Fig 2 - uploaded by Joschka Boedecker
Content may be subject to copyright.
An exemplary robot model of Fujitsu's HOAP-2 that could be used in the RoboCup-2007 simulation league competition.
Source publication
We present a road map for a joint project of the simulation league and the humanoid league that we call 3D2Real. This project is concerned with the integration of these two leagues which is becoming increasingly important as the research fields are converging. Currently, a lot of work is duplicated across the leagues, collaboration is sparse, and k...
Context in source publication
Context 1
... are very much simplified. Methods to create soccer playing agents for a team have to deal with a higher complexity of the environment, and hopefully can be transferred to humanoid robotics more easily. The current development of 3D Soccer Simulation League leads to simple two-legged agents used in technical challenges already this year (see also Fig. ...
Similar publications
In the context of imitation learning, several approaches have been developed so as to transfer human skills to robots, with demonstrations often represented in Cartesian or joint space. While learning Cartesian positions suffices for many applications, the end-effector orientation is required in many others. However, several crucial issues arising...
Citations
... The respective agents could be evaluated in separate simulation experiments or even concurrently in a competitive manner. The latter case, for instance, is applied in the RoboCup soccer simulation league [13], while the USAR-Sim system of RoboCup rescue league [3] rather pertains to the first case. However, the RoboCup simulations do not focus on distributed simulations and are limited to a specific simulation domain. ...
This paper presents the PlaSMA multiagent-based simulation system and its application for collabora tive experi- mentation. PlaSMA enables large-scale parallel and distributed simulations and relies on Java for platform indepen dence. The applied agent-based simulation approach particularly suits vir- tual organizations in research because it enables s hared comput- ing resources as well as collaborative experimentat ion with com- mon simulation scenarios.
... Dylla et al.[5] investigated to model soccer knowledge in a way that they are usable for multiple RobCup soccer leagues. Mayer et al.[11] proposed a road map to collaboration between Soccer Simulation League and Humanoid League. At the same time, the USARsim[15] simulator is used as bridging tool between the Rescue Real Robot League and the Rescue Simulation League. ...
Numerous simulators have been developed over the years to assist robotics research in the development, testing, and evaluation. Nevertheless, there is still a big gap between the simulation and the reality. This makes it difficult to transfer methods and code. The 3D simulator | SimSpark is developed and used by a big community of AI researchers in RoboCup. But up to now there are only few applications to real robots. In this paper, we discuss the general possibilities how the SimSpark simulator can be used to support research in cognitive robotics and present applications on the humanoid robot Nao. As a result of our investigation we have developed a unified team playing both in Simulation League and Standard Platform League in RoboCup.
... However, it soon became apparent that while interesting in itself, this version of Sim-3D was lacking direct relevance to the long-term goal of playing soccer with humanoid robots. Indeed in their 2007 paper, Mayer et al. [13] argued that the simulation league should embrace humanoid robots as early as possible. To this end, in 2007, Sim-3D transitioned to a more realistic and challenging humanoid model: the Soccerbot, based on the Fujitsu HOAP-2 robot 2 . ...
... In the following section, we argue that in coming to terms with the multiple challenges ahead, the three platforms considered in this paper should concentrate on separate issues, in order that their combination will be most effective in advancing towards the goals of RoboCup. As Mayer et al. [13] observe, it will become important to start integrating the various leagues in RoboCup; in the year 2050, one " league " will stand to summarize the progress made over the years. At this stage, humanoid robotics still faces a broad range of problems, chief among which is reliable bipedal locomotion . ...
In this article, we provide an overview of three humanoid soccer platforms currently in use at RoboCup: 3D simulation, the hu-manoid Standard Platform League (SPL), and the Webots-based sim-ulator released with the SPL. Although these platforms trace different historical roots, today they share the same robot model, the Aldebaran Nao. Consequently, they face a similar set of challenges, primary among which is the need to develop reliable and robust bipedal locomotion. In this paper, we compare and contrast these platforms, drawing on the experiences of our team, UT Austin Villa, in developing agents for each of them. We identify specific roles for these three platforms in advancing the overarching goals of RoboCup.
... 2). The details of the current state of the 3D Soccer Simulation League are discussed in publications that relate to the 3D2Real project [8,9]. The goal that is envisioned for the 3D2Real project is to have the finals of the soccer simulation league using real robots in the near future. ...
... For this ambitious goal several steps are necessary in the next years to create the necessary infrastructure and tools. According to the proposed road map in [8], a technical challenge would be held at the RoboCup competitions to test the ability to use the agent code of SSL participants on a predetermined real robot. We propose also the development of a central parts repository (CPR, see also [18]). ...
... at the RoboCup competitions to test the ability to use the agent code of SSL participants on a predetermined real robot. We propose also the development of a central parts repository (CPR, see also [18]). This would be a collection of real robot designs, sensor and actuator models, complete robots, as well as controllers for certain architectures. [8]). The HMDP may serve at the same time as an interface for abstract motion description between agent and simulator as well as between agent and robot (see also [9]). The abbreviations stand for 3D Soccer Simulation League (SL), Soccer Humanoid League (HL); NAO is the name of the robot used in the new Standard Platform League. For details ...
We describe the Harmonic Motion Description Protocol (HMDP), that can serve as a part in tools for rapid prototyping of behaviors in a hybrid simulation real robot environment. In particular, we are focusing on the RoboCup 3D Soccer Simulation League server that is currently evolving rapidly, and becomes a more and more useful open source, general purpose simulation environment for physical Multiagent systems. HMDP uses harmonic functions to describe motions. It allows for superposition of motions and is therefore capable of describing parametric motions. Thus, typical open loop motions (walking on spot, forward, turning, standing up) of small humanoid robots are readily available and can be optimized by hand. In conjunction with the HMDP some software tools and a small real-time motion generator (called Motion Machine) have been developed. The current implementation is very flexible to use and can easily be implemented in rather small embedded systems.
... Setting-up an experiment can be a very time-consuming process, and often requires programming knowledge or knowledge of 3D modelling languages such as VRML. We chose a different approach and designed our own description language: RoSiML (Robot Simulation Markup Language) [21, 22]. This was done for one main reason. ...
... The first version of RoSiML was used in the German Research Foundation Priority Program 1125 as a general simulator description language to make descriptions exchangeable among the program members (e.g. [21, 22]), independent of the simulation system. <RoSiML> <simulator> <!--simulator configuration --> <!--cam. ...
This paper presents YARS (Yet Another Robot Simulator), which was initially developed in the context of evolutionary robotics (ER), yet includes features which are also of benefit to those outside of this field. An experiment in YARS is defined by a single XML file, which includes the simulator configuration, the (randomisable) environment, and any number of (mobile) robots. Robots are either controlled through an automatised communication, or by dynamically loaded C++ programs. Therefore, YARS, although still under active development, is comparable with commercial and open-source robot simulators which include a physics engine such as Webots and Breve but with a much stronger focus on requirements originating from the field of evolutionary robotics.
... One has to see how the development in these leagues evolves and the what specific research topics are going to turn out to be the most interesting. One initial step towards the coordination of the 3D Soccer Simulation League and the Humanoid League is the 3D2Real project[20], [8], [19]. The aim is here to port the tactical know how of the simulation league to real humanoid robot platforms and to test how synergies of the two different leagues can be used for the benefit of the RoboCup project and the year 2050 goal. ...
We describe here the project of the humanoid challenge that is part of the RoboCup robot soccer competitions. We focus on how the relevant research issues of humanoid robotics -for example biped walking and human-like sensors and actuators- can be addressed and we investigate how the teams proceed to solve the given tasks. Thus, new technologies like artificial muscles and artificial skin might find its way into the competition very soon. We go into details of examples of these technologies and discuss in which way they may contribute to the RoboCup and in return how the RoboCup may serve as a benchmark for achievements within these technologies. Further, we describe how the RoboCup works as an open, worldwide cooperative project in robotics and AI. I. INTRODUCTION
... Webots [4], thë Ubersim [5], SimRobot [6], or Microsoft Robotics Studio [7] . Also the RoboCup Simulation League introduced a fully dynamic robot simulation , additionally aiming towards a closer cooperation with real robots [8] . All these simulations allow a detailed specification of the environment, but demand the user to do this manually what might become an exhausting task given the high number of environmental parameters. ...
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.
... On the roadmap towards simulation league finals in real robots [3], the 3D soccer simulation league has taken a big step forward to the RoboCup's ultimate long-term goal this year: the humanoid robot simulation was used for the first time in RoboCup-2007 1 (See Fig. 1). Fig. 1. ...
This paper proposes the layered omnidirectional walking controller for the humanoid soccer robot. The gait of the robot can be parameterized using the destination posititon and the desired direction while reaching the destination. Its implementation in our RoboCup simulation team - SEU-3D is detailed in this paper. Our approach generates smooth robot trajectories without stop before changing direction or turning, and is fast enough to meet the real-time requirements. The proposed approach has been tested in the RoboCup soccer 3D server platform. The results showed that omnidirectional walking has advantages in dynamic environments.
... Currently, work on a more realistic camera perceptor has begun which will deliver images rendered via OpenGL hardware accelerated offscreen buffers. Besides that, a new effector has reached its first usable version which implements the Harmonic Motion Description Protocol (HMDP) as presented in [16] (both as part of the 3D2real project [17]). In the near future, several significant changes are planned to parts of the SimSpark architecture; in order to take advantage of the rapid improvement of several freely available physics engines, it is planned to integrate support for the Physics Abstraction Layer [18] library, providing support for ODE, PhysX [19], and others; the internal rendering library Kerosin is planned to be replaced by support for state-of-the-art open-source solutions like OGRE 3D [20] and the GUI will be extended; the network protocols will be optimized for faster information transmission. ...
Simulators have a long tradition within RoboCup. In the Soccer Simulation League, one of the earliest leagues within RoboCup, traditionally single-purpose robot simulators with high levels of abstrac-tion have been used. These systems proved valuable as tools for mul-tiagent research, but were difficult to extend. Furthermore, there were concerns that research results would not be easily transferable to the real world due to the extensive simplifications. In order to alleviate these problems, a new, more generic, and more realistic robot simulation sys-tem was proposed in the year 2003 and first used in the league's com-petitions in 2004. This paper presents the architecture and concepts of this system, and its application in the RoboCup 3D Soccer Simulation League. Moreover, it presents ongoing and future development plans.
... Otherwise, the simulated robot might not only behave unrealistic but could fail completely. All previously listed simulators789 as well as the RoboCup Simulation League's fully dynamic robot simulation SimSpark, which additionally aims towards a closer cooperation with real robots [14], allow a detailed specification of the environment. But all of them demand the user to do this manually what might become an exhausting task given the high number of environmental parameters . ...
This paper describes SimRobot, a robot simulator which is able to simulate arbitrary user-defined robots in three-dimensional space. It includes a physical model which is based on rigid body dynamics. To allow an extensive flexibility in building accurate models, a variety of different generic bodies, sensors, and actuators has been implemented. Furthermore, the simulator follows a user-oriented approach by includ-ing several mechanisms for visualization, direct actuator manipulation, and interaction with the simulated world. To allow a more detailed sim-ulation, algorithms for simulating image disturbances as well as for ac-tuator parameter optimization have been added. During the past years, SimRobot has been used to simulate several different robots, of which some are presented in this paper.
