International Journal of Humanoid Robotics

We describe here the humanoid challenge that is part of the RoboCup robot soccer competitions. We focus on how relevant research issues of humanoid robotics — for example, biped walking and human-like sensors and actuators — can be addressed, and we investigate how teams proceed to solve the given tasks. Thus, new technologies like artificial muscles and artificial skin might find their way into the competition very soon. We detail examples of these technologies and discuss in which way they may contribute to RoboCup, and in return how RoboCup may serve as a benchmark for achievements within these technologies. Further we describe how RoboCup works as an open, worldwide cooperative project in robotics and AI.

The RoboCup, which is a worldwide robot soccer competition, has set an ambitious goal for itself: to have a humanoid robot team win against human teams in World Cup Soccer by 2050. In order to achieve this goal, the robots require highly sophisticated sensory-data processing and decision-making functions. The development of robots for the RoboCup Humanoid League also has significant meaning for the development of robotics. However, this development is not easy and there are few papers covering it and its design policy. This paper reports the design policy for humanoids developed by Team Osaka, whose robots have been selected as the best humanoid robots four times in the last four years. In addition to the design policy, this paper also reports on the developmental process and comparisons among humanoid versions developed by Team Osaka. We believe that this paper will offer much information to other researchers who are developing humanoids for the RoboCup.

Humanoid soccer serves as a benchmark problem for artificial intelligence research and robotics. Every year, more teams are competing, for example, in the RoboCup Humanoid League. As the robots improve in the basic skills of walking, kicking, and getting up, teams can focus on soccer skills and team coordination. The complexity of soccer behaviors and team play calls for structured behavior engineering.

In this paper, we describe the design of the behavior control software for the Humanoid League team NimbRo. The software is based on a framework that supports a hierarchy of reactive behaviors. It is structured both as an agent hierarchy (joint–body part–player–team) and as a time hierarchy. The speed of sensors, behaviors, and actuators decreases when moving up in the hierarchy. The lowest levels of this framework contain position control of individual joints and kinematic interfaces for body parts. At the next level, basic skills are implemented. Soccer behaviors like searching for the ball, approaching the ball, avoiding obstacles, and defending the goal reside on the player level. Finally, on the tactical level, the robots communicate via a wireless network to negotiate roles and use allocentric information to configure the soccer behaviors.

Our robots performed very well at RoboCup 2007, which took place in Atlanta, Georgia, USA. They won both the KidSize and the TeenSize humanoid soccer competitions.

This paper presents an embedded control architecture constructed for Robo-Erectus, a soccer-playing humanoid robot developed at the Advanced Robotics and Intelligent Control Centre of Singapore Polytechnic. The Robo-Erectus team has participated in the KidSize category of RoboCup's Humanoid League since 2002, collecting different awards. The latest version of Robo-Erectus has many capabilities that can be exploited to improve the robot's behavior. The new embedded controller has made possible the first stage of the performance (displayed during RoboCup 2007), including network communication, mapping, and localization. The new mechanical, electronic design, embedded control architecture, and control schemes are described in this paper. In addition to the hardware, the paper presents details of the modules for gait generation, vision, behavior control, and communication.

Autonomous soccer games represent an extraordinary challenge for autonomous humanoid robots which must act fast and stable while carrying all needed onboard computers, sensors and batteries. In this paper, the development and system integration of hardware and software modules of the 55-cm tall, autonomous humanoid soccer robot Bruno is described to cope with this challenge. Although based on a "minimalistic" design which only uses gyroscopes in the hip but not foot-ground contact sensors for control of balance, versatile and high-quality walking motions have been developed. Fast forward walking of about 1.5 km/h has been obtained using an efficient sequential surrogate optimization method and walking through uneven terrain with a newly designed passively compliant foot sole. Further modules of the software and control architecture which are needed for an adaptive selection of different motions and autonomous robot behavior are briefly described. Experimental results are reported, which have been obtained under the conditions of a live competition. The robot's hardware is mainly based on standard components which can therefore be easily adapted by new designers, as no comparable, standard humanoid robot platforms are available.

This paper gives an overview of locomotion planning and control of a TeenSize humanoid soccer robot, Robo-Erectus Senior (RESr-1), which has been developed as an experimental platform for human–robot interaction and cooperative research in general and robotics soccer games in particular. The locomotion planning and control, along with an introduction of hierarchical control architecture, vision-based behavior and its application in the Humanoid TeenSize soccer challenge, are elaborated. The Estimation of Distribution Algorithm (EDA) is used in locomotion generation and optimization to achieves dynamically stable walk and a powerful kick. By setting different objective functions, smooth walking and powerful kicking can be generated quickly. RESr-1 made its debut at RoboCup 2007, and got fourth place in the Humanoid TeenSize penalty kick competition. In addition, some experimental results on RESr-1's walking, tracking and kicking are presented.

We present a team of two humanoid robot commentators for AIBO robot soccer games. The two humanoids stand by the side lines of the playing field, autonomously observe the game, wirelessly listen to a "game controller" computer, recognize events, and select announcing actions that may require coordination with each other. Given the large degree of uncertainty and dynamics of the robot soccer games, we further introduce a "Puppet Master" control that allows humans to intervene, prompting the robots to commentate an event if previously undefined or undetected. The robots recognize events based on input from these three sources, namely own and shared vision, game controller, and occasional Puppet Master. We present the two-humanoid behavioral architecture and the vision-based event recognition, including a SIFT-based vision processing algorithm that allows for the detection of multiple similar objects, such as the identical shaped robot players. We introduce the commentating algorithm that probabilistically selects a commentating action from a set of weighted actions corresponding to a detected event. The probabilistic selection uses the game history and updates the action weights to effectively avoid repetition of comments to enable entertainment. Our work, corresponding to a fully implemented system, CMCast, with two QRIO robots, contributes a team of two humanoids fully executing a challenging observation, modeling, coordination, and reporting task.

Motion planning for humanoids faces several challenging issues: high dimensionality of the configuration space, necessity to address balance constraints in single and double support mode, higher levels of planning for coordination of different skills, etc. While the above challenges hold for any humanoid robot, the soccer scenario adds difficulties rarely addressed in humanoid motion planning research, as for example: dynamic environments with active opponents, the requirement to perform short- and long-term plans for performing soccer-relevant actions, and the necessity to plan movements purposely terminating with a collision with the ball. These aspects open a completely new scenario for researchers. This paper surveys state-of-the-art research in motion planning for humanoid robots with a focus on outlining connections, differences, and identifying the key aspects that ought to be addressed when developing effective humanoid soccer players.

In this paper, we develop a general Lie group framework for analysis of kicking motion in humanoid soccer robots which aims to capture the multidimensional kicking information and hence to study how to develop more powerful and skilful kicking motion for humanoid soccer robots. To maintain dynamic stability while kicking is being performed, the zero-moment point (ZMP) is used to evaluate the performance of the humanoid kick. The proposed Lie-group-formulation-based compensation approach for force/torque sensing from the humanoid ankle has been implemented on, Robo-Erectus, our humanoid soccer robot. Humanoid kicking experiments have been conducted to verify the effectiveness of the proposed approach.

This paper describes our experience in the simulation of humanoid soccer robots using two general purposes 3D simulators, namely USARSim and Microsoft Robotics Studio. We address the problem of the simulation of a soccer match among two teams of small humanoid robots in the RoboCup Soccer Kid-Size Humanoid competitions. The paper reports the implementation of the virtual models of the Robovie-M humanoid robot platform in the two simulators. Robovie-M was the robot used by our team "Artisti" in the RoboCup 2006 competitions. This paper focuses on the procedures needed to implement the virtual models of the robot and in the details of the models. We describe experiments assessing the feasibility and the fidelity of the two simulators.