RoboCup 2008 Publications

From RoboCup Wiki
Jump to: navigation, search
Author Title Year Journal/Proceedings Reftype DOI/URL
Antonelli, M., Libera, F.D., Menegatti, E., Minato, T. & Ishiguro, H.

Intuitive Humanoid Motion Generation Joining User-Defined Key-Frames and Automatic Learning


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 13-24 inproceedings

DOI URL

Abstract: In this paper we present a new method for generating humanoid robot movements. We propose to merge the intuitiveness of the widely used key-frame technique with the optimization provided by automatic learning algorithms. Key-frame approaches are straightforward but require the user to precisely define the position of each robot joint, a very time consuming task. Automatic learning strategies can search for a good combination of parameters resulting in an effective motion of the robot without requiring user effort. On the other hand their search usually cannot be easily driven by the operator and the results can hardly be modified manually. While the fitness function gives a quantitative evaluation of the motion (e.g. "How far the robot moved?"), it cannot provide a qualitative evaluation, for instance the similarity to the human movements. In the proposed technique the user, exploiting the key-frame approach, can intuitively bound the search by specifying relationships to be maintained between the joints and by giving a range of possible values for easily understandable parameters. The automatic learning algorithm then performs a local exploration of the parameter space inside the defined bounds. Thanks to the clear meaning of the parameters provided by the user, s/he can give qualitative evaluation of the generated motion (e.g. "This walking gait looks odd. Let's raise the knee more") and easily introduce new constraints to the motion. Experimental results proved the approach to be successful in terms of reduction of motion-development time, in terms of natural appearance of the motion, and in terms of stability of the walking.
Arenas, M., del Solar, J.R., Norambuena, S. & Cubillos, S.

A Robot Referee for Robot Soccer


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 288-295 inproceedings

DOI URL

Abstract: The aim of this paper is to propose a robot referee for robot soccer. This idea is implemented using a service robot that moves along one of the field sides, uses its own cameras to analyze the game, and communicates its decisions to the human spectators using speech, and to the robot players using wireless communication. The robot uses a video-based game analysis toolbox that is able to analyze the actions at up to 20fps. This toolbox includes robots, ball, landmarks, and lines detection and tracking, as well as refereeing decision-making. This robot system is validated and characterized in real game situations with humanoid robot players.
This research was partially supported by FONDECYT (Chile) under Project Number 1061158.
Asano, K., Murakami, K. & Naruse, T.

Detection of Basic Behaviors in Logged Data in RoboCup Small Size League


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 439-450 inproceedings

DOI URL

Abstract: This paper describes a method that extracts the basic behaviors of robots such as kicking and passing from the history data of the positions and velocities of the robots and the ball in RoboCup Small Size League (SSL). In this paper, as a first step, we propose an offline method that extracts the basic behaviors of robots from the logged data which is a record of the positions and velocities of the robots and the ball as the time series data. First, paying attention to the ball movement, we extract the line segments in the ball trajectory which satisfy our proposed conditions. These segments arise from the kicking actions. Then we classify the extracted line segments into the detailed kicking actions by analysing the intention of the kicked robot. We also propose algorithms that detect and classify the covering actions. Experimental results show that 98% of the kicking actions are correctly detected and more than 80% of the detected kicking actions are correctly classified, and that 90% of the covering actions are also correctly classified.
Beck, D., Ferrein, A. & Lakemeyer, G.

Landmark-Based Representations for Navigating Holonomic Soccer Robots


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 25-36 inproceedings

DOI URL

Abstract: For navigating mobile robots the central problems of path planning and collision avoidance have to be solved. In this paper we propose a method to solve the (local) path planning problem in a reactive fashion given a landmark-based representation of the environment. The perceived obstacles define a point set for a Delaunay tessellation based on which a traversal graph containing possible paths to the target position is constructed. By applying A* we find a short and safe path through the obstacles. Although the traversal graph is recomputed in every iteration in order to achieve a high degree of reactivity the method guarantees stable paths in a static environment; oscillating behavior known from other local methods is precluded. This method has been successfully implemented on our Middle-size robots.
Becker, D. & Risler, M.

Mutual Localization in a Team of Autonomous Robots Using Acoustic Robot Detection


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 37-48 inproceedings

DOI URL

Abstract: In order to improve self-localization accuracy we are exploring ways of mutual localization in a team of autonomous robots. Detecting team mates visually usually leads to inaccurate bearings and only rough distance estimates. Also, visually identifying teammates is not possible. Therefore we are investigating methods of gaining relative position information acoustically in a team of robots.
The technique introduced in this paper is a variant of code-multiplexed communication (CDMA, code division multiple access). In a CDMA system, several receivers and senders can communicate at the same time, using the same carrier frequency. Well-known examples of CDMA systems include wireless computer networks and the Global Positioning System, GPS. While these systems use electro-magnetic waves, we will try to adopt the CDMA principle towards using acoustic pattern recognition, enabling robots to calculate distances and bearings to each other.
First, we explain the general idea of cross-correlation functions and appropriate signal pattern generation. We will further explain the importance of synchronized clocks and discuss the problems arising from clock drifts.
Finally, we describe an implementation using the Aibo ERS-7 as platform and briefly state basic results, including measurement accuracy and a runtime estimate. We will briefly discuss acoustic localization in the specific scenario of a RoboCup soccer game.
Birbach, O., Kurlbaum, J., Laue, T. & Frese, U.

Tracking of Ball Trajectories with a Free Moving Camera-Inertial Sensor


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 49-60 inproceedings

DOI URL

Abstract: This paper is motivated by the goal of a visual perception system for the RoboCup 2050 challenge to win against the human world-cup champion. Its contribution is to answer two questions on the subproblem of predicting the motion of a flying ball. First, if we could detect the ball in images, is that enough information to predict its motion precise enough? And second, how much do we lose by using the real-time capable Unscented Kalman Filter (UKF) instead of non-linear maximum likelihood as a gold standard? We present experiments with a camera and an inertial sensor on a helmet worn by a human soccer player. These confirm that the precision is roughly enough and using an UKF is feasible.
Birk, A., Poppinga, J. & Pfingsthorn, M.

Using Different Humanoid Robots for Science Edutainment of Secondary School Pupils


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 451-462 inproceedings

DOI URL

Abstract: Robotics camps that involve design, construction and programming tasks are a popular part of various educational activities. This paper presents the results of a survey that accompanied the Innovationscamp, a one week intensive workshop for promoting science and engineering among secondary school pupils through humanoid robots. Two very different types of platforms were used in this workshop: LEGO mindstorms, which are widely used for educational activities, and Bioloid humanoids, which are more commonly used for professional research. Though the workshop participants were robotics novices, the survey indicates through several statistically significant results that the Bioloid robots are preferred by the pupils over the LEGO robots as educational tools.
Birk, A., Poppinga, J., Stoyanov, T. & Nevatia, Y.

Planetary Exploration in USARsim: A Case Study Including Real World Data from Mars


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 463-472 inproceedings

DOI URL

Abstract: Intelligent Mobile Robots are increasingly used in unstructured domains; one particularly challenging example for this is planetary exploration. The preparation of according missions is highly non-trivial, especially as it is difficult to carry out realistic experiments without very sophisticated infrastructures. In this paper, we argue that the Unified System for Automation and Robot Simulation (USARSim) offers interesting opportunities for research on planetary exploration by mobile robots. With the example of work on terrain classification, it is shown how synthetic as well as real world data from Mars can be used to test an algorithm's performance in USARSim. Concretely, experiments with an algorithm for the detection of negotiable ground on a planetary surface are presented. It is shown that the approach performs fast and robust on planetary surfaces.
This work was supported by the German Research Foundation (DFG).
Correa, M., del Solar, J.R. & Bernuy, F.

Face Recognition for Human-Robot Interaction Applications: A Comparative Study


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 473-484 inproceedings

DOI URL

Abstract: The major task of police force agents in rescue simulation environment is to connect the isolated parts of the city. To achieve this goal, the best blocked roads should be chosen to clear. This selection is based on some issues such as number of burning buildings and victims existing in the mentioned parts. A linear combination of these factors is essential to determine a priority for each road. In this paper we propose an integrated Genetic Algorithm (GA) and Fuzzy Logic approach to optimize the combination statement. The parameters are learned via GA for some training maps. Then, because of differences between test and train maps, the agent should decide which parameters to choose according to the new map. The agents? decision is based on similarity measures between characteristics of maps using Fuzzy Logic. After utilizing this method, the simulation score increased between 2% and 7% in 20 test maps.
Doostdar, M., Schiffer, S. & Lakemeyer, G.

A Robust Speech Recognition System for Service-Robotics Applications


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 1-12 inproceedings

DOI URL

Abstract: Mobile service robots in human environments need to have versatile abilities to perceive and to interact with their environment. Spoken language is a natural way to interact with a robot, in general, and to instruct it, in particular. However, most existing speech recognition systems often suffer from high environmental noise present in the target domain and they require in-depth knowledge of the underlying theory in case of necessary adaptation to reach the desired accuracy. We propose and evaluate an architecture for a robust speaker independent speech recognition system using off-the-shelf technology and simple additional methods. We first use close speech detection to segment closed utterances which alleviates the recognition process. By further utilizing a combination of an FSG based and an N-gram based speech decoder we reduce false positive recognitions while achieving high accuracy.
Enderle, S., Guenther, W., Hilscher, H.-J. & Kenn, H.

xROB-S and iCon-X: Flexible Hardware, Visual Programming and Software Component Reuse


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 485-494 inproceedings

DOI URL

Abstract: The following article describes an optimized system solution for RoboCup-applications, which reaches from RoboCup-Junior to humanoid robots.
It describes the versatile multiprocessor-hardware xROB-CUx. The control unit contains the communication processor netX®, based on an arm9-Core. A variety of communication- and hardware-interfaces as well as the simple scaling enables the use as a communication junction, e.g. for sub-layered DSP or FPGA-axis-controllers in humanoid robots or as a control unit for construction systems (e.g. fischertechnik®, Lego®, qfix®).
The paper also contains a description of the Windows® based software-system xROB-L , which contains the open visual programming-system iCon-L® as an essential part. The creation of user-programs, PC-simulation, data-archiving and the download to different target-systems is illustrated as well.
While the up to now available developer-environment iCon-MFB (Make Function Blocks) only supported the design-process of advanced function blocks in Visual C/C++, it is now possible to integrate ANSI-C written function blocks via the GNU-Compiler with the tool xROB-C. The main advantage to other Solutions is the Vision system, the relation of cost to performance and the wide area of possible applications. Because of the easy to use visual programming technology it can be used by all age groups.
Gabel, T., Riedmiller, M. & Trost, F.

A Case Study on Improving Defense Behavior in Soccer Simulation 2D: The NeuroHassle Approach


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 61-72 inproceedings

DOI URL

Abstract: While a lot of papers on RoboCup's robotic 2D soccer simulation have focused on the players' offensive behavior, there are only a few papers that specifically address a team's defense strategy. In this paper, we consider a defense scenario of crucial importance: We focus on situations where one of our players must interfere and disturb an opponent ball leading player in order to scotch the opponent team's attack at an early stage and, even better, to eventually conquer the ball initiating a counter attack. We employ a reinforcement learning methodology that enables our players to autonomously acquire such an aggressive duel behavior, and we have embedded it into our soccer simulation team's defensive strategy. Employing the learned NeuroHassle policy in our competition team, we were able to clearly improve the capabilities of our defense and, thus, to increase the performance of our team as a whole.
Guerrero, P., del Solar, J.R. & Romero, M.

Explicitly Task Oriented Probabilistic Active Vision for a Mobile Robot


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 85-96 inproceedings

DOI URL

Abstract: A mobile robot has always uncertainty about the world model. Reducing this uncertainty is very hard because there is a huge amount of information and the robot must focus on the most relevant one. The selection of the most relevant information must be based on the task the robot is executing, but there could be several sources of information where the robot would like to focus on. This is also true in robot soccer where the robot must pay attention to landmarks in order to self-localize and to the ball and robots in order to follow the status of the game. In the presented work, an explicitly task oriented probabilistic active vision system is proposed. The system tries to minimize the most relevant components of the uncertainty for the task that is been performed and it is explicitly task oriented in the sense that it explicitly considers a task specific value function. As a result, the system estimates the convenience of looking towards each of the available objects. As a test-bed for the presented active vision approach, we selected a robot soccer attention problem: goal covering by a goalie player.
This research was partially supported by FONDECYT (Chile) under Project Number 1090250.
Göhring, D., Mellmann, H. & Burkhard, H.-D.

Constraint Based Belief Modeling


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 73-84 inproceedings

DOI URL

Abstract: In this paper we present a novel approach using constraint based techniques for world modeling, i.e. self localization and object modeling. Within the last years, we have seen a reduction of landmarks as beacons, colored goals, within the RoboCup domain. Using other features as line information becomes more important. Using such sensor data is tricky, especially when the resulting position belief is stretched over a larger area. Constraints can overcome this limitations, as they have several advantages: They can represent large distributions and are easy to store and to communicate to other robots. Propagation of a several constraints can be computationally cheap. Even high dimensional belief functions can be used. We will describe a sample implementation and show experimental results.
Iravani, P.

Multi-level Network Analysis of Multi-agent Systems


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 495-506 inproceedings

DOI URL

Abstract: This paper presents a multi-level network-based approach to study complex systems formed by multiple autonomous agents. The fundamental idea behind this approach is that elements of a system (represented by network vertices) and their interactions (represented by edges) can be assembled to form structures. Structures are considered to be at one hierarchical level above the elements and interactions that form them, leading to a multi-level organisation.
Analysing complex systems represented by multi-level networks make possible the study of the relationships between network topology and dynamics to the system?s global outcome. The framework proposed in this paper is exemplified using data from the RoboCup Football Simulation League.
Ji, X., Zhang, H., Hai, D. & Zheng, Z.

An Incremental SLAM Algorithm with Inter-calibration between State Estimation and Data Association


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 97-108 inproceedings

DOI URL

Abstract: In most SLAM (Simultaneous Localization and Mapping) ap proaches, there is only unilateral data stream from data association (DA) to state estimation (SE), and the SE model estimates states according to the results of DA. This paper focuses on the reciprocity between DA and SE, and an incremental algorithm with inter-calibration between SE and DA is presented. Our approach uses a tree model called correspondence tree (CT) to represent the solution space of data association. CT is layered according to time steps and every node in it is a data association hypothesis for all the measurements gotten at the same time step. A best-first search with limited back-tracking is designed to find the optimal path in CT, and a state estimation approach based on the least-squares method is used to compute the cost of nodes in CT and update state estimation incrementally, so direct feedback is introduced from the SE to DA. With the interaction between DA and SE, and combining with tree pruning techniques, our approach can get accurate data association and state estimation for on-line SLAM applications.
Ji, X., Zhang, H., Hai, D. & Zheng, Z.

A Decision-Theoretic Active Loop Closing Approach to Autonomous Robot Exploration and Mapping


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 507-518 inproceedings

DOI URL

Abstract: One of the challenges of SLAM (Simultaneous Localization and Mapping) for autonomous robots is the loop closing problem. In this paper, a decision-theoretic active loop closing approach is presented, which integrates the exploration planning with loop closing. In our approach, the active loop closing process is modeled as a multi-stage decision problem, and a frontier-based auxiliary topological map is build to assist the decision process. The autonomous robot chooses its actions according to the sequential decision results. The unknown range most likely to close a loop is selected to explored, and a particle-filter-based localization and smoothing method applied to partial maps is used in the loop validating and loop constraints building process. Experiments have shown that our approach can practically implement loop closure and obviously improve the mapping precision compared to passive exploration strategy.
Knox, W.B., Lee, J. & Stone, P.

Domestic Interaction on a Segway Base


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 519-531 inproceedings

DOI URL

Abstract: To be useful in a home environment, an assistive robot needs to be capable of a broad range of interactive activities such as locating objects, following specific people, and distinguishing among different people. This paper presents a Segway-based robot that successfully performed all of these tasks en route to a second place finish in the RoboCup@Home 2007 competition. The main contribution is a complete description and analysis of the robot system and its implemented algorithms that enabled the robot's successful human-robot interaction in this broad and challenging forum. We describe in detail a novel person recognition algorithm, a key component of our overall success, that included two co-trained classifiers, each focusing on different aspects of the person (face and shirt color).
Kobayashi, H., Osaki, T., Okuyama, T., Ishino, A. & Shinohara, A.

Development of an Augmented Environment and Autonomous Learning for Quadruped Robots


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 109-120 inproceedings

DOI URL

Abstract: This paper describes an interactive experimental environment for autonomous soccer robots, which is a soccer field augmented by utilizing camera input and projector output. This environment, in a sense, plays an intermediate role between simulated environments and real environments. We can simulate some parts of real environments, e.g., real objects such as robots or a ball, and reflect simulated data into the real environments, e.g., to visualize the positions on the field, so as to create a situation that allows easy debugging of robot programs. As an application in the augmented environment, we address the learning of goalie strategies on real quadruped robots in penalty kicks. Our robots learn and acquire sophisticated strategies in a fully simulated environment, and then they autonomously adapt to real environments in the augmented environment.
Laue, T. & Hebbel, M.

Automatic Parameter Optimization for a Dynamic Robot Simulation


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 121-132 inproceedings

DOI URL

Abstract: One common problem of dynamic robot simulations is the accuracy of the actuators' behavior and their interaction with the environment. Especially when simulating legged robots which have optimized gaits resulting from machine learning, manually finding a proper configuration within the high-dimensional parameter space of the simulation environment becomes a demanding task. In this paper, we describe a multi-staged approach for automatically optimizing a large set of different simulation parameters. The optimization is carried out offline through an evolutionary algorithm which uses the difference between the recorded data of a real robot and the behavior of the simulation as fitness function. A model of an AIBO robot performing a variety of different walking gaits serves as an example of the approach.
Lu, H., Zhang, H., Xiao, J., Liu, F. & Zheng, Z.

Arbitrary Ball Recognition Based on Omni-Directional Vision for Soccer Robots


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 133-144 inproceedings

DOI URL

Abstract: Recognizing the arbitrary standard FIFA ball is a significant ability for soccer robots to play competition without the constraint of current color-coded environment. This paper describes a novel method to recognize arbitrary FIFA ball based on omni-directional vision system. Firstly the omni-directional vision system and its calibration for distance map are introduced, and the conclusion that the ball on the field can be imaged to be ellipse approximately is derived by analyzing the imaging character. Then the arbitrary FIFA ball is detected by using image processing algorithm to search the ellipse imaged by the ball according to the derivation. In the actual application, a simple but effective ball tracking algorithm is also used to reduce the running time needed after the ball has been recognized globally. The experiment results show that the novel method can recognize the arbitrary FIFA ball effectively and in real-time.
Ma, J. & Cameron, S.

Combining Policy Search with Planning in Multi-agent Cooperation


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 532-543 inproceedings

DOI URL

Abstract: It is cooperation that essentially differentiates multi-agent systems (MASs) from single-agent intelligence. In realistic MAS applications such as RoboCup, repeated work has shown that traditional machine learning (ML) approaches have difficulty mapping directly from cooperative behaviours to actuator outputs. To overcome this problem, vertical layered architectures are commonly used to break cooperation down into behavioural layers; ML has then been used to generate different low-level skills, and a planning mechanism added to create high-level cooperation. We propose a novel method called Policy Search Planning (PSP), in which Policy Search is used to find an optimal policy for selecting plans from a plan pool. PSP extends an existing gradient-search method (GPOMDP) to a MAS domain. We demonstrate how PSP can be used in RoboCup Simulation, and our experimental results reveal robustness, adaptivity, and outperformance over other methods.
McGrath, S., Anderson, J. & Baltes, J.

Model-Free Active Balancing for Humanoid Robots


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 544-555 inproceedings

DOI URL

Abstract: To be practical, humanoid robots must be able to manoeuvre over a variety of flat and uneven terrains, at different speeds and with varying gaits and motions. This paper describes three balancing-reflex algorithms (threshold control, PID control, and hybrid control) that were implemented on a real 8 DOF small humanoid robot equipped with a two-axis accelerometer sensor to study the capabilities and limitations of various balancing algorithms when combined with a single sensor. We term this approach a model-free approach, since it does not require a mathematical model of the underlying robot. Instead the controller attempts to recreate successful previous motions (so-called baseline motions). In our extensive tests, the basic threshold algorithm proves the most effective overall. All algorithms are able to balance for simple tasks, but as the balancing required becomes more complex (e.g. controlling multiple joints over uneven terrain), the need for more sophisticated algorithms becomes apparent.
McKinnon, B., Baltes, J. & Anderson, J.

Stereo-Vision Based Control of a Car Using Fast Line-Segment Extraction


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 556-567 inproceedings

DOI URL

Abstract: This paper describes our work on applying stereo vision to the control of a car or car-like mobile robot, using cheap, low-quality cameras. Our approach is based on line segments, since those provide significant information about the environment, provide more depth information than point features, and are robust to image noise and colour variation. However, stereo matching with line segments is a difficult problem, due to poorly localized end points and perspective distortion. Our algorithm uses integral images and Haar features for line segment extraction. Dynamic programming is used in the line segment matching phase. The resulting line segments track accurately from one frame to the next, even in the presence of noise.
Meriçli, T., Meriçli, Ç. & Akın, H.L.

A Robust Statistical Collision Detection Framework for Quadruped Robots


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 145-156 inproceedings

DOI URL

Abstract: In order to achieve its tasks in an effective manner, an autonomous mobile robot must be able to detect and quickly recover from collisions. This paper proposes a new solution to the problem of detecting collisions during omnidirectional motion of a quadruped robot equipped with an internal accelerometer. We consider this as an instance of general signal processing and statistical anomaly detection problems. We find that temporal accelerometer readings examined in the frequency domain are good indicators of regularities (normal motion) and novel situations (collisions). In the course of time, the robot builds a probabilistic model that captures its proprioceptive properties while walking without obstruction and uses that model to determine whether there is an abnormality in the case of an unfamiliar pattern. The approach does not depend on walk characteristics and the walking algorithm used, and is insensitive to the surface texture that the robot walks on as long as the surface is flat. The experiments demonstrate quite fast and successful detection of collisions independent of the point of contact with an acceptably low false positive rate.
MeriÇli, Ç. & Akın, H.L.

A Layered Metric Definition and Evaluation Framework for Multirobot Systems


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 568-579 inproceedings

DOI URL

Abstract: In order to accomplish it successfully, the top-level goal of a multi-robot team should be decomposed into a sequence of sub-goals and proper sequences of actions for achieving these subgoals should be selected and refined through execution. Selecting the proper actions at any given time requires the ability to evaluate the current state of the environment, which can be achieved by using metrics that give quantitative information about the environment. Defining appropriate metrics is already a challenging problem; however, it is even harder to assess the performance of individual metrics. This work proposes a layered evaluation scheme for robot soccer where the environment is represented in different time resolutions at each layer. A set of metrics defined on these layers together with a novel metric validation method for assessing the performance of the defined metrics are proposed.
Michel, O., Bourquin, Y. & Baillie, J.-C.

RobotStadium: Online Humanoid Robot Soccer Simulation Competition


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 580-590 inproceedings

DOI URL

Abstract: This paper describes robotstadium: an online simulation contest based on the new RoboCup Nao Standard League. The simulation features two teams with four Nao robots each team, a ball and a soccer field corresponding the specifications of the real setup used for the new RoboCup Standard League using the Nao robot. Participation to the contest is free of charge and open to anyone. Competitors can simply register on the web site and download a free software package to start programming their team of soccer-playing Nao robots. This package is based on the Webots simulation software, the URBI middleware and the Java programming language. Once they have programmed their team of robots, competitors can upload their program on the web site and see how their team behaves in the competition. Matches are run every day and the ranking is updated accordingly in the "hall of fame". New simulation movies are made available on a daily basis so that anyone can watch them and enjoy the competition on the web. The contest is running online for a given period of time after which the best ranked competitors will be selected for a on-site final during the next RoboCup event. This contest is sponsored by The RoboCup federation, Aldebaran Robotics, Cyberbotics and Gostai. Prizes includes a Nao robot, a Webots package and a URBI package.
Moballegh, H.R., Mohajer, M. & Rojas, R.

Increasing Foot Clearance in Biped Walking: Independence of Body Vibration Amplitude from Foot Clearance


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 157-165 inproceedings

DOI URL

Abstract: A method of foot clearance achievement and frontal plane stability of 3D bipeds is mathematically analyzed. The independency of the robot's body vibration amplitude from the foot clearance is also proven. The analyzed method takes advantage of the sideways vibration of the body generated by periodically shortening each leg to obtain the required foot clearance. A mathematical model of the biped in the frontal plane is suggested and analyzed in two separate phases, resulted in the calculation of the steady state working point. It is demonstrated that in steady state, the amplitude of the body vibration becomes independent from the leg length vibration amplitude. A direct advantage of the proof is the possibility of achieving of high foot-to-ground distances by increasing the leg length vibration amplitude as the robot reaches its steady state. The results have been verified using both simulations and real robot experiments. To guarantee the stability of the robot in the transient phases a method is suggested based on an energy criterion.
Mohan, R.E., Calderon, C.A.A., Zhou, C., Yang, T., Zhang, L. & Yang, Y.

Adapting ADDIE Model for Human Robot Interaction in Soccer Robotics Domain


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 166-176 inproceedings

DOI URL

Abstract: Though human robot interaction has attracted lots of attention in the recent years due to increasing presence of robots in the marketplace, very little work has been done with systematic human robot interaction development using process models. Analysis, Design, Develop, Implement and Evaluate (ADDIE) model is one of the well established systematic process models for developing instructions in instructional design (ID) community. This paper focuses on two issues: 1) guidance in adapting ADDIE model for human robot interaction, and 2) performance improvement in adopting ADDIE model for human robot interaction in soccer robotics domain. Evaluations were performed and adequate results were obtained. ADDIE modelled human robot interaction was used by our Robo-Erectus Junior humanoid robots in the 2 versus 2 humanoid leagues of RoboCup 2007.
Nagasaka, Y., Kitahara, T. & Takahashi, T.

A Proposal of Bridging Activities between RoboCupJunior and Senior Leagues


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 177-188 inproceedings

DOI URL

Abstract: We propose new bridging activities between RoboCupJunior (RCJ) and RoboCup senior leagues. Many of RCJ graduates cannot find a suitable way to progress to the senior leagues. Therefore, they often retire from the RoboCup at this stage. A major problem is the technical gap between RCJ and senior leagues. However, if some top-level RCJ teams are provided with sufficient technical advice, they can join senior leagues in a short time. Our proposal aims to achieve this. Since authers have managed RCJ Soccer Secondary class (RCJSS) and RoboCup Soccer Small-Size League (SSL) in RoboCup national competition and know them well, the bridging activities for them are proposed. The first-level activities include the following. (1) RCJSS members visit an SSL competition, watch the game, and understand its rules, and vice versa. (2) We have realized an inter-league game using only the current field and robots in RoboCup National Open 2007. (3) A greater level of interaction is required. When the RCJSS teams require technical assistance or advice, if they are friendly with the members of any SSL team, they can freely seek assistance. The second-level is to form a bridge league between RCJSS and SSL. (1) Rules are introduced from SSL since this league aims to be a stepping-stone to full SSL. The field size is reduced to half the current size of an SSL field. The number of robots is reduced from five to three in order to reduce the parts cost. (2) We prepare development kits for transition. RCJSS teams can develop their robot systems rapidly by utilizing these kits for transition.
Pachur, D., Laue, T. & Röfer, T.

Real-Time Simulation of Motion-Based Camera Disturbances


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 591-601 inproceedings

DOI URL

Abstract: In the RoboCup domain, many robot systems use low-cost image sensors to perceive the robot?s environment and to locate the robot in its environment. The image processing typically has to handle image distortions such as motion blur, noise, and the properties of the shutter mechanism. If a simulator is used in the development of the control software, the simulation has to take account for these artifacts. Otherwise, the performance of the image processing system in the simulation may not correspond to its performance on the real robot; it may even perform worse.
The effect of motion blur has been widely used for special effects both for movies and for computer games. While real-time algorithms using modern graphics hardware came up in recent years, the image distortion resulting from a so-called rolling shutter has not been in focus so far. In fact, this effect is not relevant for gaming, but it is for simulating low-cost cameras of robots.
In this paper, we present an efficient way to simulate the rolling shutter effect using per-pixel velocities. In addition, we improve the velocity buffer method for creating motion blur using the current speed of each pixel in real-time. The application of our approach is shown exemplarily for the head-mounted camera of a humanoid soccer robot.
Palamara, P.F., Ziparo, V.A., Iocchi, L., Nardi, D. & Lima, P.

Teamwork Design Based on Petri Net Plans


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 200-211 inproceedings

DOI URL

Abstract: This paper presents a design of cooperative behaviors through Petri Net Plans, based on the principles provided by Cohen and Levesque?s Joint Commitments Theory. Petri Net Plans are a formal tool that has proved very effective for the representation of multi-robot plans, providing all the means necessary for the design of cooperation. The Joint Commitment theory is used as a guideline to present a general multi-robot Petri Net Plan for teamwork, that can be used to model a wide range of cooperative behaviors. As an example we describe the implementation of a robotic-soccer passing task, performed by Sony AIBO robots.
Palma-Amestoy, R., Guerrero, P., del Solar, J.R. & Garretón, C.

Bayesian Spatiotemporal Context Integration Sources in Robot Vision Systems


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 212-224 inproceedings

DOI URL

Abstract: Having as a main motivation the development of robust and high performing robot vision systems that can operate in dynamic environments, we propose a bayesian spatiotemporal context-based vision system for a mobile robot with a mobile camera, which uses three different context-coherence instances: current frame coherence, last frame coherence and high level tracking coherence (coherence with tracked objects). We choose as a first application for this vision system, the detection of static objects in the RoboCup Standard Platform League domain. The system has been validated using real video sequences and has presented satisfactory results. A relevant conclusion is that the last frame coherence appears to be not very important in the tested cases, while the coherence with the tracked objects appears to be the most important context level considered.
This research was partially supported by FONDECYT (Chile) under Project Number 1090250.
Pfingsthorn, M., Nevatia, Y., Stoyanov, T., Rathnam, R., Markov, S. & Birk, A.

Towards Cooperative and Decentralized Mapping in the Jacobs Virtual Rescue Team


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 225-234 inproceedings

DOI URL

Abstract: The task of mapping and exploring an unknown environment remains one of the fundamental problems of mobile robotics. It is a task that can intuitively benefit significantly from a multi-robot approach. In this paper, we describe the design of the multi-robot mapping system used in the Jacobs Virtual Rescue team. The team competed in the World Cup 2007 and won the second place. It is shown how the recently proposed pose graph map representation facilitates not only map merging but also allows transmitting map updates efficiently.
Phillips, M. & Veloso, M.

Robust Supporting Role in Coordinated Two-Robot Soccer Attack


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 235-246 inproceedings

DOI URL

Abstract: In spite of the great success of the fully autonomous distributed AIBO robot soccer league, a standing challenge is the creation of an effective planned, rather than emergent, coordinated two-robot attack, where one robot is the main attacker and goes to the ball and the other robot "supports the attack." While the main attacker has its navigation conceptually driven by following the ball and aiming at scoring, the supporter objectives are not as clear. In this work, we investigate this distributed, limited perception, two-robot soccer attack with emphasis on the overlooked supporting robot role. We contribute a region-based positioning of the supporting robot for a possible pass or for the recovery of a lost ball. The algorithm includes a safe path navigation that does not endanger the possible scoring of the teammate attacker by crossing in between it and the goal. We then further present how the supporter enables pass evaluation, under the concept that it is in a better position to visually assess a pass than the attacker, which is focused on the ball and surrounded by the opponent defense. We show extensive statistically significant lab experiments, using our AIBO robots, which show the effectiveness of the positioning algorithm compared both to a previous supporter algorithm and to a single attacker. Additional experimental results provide solid evidence of the effectiveness of our passing evaluation algorithm. The algorithms are ready to incorporate in different RoboCup standard platform robot teams.
Poppinga, J. & Birk, A.

A Novel Approach to Efficient Error Correction for the SwissRanger Time-of-Flight 3D Camera


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 247-258 inproceedings

DOI URL

Abstract: 3D data acquisition gets increasingly important for mobile robotics in general and for Safety, Security and Rescue Robotics (SSRR) in particular. 3D data allows for example to estimate the size of gaps, to construct realistic maps of unstructured disaster environments, or to detect human victims from shape. The SwissRanger SR-3000 time-of-flight 3D camera is a popular device for acquiring 3D range data as it offers the fast update rates of a camera with a resolution of 176 × 144 pixel at a reasonable cost. But the SR-3000 suffers - like any device using phase differences of modulated light - from the fundamental problem of wrap around error, i.e., distances that are a multiple of the wavelength of the modulated ranging signal can not be distinguished. The standard solution to this problem is to use an amplitude threshold, i.e., to discard pixels which are relatively dark and hence assumed to be far away. Here, a significant improvement to the standard method is presented that relates measured brightness and distance and also takes the geometry of the modulated light source into account. It is shown that a significantly higher amount of valid range data can be acquired with this new method.
This work was supported by the German Research Foundation (DFG).
Qining Wang, H.L., Huang, F., Xie, G. & Wang, L.

Collaborative Localization Based Formation Control of Multiple Quadruped Robots


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 649-659 inproceedings

DOI URL

Abstract: In this article, we present a new formation control method based on collaborative localization. Due to the accurate position estimation, the proposed approach is computationally light and well-performing in switching between different formations of real-world multi-robot systems. Satisfactory experimental results of multi-robot formation control are obtained in the RoboCup environment.
Rong, C., Wang, Q., Huang, Y., Xie, G. & Wang, L.

Autonomous Evolution of High-Speed Quadruped Gaits Using Particle Swarm Optimization


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 259-270 inproceedings

DOI URL

Abstract: This paper presents a novel evolutionary computation approach to optimize fast forward gaits for a quadruped robot with three motor-driven joints on each limb. Our learning approach uses Particle Swarm Optimization to search for a set of parameters automatically aiming to develop the fastest gait that an actual quadruped robot can possibly achieve, based on the concept of parameterized representation for quadruped gaits. In addition, we analyze the computational cost of Particle Swarm Optimization taking the memory requirements and processing limitation into consideration. Real robot experiments show that the evolutionary approach is effective in developing quadruped gaits. Satisfactory results are obtained in less than an hour by the autonomous learning process, which starts with randomly generated parameters instead of any hand-tuned parameters.
Sarika, R., Siddhartha, H. & Karlapalem, K.

Database Driven RoboCup Rescue Server


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 602-613 inproceedings

DOI URL

Abstract: The RoboCup Rescue Simulation, a generic urban disaster simulation environment constructed on a network of computers, has been in existence for many years. The server used in the simulation league has problems of scaling up. Further, it requires considerable effort to understand the server code to make any additional changes. Therefore, it is difficult for newcomers to quickly enhance the server. The architecture and the functional design of the current server are excellent. This helps us provide a database driven architecture that can scale up the current server to 10-15 times the number of agents that can be simulated. Moreover, it is now easy for others to implement many other subsystems that can provide additional functionality. We have also shown in this paper a new scoring strategy for agent teams which can be customized to emphasize, test and evaluate different concepts and strategies employed by the agent teams.
Savage, J., Weitzenfeld, A., Ayala, F. & Cuellar, S.

The Use of Scripts Based on Conceptual Dependency Primitives for the Operation of Service Mobile Robots


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 284-295 inproceedings

DOI URL

Abstract: This paper describes a Human-Robot interaction subsystem that is part of a robotics architecture, the ViRbot, used to control the operation of service mobile robots. The Human/Robot Interface subsystem consists of tree modules: Natural Language Understanding, Speech Generation and Robot's Facial Expressions. To demonstrate the utility of this Human-Robot interaction subsystem it is presented a set of applications that allows a user to command a mobile robot through spoken commands. The mobile robot accomplish the required commands using an actions planner and reactive behaviors. In the ViRbot architecture the actions planner module uses Conceptual Dependency (CD) primitives as the base for representing the problem domain. After a command is spoken a CD representation of it is generated, a rule base system takes this CD representation, and using the state of the environment generates other subtasks represented by CDs to accomplish the command. In this paper is also presented how to represent context through scripts. Using scripts it is easy to make inferences about events for which there are incomplete information or are ambiguous. Scripts serve to encode common sense knowledge. Scripts are also used to fill the gaps between seemingly unrelated events.
Schmits, T. & Visser, A.

An Omnidirectional Camera Simulation for the USARSim World


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 296-307 inproceedings

DOI URL

Abstract: Omnidirectional vision is currently an important sensor in robotic research. The catadioptric omnidirectional camera with a hyperbolic convex mirror is a common omnidirectional vision system in the robotics research field as it has many advantages over other vision systems. This paper describes the development and validation of such a system for the RoboCup Rescue League simulator USARSim.
After an introduction of the mathematical properties of a real catadioptric omnidirectional camera we give a general overview of the simulation method. We then compare different 3D mirror meshes with respect to quality and system performance. Simulation data also is compared to real omnidirectional vision data obtained on an 4-Legged League soccer field. Comparison is based on using color histogram landmark detection and robot self-localization based on an Extended Kalman filter.
Siedentop, C., Schwarz, M. & Pfülb, S.

Introducing Image Processing to RoboCupJunior PALB VISION – A First Implementation of Live Image Processing in RCJ Soccer


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 308-317 inproceedings

DOI URL

Abstract: For a long time, teams in the RoboCup Junior competition have relied on the same basic sensorical appliances. It is time to evolve. We believe that, the introduction of image processing to RoboCupJunior will take Soccer to a new level of intelligent and less aggressive gameplay. We found it especially challenging to design a system that could not only successfully detect the goal, but also any obstructions, i.e. robots from the opponent?s team, in order to score goals more precisely. This work aims to prove that the implementation of image processing does not need to be as much work as one might assume. Our project Palb Vision is designed to serve as an example for other teams in RCJ who plan to use camera-based detection software on their future robots.
We use a CMUcam3 with onboard processing connected to an ATmega 2560. The code for visual detection is written in C and runs directly on the CMUcam. The camera only takes into account the small area between the upper and lower edge of the wall and applies three simple filters for each pixel on a horizontal line.
We have created a quick and reliable vision system which processes ten frames per second and is very resistant to changing illumination. By implementing a special calibration mode, the pre-game setup is reduced to less than one minute. Palb Vision has turned out to be an improvement to the game of RoboCupJunior Soccer and may provide a robust framework for other teams who wish to adopt live image processing into their strategy.
del Solar, J.R., Palma-Amestoy, R., Vallejos, P., Marchant, R. & Zegers, P.

Designing Fall Sequences That Minimize Robot Damage in Robot Soccer


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 271-283 inproceedings

DOI URL

Abstract: In this paper is proposed a methodology for the analysis and design of fall sequences that minimize robot damage. This methodology minimizes joint/articulation injuries, as well as damage of valuable body parts (cameras and processing units). The methodology is validated using humanoid Nao robots and a realistic simulator. The obtained results show that fall sequences designed using the proposed methodology produce less damage than standard, uncontrolled falls.
This research was partially supported by FONDECYT (Chile) under Project Number 1061158.
Sprado, J. & Gottfried, B.

What Motion Patterns Tell Us about Soccer Teams


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 614-625 inproceedings

DOI URL

Abstract: A qualitative representation of motion patterns is presented that forms an interface between low-level concepts of behaviours and high-level concepts of reasoning. How the patterns can be employed for characterising interaction patterns in soccer is demonstrated using the simulation league; also, specific soccer scenes from real games prove their adequacy. The advantages of our approach are: it supports the limited abilities of robots in the different RoboCup leagues, i. e. it relies on coarse positional distinctions that are reliably obtainable and easily translated into action; the analysis is directly derived from raw data without the need for any preprocessing steps; both situations can be dealt with, egocentric viewpoints of individuals and the bird?s eye view; the approach is independent on the domain, i. e. generalises to arbitrary spatiotemporal interaction patterns.
Stanton, C.

Designing Grounded Agents: From RoboCup to the Real-World


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 626-637 inproceedings

DOI URL

Abstract: This paper discusses the nature and role of "grounding" in designing programs for controlling autonomous mobile robots. Since its inception, artificial intelligence has been plagued by problems of scaling and brittleness. A fundamental problem impeding the development of artificial intelligence is our dependence on grounding agents by design. That is, currently agents tend to be grounded by their designer's understanding of the world, task, and robot. However, little (if any) of the knowledge of "how to ground" is embedded in the artificial agent. Consequently, brittle, purpose-built systems result. This paper explores how the intellectual burden of grounding can be shifted from the programmer to the program by designing robots capable of grounding themselves. An overview of a grounding oriented design methodology (Go-Design) is presented - an initial step towards the longer-term objective of developing autonomous grounding capabilities.
Sun, D., Kleiner, A. & Wendt, T.M.

Multi-robot Range-Only SLAM by Active Sensor Nodes for Urban Search and Rescue


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 318-330 inproceedings

DOI URL

Abstract: To jointly map an unknown environment with a team of autonomous robots is a challenging problem, particularly in large environments, as for example the area of devastation after a disaster. Under such conditions standard methods for Simultaneous Localization And Mapping (SLAM) are difficult to apply due to possible misinterpretations of sensor data, leading to erroneous data association for loop closure. We consider the problem of multi-robot range-only SLAM for robot teams by solving the data association problem with wireless sensor nodes that we designed for this purpose. The memory of these nodes is utilized for the exchange of map data between multiple robots, facilitating loop-closures on jointly generated maps. We introduce RSLAM, which is a variant of FastSlam, extended for range-only measurements and the multi-robot case. Maps are generated from robot odometry and range estimates, which are computed from the RSSI (Received Signal Strength Indication). The proposed method has been extensively tested in USARSim, which serves as basis for the Virtual Robots competition at RoboCup, and by real-world experiments with a team of mobile robots. The presented results indicates that the approach is capable of building consistent maps in presence of real sensor noise, as well as to improve mapping results of multiple robots by data sharing.
Takahashi, T.

Analysis Methods of Agent Behavior and Its Interpretation in a Case of Rescue Simulations


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 331-342 inproceedings

DOI URL

Abstract: The agent-based approach has been proved to be useful for the modeling phenomena of traditional social fields. In order to apply agent-based simulation results to practical usages, it is necessary to show potential users the validity of the simulation outputs that arise out of the agents? behaviors. In cases that involve human actions, it is difficult to obtain sufficient amounts of data on real cases or experimental data to validate the simulation results.
In this paper, we review the metrics that have been used to evaluate rescue agents in the Rescue Simulation Agent competition and propose a method to analyze the simulation output by presenting the agent behavior with a probability model. We present that the analysis results of the method are comparable to a human-readable interpretation and with task-dependent knowledge and discuss its applicability to real-world cases.
Takahashi, Y., Tamura, Y. & Asada, M.

Spiral Development of Behavior Acquisition and Recognition Based on State Value


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 343-354 inproceedings

DOI URL

Abstract: Both self-learning architecture (embedded structure) and explicit/implicit teaching from other agents (environmental design issue) are necessary not only for one behavior learning but more seriously for life-time behavior learning. This paper presents a method for a robot to understand unfamiliar behavior shown by surrounding players through the collaboration between behavior acquisition and recognition of observed behavior, where the state value has an important role not simply for behavior acquisition (reinforcement learning) but also for behavior recognition (observation). That is, the state value updates can be accelerated by observation without real trials and errors while the learned values enrich the recognition system since it is based on estimation of the state value of the observed behavior. The validity of the proposed method is shown by applying it to a soccer robot domain.
Taleghani, S., Aslani, S. & Shiry, S.

Robust Moving Object Detection from a Moving Video Camera Using Neural Network and Kalman Filter


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 638-648 inproceedings

DOI URL

Abstract: Detecting motion of objects in images, while the camera is moving, is a complicated task. In this paper, we propose a novel method to effectively solve this problem by using Neural Network and Kalman Filter. This technique uses parameters of camera motion to overcome problems caused by error in the image processing outputs. We have implemented this technique in the MRL Middle Size Soccer Robots. The experimental results show a low error rate of 2.2% which suggests that the combined approach performs significantly better than the traditional techniques.
Varsadan, I., Birk, A. & Pfingsthorn, M.

Determining Map Quality through an Image Similarity Metric


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 355-365 inproceedings

DOI URL

Abstract: A quantitative assessment of the quality of a robot generated map is of high interest for many reasons. First of all, it allows individual researchers to quantify the quality of their mapping approach and to study the effects of system specific choices like different parameter values in an objective way. Second, it allows peer groups to rank the quality of their different approaches to determine scientific progress; similarly, it allows rankings within competition environments like RoboCup. A quantitative assessment of map quality based on an image similarity metric Ψ is introduced here. It is shown through synthetic as well as through real world data that the metric captures intuitive notions of map quality. Furthermore, the metric is compared to a seemingly more straightforward metric based on Least Mean Squared Euclidean distances (LMS-ED) between map points and ground truth. It is shown that both capture intuitive notions of map quality in a similar way, but that Ψ can be computed much more efficiently than the LMS-ED.
Warden, T., Lattner, A.D. & Visser, U.

Real-Time Spatio-Temporal Analysis of Dynamic Scenes in 3D Soccer Simulation


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 366-378 inproceedings

DOI URL

Abstract: We propose a framework for spatio-temporal real-time analysis of dynamic scenes. It is designed to improve the grounding situation of autonomous agents in (simulated) physical domains. We introduce a knowledge processing pipeline ranging from relevance-driven compilation of a qualitative scene description to a knowledge-based detection of complex event and action sequences, conceived as a spatio-temporal pattern matching problem. A methodology for the formalization of motion patterns and their inner composition is introduced and applied to capture human expertise about domain-specific motion situations. We present extensive experimental results from the 3D soccer simulation that substantiate the online applicability of our approach under tournament conditions, based on 5 Hz a) precise and b) noisy/incomplete perception.
Weitzenfeld, A., Ramos, C. & Dominey, P.F.

Coaching Robots to Play Soccer via Spoken-Language


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 379-390 inproceedings

DOI URL

Abstract: The objective of this paper and our current research is to develop a human-robot interaction architecture that will let human coaches train robots to play soccer via spoken language. This work exploits recent developments in cognitive science, particularly notions of grammatical constructions as form-meaning mappings in language, and notions of shared intentions as distributed plans for interaction and collaboration between humans and robots linking perceptions to action responses. We define two sets of voice-driven commands for human-robot interaction. The first set involves action commands requiring robots to perform certain behaviors, while the second set involves interrogation commands requiring a response from the robot. We then define two training levels to teach robots new forms of soccer-related behaviors. The first level involves teaching new basic behaviors based on action and interrogation commands. The second level involves training new complex behaviors based on previously learnt behaviors. We explore the two coaching approaches using Sony AIBO robots in the context of RoboCup soccer standard platform league previously known as the four-legged league. We describe the coaching process, experiments, and results. We also discuss the state of advancement of this work.
Work, H., Chown, E., Hermans, T., Butterfield, J. & McGranaghan, M.

Player Positioning in the Four-Legged League


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 391-402 inproceedings

DOI URL

Abstract: As RoboCup continues its march towards the day when robots play soccer against people, the focus of researchers' efforts is slowly shifting from low-level systems (vision, motion, self-localization, etc) to high-level systems such as strategy and cooperation. In the Four-Legged League (recently renamed to the Standard Platform League), teams are still struggling with this transition. While the level of play has consistently risen each year, teams continue to remain focused on low-level tasks. Surprisingly few of the 24 Four-Legged teams that competed at RoboCup 2007 were able to self-position at the beginning of the game, despite penalties incurred for not doing so. Things considered to be standard in 'real' soccer – positioning, passing, overall strategies – are still, after 10 years of research, far from a given within the league, and are arguably in their infancy compared to other RoboCup leagues (Small-Sized, Mid-Sized, Simulation). Conversely, for the top teams, many of these low-level systems have been pushed far enough that there is little to be gained in soccer performance from further low-level system work. In this paper we present a robust and successful player positioning system for the Four-Legged League.
Zhao, M., Zhang, J., Dong, H., Liu, Y., Li, L. & Su, X.

Humanoid Robot Gait Generation Based on Limit Cycle Stability


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 403-413 inproceedings

DOI URL

Abstract: This paper presents the gait generation and mechanical design of a humanoid robot based on a limit cycle walking method-Virtual Slope Control. This method is inspired by Passive Dynamic Walking. By shortening the swing leg, the robot walking on level ground can be considered as on a virtual slope. Parallel double crank mechanisms and elastic feet are introduced to the 5 DoF robot leg, to make the heelstrike of the swing leg equivalent to the point-foot collision used in Virtual Slope Control. In practical walking, the gait is generated by connecting the two key frames in the sagittal and lateral plane with sinusoids. With the addition of leg rotational movement, the robot achieves a fast forward walking of 2.0leg/s and accomplishes omnidirectional walking favorably.
Zickler, S. & Veloso, M.

Playing Creative Soccer: Randomized Behavioral Kinodynamic Planning of Robot Tactics


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 414-425 inproceedings

DOI URL

Abstract: Modern robot soccer control architectures tend to separate higher level tactics and lower level navigation control. This can lead to tactics which do not fully utilize the robot's dynamic actuation abilities. It can furthermore create the problem of the navigational code breaking the constraints of the higher level tactical goals when avoiding obstacles.
We aim to improve such control architectures by modeling tactics as sampling-based behaviors which exist inside of a probabilistic kinodynamic planner, thus treating tactics and navigation as a unified dynamics problem. We present a behavioral version of Kinodynamic Rapidly- Exploring Random Trees and show that this algorithm can be used to automatically improvise new ball-manipulation strategies in a simulated robot soccer domain. We furthermore show how opponent-models can be seamlessly integrated into the planner, thus allowing the robot to anticipate and outperform the opponent's motions in physics-space.
Özkucur, N.E., Kurt, B. & Akın, H.L.

A Collaborative Multi-robot Localization Method without Robot Identification


2009 RoboCup 2008: Robot Soccer World Cup XII, pp. 189-199 inproceedings

DOI URL

Abstract: This paper introduces a method for multi-robot localization which can be applied to more than two robots without identifying each of them individually. The Monte Carlo localization for the single robot case is extended using negative landmark information and shared belief state in addition to perception. A robot perceives a teammate and broadcasts its observation without identifying the teammate, and whenever a robot receives an observation, the observation is processed only if the robot decides that the observation concerns itself. The experiments are based on scenarios where it is impossible for a single robot to localize precisely and the shared information is ambiguous. We demonstrate successful robot identification and localization results in different scenarios.