoid: RoboCupSoccer using humanoid robots under the global vision

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 60-71 inproceedings


Abstract: We propose a new RoboCup league called "SSL-Humanoid". In the SSL-Humanoid league, the global vision system used in the RoboCup soccer Small Size Robot League (SSL) is also employed, and two teams of up to five humanoid robots play on a field whose size is half of the SSL field. This league aims to advance AI research for humanoid robots such as cooperation among robots, tactics and strategies. This contribution is possible due to the substitution of local to a global vision system. This is expected to attract new participants who have been interested in RoboCup but did not attend so far. We propose to organize this league as a sub-league of the original SSL. In this paper, we describe background of the proposal, research topics, road map toward 2014 and a summary of games played in 2009. Finally, we introduce team ODENS' system as an example of an SSL-Humanoid team.
Niemueller, T., Ferrein, A., Podbregar, P., Kellner, T., Rath, C. & Steinbauer, G.

Providing Ground-truth Data for the Nao Robot Platform

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 133-144 inproceedings


Abstract: Collecting ground truth-data for real-world applications is a non-trivial but very important task. In order to evaluate new algorithmic approaches or to benchmark system performance, they are inevitable. This is particularly true for robotics applications. In this paper we present our data collection for the biped humanoid robot Nao. Reflective markers were attached to Nao's body, and the positions and orientation of its body and head were tracked in 6D with an accurate professional vision-based body motion tracking system. While doing so, the data of Nao's internal state, i.e., the readings of all its servos, the inertial measurement unit, the force receptors plus a camera stream of the robot's camera were stored for different, typical robotic soccer scenarios in the context of the RoboCup Standard Platform League. These data will be combined in order to compile an accurate ground-truth data set. We describe how the data were recorded, in which format they are stored, and show the usability of the logged data in some first experiments on the recorded data sets. The data sets will be made publicly available for the RoboCup’s Standard Platform League community.
Pedro Abreu, Israel Costa, D.C.L.P.R. & Garganta, J.

Human vs. Robotic Soccer: How Far are they? A Statistical Comparison

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 242-253 inproceedings


Abstract: In soccer games, a performance indicator is defined as a selection of action variables that aims to define all aspects of accomplishment of the game goals. However their perception during the match is extremely difficult. Over the years, soccer has been used in many research areas including the robotic international soccer competition, RoboCup. The aim of this research project is to present a comparison study, performed to detect similarities between these two games (Human versus Robotic Simulation 2D soccer). Having an off-line automatic event detection tool as a base, a collection of final game statistics was done and the Mann-Whitney test was used to verify their statistical significance. The results show that the most frequent events occurred in both types of game are successful passes. In what concerns stopped game situation types, in both types of games, the most frequent one is the Throw in situation (Human-59,8%, versus Robotic-74,1%) and the less frequent is the Corner situation (Human-13,7%, versus Robotic-10,3%). Some differences still reside, especially in the frequency of set pieces and the action prior the goal.
Pellenz, J., Neuhaus, F., Dillenberger, D. & Paulus, D.

Mixed 2D/3D Perception for Autonomous Robots in Unstructured Environments

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 303-313 inproceedings


Abstract: Autonomous robots in real world applications have to deal with a complex 3D environment, but are often equipped with standard 2D laser range finders (LRF) only. By using the 2D LRF for both, the 2D localization and mapping (which can be done efficiently and precisely) and for the 3D obstacle detection (which makes the robot move safely), a completely autonomous robot can be built with affordable 2D LRFs. We use the 2D LRF to perform particle filter based SLAM to generate a 2D occupancy grid, and the same LRF (moved by two servo motors) to acquire 3D scans to detect obstacles not visible in the 2D scans. The 3D data is analyzed with a recursive principal component analysis (PCA) based method, and the detected obstacles are recorded in a separate obstacle map. This obstacle map and the occupancy map are merged for the path planning. Our solution was tested on our mobile system Robbie during the RoboCup Rescue competitions in 2008 and 2009, winning the mapping challenge at the world championship 2008 and the German Open in 2009.
This shows that the benefit of a sensor can dramatically be increased by actively controlling it, and that mixed 2D/3D perception can efficiently be achieved with a standard 2D sensor by controlling it actively.
Petersen, K., Stoll, G. & von Stryk, O.

A Supporter Behavior for Soccer Playing Humanoid Robots

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 386-396 inproceedings


Abstract: In this paper, an efficient behavior for a humanoid soccer robot supporting a teammate manipulating the ball is proposed and evaluated. This approach has been derived from human soccer tactics, requires only minimal information about the current state of the game and is therefore well suited for robots with directed vision and limited motion abilities like humanoid robots. The position of the supporter is chosen in a way to reduce the chances of the opponent team for scoring a goal and to increase the chances of the robot to take over the role of its teammate manipulating ball in case the teammate falls down or is being dodged by the opponent team. By different parameterizations more defensive or more offensive player tactics can be realized as well as adaptations to specific tactics and skills of the opposing robot players. The developed behavior was evaluated in a simulation based statistical analysis. Also an evaluation in games during RoboCup 2009 is given.
Randelli, G., Marchetti, L. & Marino, F.A.

Multi-Agent Behavior Composition through Adaptable Software Architectures and Tangible Interfaces

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 278-290 inproceedings


Abstract: Cooperative behavior realization is an important aspect of Multi-Agent Systems, and it has been widely addressed by the robotics community. However, the interaction among multiple robots and a human operator still requires a non-negligible effort to be effective. The availability of modern input devices can ease the behavior composition process, allowing a user to train Multi-Agent Systems by using alternative methods. In this paper, we introduce a novel approach to integrate Tangible User Interfaces within a reconfigurable software architecture, for cooperative Multi-Robot Systems. To address a real test case, we implemented a strategy training system for a humanoid RoboCup soccer team. Such a system allows a human coach to train several robotic players by using multiple input interfaces, directly onto the application field.
Santos, P.L., Oliveira, R., Ahmad, A., Lima, P.U. & Santos, J.

Cooperative Localization Based on Visually Shared Objects

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 350-361 inproceedings


Abstract: In this paper we describe a cooperative localization algorithm based on a modification of the Monte Carlo Localization algorithm where, when a robot detects it is lost, particles are spread not uniformly in the state space, but rather according to the information on the location of an object whose distance and bearing is measured by the lost robot. The object location is provided by other robots of the same team using explicit (wireless) communication. Results of application of the method to a team of real robots are presented.
Schulz, H., Liu, W., Stückler, J. & Behnke., S.

Utilizing the Structure of Field Lines for Efficient Soccer Robot Localization

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 397-408 inproceedings


Abstract: The rules in RoboCup soccer more and more discourage a solely colorbased orientation on the soccer field.While the field size increases, field boundary markers and goals become smaller and less colorful. For robust game play, robots therefore need to maintain a state and rely on more subtle environmental clues. Field lines are particularly interesting, because they are hardly completely occluded and observing them significantly reduces the number of possible poses on the field.
In this work we present a method for line-based localization. Unlike previous work, our method first recovers a line structure graph from the image. From the graph we can then easily derive features such as lines and corners. Finally, we describe optimizations for efficient use of the derived features in a particle filter. The method described in this paper is used regularly on our humanoid soccer robots.
Seekircher, A., Röfer, T. & Laue, T.

Entropy-Based Active Vision for a Humanoid Soccer Robot

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 1-12 inproceedings


Abstract: In this paper, we show how the estimation of a robot's world model can be improved by actively sensing the environment through considering the current world state estimate through minimizing the entropy of an underlying particle distribution. Being originally computationally expensive, this approach is optimized to become executable in real-time on a robot with limited resources. We demonstrate the approach on a humanoid robot, performing self-localization and ball tracking on a RoboCup soccer field.
Seib, V., Gossow, D., Vetter, S. & Paulus, D.

Hierarchichal Multi-Robot Coordination

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 314-323 inproceedings


Abstract: The complexity and variety of household chores creates conflicting demands on the technical design of domestic robots. One solution for this problem is the coordination of several specialized robots based on the master-slave principle. One robot acts as a master system, tracking and remotely controlling the slave robots. This way, only the master robot needs to be equipped with sophisticated sensors and computing hardware. We implemented a tracking system using an infra-red camera for the master and active markers on the slave robot. The master system is able to interact with the user using natural language. It builds a map of its environment automatically using a laser range finder. It can track a cleaning robot for which we use the commercially available platform "Roomba" by iRobot. The master safely navigates it to a given destination, avoiding obstacles. We successfully demonstrated the system during the RoboCup@Home competitions 2009 in Graz, Austria. We evaluate the performance of the two systems and describe the accuracy of localization and navigation.
Shafii, N., Reis, L.P. & Lau, N.

Biped Walking using Coronal and Sagittal Movements based on Truncated Fourier Series

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 324-335 inproceedings


Abstract: Biped walking by using all joint movements and DOFs in both directions (sagittal plane and coronal plane) is one of the most complicated research topics in robotics. In this paper, angular trajectories of a stable biped walking for a humanoid robot are generated by a Truncated Fourier Series (TFS) approach. The movements of legs and arms in sagittal plane are implemented by an optimized gait generator and a new model is proposed that can also produce the movement of legs in coronal plane based on TFS. Particle Swarm Optimization (PSO) is used to find the best angular trajectories and optimize TFS. Experimental results show that the using joints movements in sagittal and coronal planes to compose the walking skill allowed the biped robot to walk faster than previous methods that only used the joints in sagittal plane.
Stückler, J. & Behnke, S.

Improving People Awareness of Service Robots by Semantic Scene Knowledge

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 157-168 inproceedings


Abstract: Many mobile service robots operate in close interaction with humans. Being constantly aware of the people in the surrounding of the robot thus poses an important challenge to perception and behavior design.
In this paper, we present an approach to people awareness for mobile service robots that utilizes knowledge about the semantics of the environment. The known semantics, e.g., about walkable floor, chairs, and shelves, provides the robot with prior information. We utilize information about the a-priori likelihood that people are present at semantically distinct places. Together with reasonable face heights inferred from scene semantics, this information supports robust detection and awareness of people in the robot's environment. For efficient exploration of the environment for people, we propose a strategy which chooses search locations that maximize the expected detection rate of new persons.
We evaluate our approach with our domestic service robot that competes in the RoboCup@Home league.
Visser, U.

TopLeague and Bundesliga Manager: New Generation Online Soccer Games

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 230-241 inproceedings


Abstract: This paper describes a new generation online soccer manager. More than 210,000 users operate TopLeague and the Official Bundesliga Manager, complex real-time soccer simulators that are based on actual data of real professional soccer players. The underlying technology is a hierarchical three-tier multiagent system that consists of autonomous BDI agents and allows dynamic group structures (e.g. an emergent situation for a wing attack). The German Bundesliga, one of the most prestigious soccer leagues in the world, adopted this AI system for their official web site. The online games run seamlessly in a web browser with a state-of-the-art 3D visualization engine. Fundamental research within the domain of the RoboCup simulation league is the basis for this technology. We will describe the architecture of the multiagent system (MAS) in this paper, discuss motion capture techniques for graphical animation, and reveal details about user acceptance of the games.
Özkucur, N.E. & Akın, H.L.

Localization with Non-unique Landmark Observations

2011 RoboCup 2010: Robot Soccer World Cup XIV, pp. 72-81 inproceedings


Abstract: In the probabilistic robot localization problem, when the associations of observations with the landmarks in the map are given, the solution is straightforward. However, when the observations are non-unique (e.g. the association with the map is not given) the problem becomes more challenging. In the Standard Platform League (SPL) and other similar categories of RoboCup, as the field setups evolve over years, the observations become less informative. In the localization level, we have to seek solutions with non-unique landmark observations. In this paper, we established the probabilistic model of the problem and showed the difficulty of optimal solution. After that, we introduce our importance sampling based approximate solution and implicit hypothesis pruning. We give results from simulation tests in the SPL setup using corners and goal bar observations and discuss characteristics of our approach.