HRP-2W: A humanoid platform for research on support behavior in daily life environments

doi:10.1016/j.robot.2008.10.014

Robotics and Autonomous Systems

Volume 57, Issue 2, 28 February 2009, Pages 145-154

https://doi.org/10.1016/j.robot.2008.10.014 Get rights and content

Abstract

We introduce a concept of a real-world-oriented humanoid robot that can support humans’ activities in daily life. In such environments, robots have to watch humans, understand their behavior, and support their daily life tasks. In particular, these robots must be capable of such real-world behavior as handling tableware and delivering daily commodities by hand. We developed a humanoid robot, HRP-2W, which has an upper body of HRP-2 [K. Kaneko, F. Kanehiro, S. Kajita, H. Hirukawa, T. Kawasaki, M. Hirata, K. Akachi, T. Isozumi, Humanoid Robot HRP-2, in: Proceedings of the 2004 IEEE International Conference on Robotics & Automation, 2004, pp. 1083–1090] and a wheel module instead of legs, as a research platform to fulfill this aim. We also developed basic software configuration in order to integrate our platform with other research groups. Through experiments, we demonstrated the feasibility of the humanoid robot platform and the potential of the software architecture.

Introduction

Agent systems, which can communicate with humans and offer solutions for problems, have recently been gaining a great deal of attention. At the same time, development of humanoid robots is also under intense study. Integrating these two technologies would bring us closer to the realization of real-world-oriented humanoid robots. Such real-world-oriented robots have to watch humans, understand their behaviors, and support their daily life tasks. Real-world behavior, especially such tasks as handling tableware and delivering daily commodities by hand, is gaining importance in real-world environments.

Following this background, research on humanoid robots tends to focus on content-oriented systems rather than individual functions such as manipulation, image understanding, and motion control. Interaction between humans, including gesture understanding, natural language communication, teaching by showing, and storage and reuse of communication experience, has been well studied.

Bischoff et al. have developed HERMES [2], a humanoid robot that can interact with humans and understand their instructions by natural language. In addition, Thrun et al. have developed a tour-guide robot in a museum environment. It can build environment maps using distance sensors, navigate in a museum, and communicate with audiences. Our approach considers not only such communication functions but also physical interaction between humans as well as functions of watching humans’ behavior. Fig. 1 shows an image of the ultimate goal of our project: producing robots that always pay attention to humans’ behavior, forecast the humans’ intentions, and support the daily life behavior before the humans ask for help. In this paper, we describe the methodology for realizing real-world-oriented humanoid robots.

In Section 2, we propose a humanoid robot HRP-2W as a research platform. Hardware and software configuration of the robot are shown. In Section 3, sub-functions such as object recognition, human gesture recognition and hearing of sound are explained. Garbage segregation application is shown as an instance of an integrated system on HRP-2W. In Section 4, an on-line motion acquisition subsystem is proposed for real-time and flexible behavior design in unexpected real environments. In Section 5, we discuss and conclude the performance of the humanoid robot HRP-2W.

Section snippets

Concept

We developed a humanoid robot platform “HRP-2W” which has an upper body of HRP-2 [1], for research on support behavior in daily life environments. The outline of the robot is shown in Fig. 2. Similar humanoids with wheel unit have already been proposed [2], [3]. One of the differences between those projects and ours is the realization of long-term behavior acquisition and performance in both indoor and outdoor environments. Four large battery packs inside its body enable about four hours of

Basic functions for daily life supporting behavior

In this section, we pick up and explain fundamental sub-functions of HRP-2W from the viewpoint of realization of examples of daily life supporting behavior (Fig. 4). Required sub-functions for supporting behavior consists of observation, manipulation and recognition of target objects and users; understanding of human gestures and making communication are also important sub-functions.

On-line behavior learning and acquisition

The main target tasks of this research are daily life behavior, such as handling of tableware, cleaning of furniture, and operation of home information appliances. Designing autonomous behavior in advance to achieve above tasks is difficult for developers, because models of the tools are difficult to predict and acquire in unknown environments. Therefore, we have added on-line behavior learning and acquisition functions to act in each case.

As user interfaces for the learning and acquisition of

Summary and conclusions

We have introduced a concept of a real-world-oriented humanoid robot that can support humans’ activities in daily life. As a research platform, we developed a humanoid robot, HRP-2W, which has a wheel module instead of legs. For realization of supporting behavior in daily life environments, the following functions are embedded as a basic library: (1) understanding of humans’ gesture based on binocular cameras, (2) construction of daily commodity database and recognition of objects using vision

References (10)

K. Kaneko, F. Kanehiro, S. Kajita, H. Hirukawa, T. Kawasaki, M. Hirata, K. Akachi, T. Isozumi, Humanoid Robot HRP-2,...
R. Bischoff et al.
Hermes–an intelligent humanoid robot, designed and tested for dependability
S. Hashimoto, H.T., et al. Humanoid Robots in Waseda University - Hadaly-2 and WABIAN -, in: Proc. of IEEE-RAS...
H. Ishiguro et al.
Robovie: An interactive humanoid robot
International Journal of Industrial Robotics
(2001)
S. Thrun, M. Bennewitz, W. Burgard, A. Cremers, F. Dellaert, D. Fox, D. Hahnel, C. Rosenberg, N. Roy, J. Schulte, D....

There are more references available in the full text version of this article.

Cited by (19)

Autonomous SLAM based humanoid navigation in a cluttered environment while transporting a heavy load
2018, Robotics and Autonomous Systems
Citation Excerpt :
Robots with humanoid arms have better control on the structure of an object and are therefore more suitable to control various shaped objects. More advanced and specialized models of wheeled robots can possess humanoid torsos and arms which give them the same capacity to keep a high level of control when holding or transporting objects [7–10]. However, most environments, made by and for humans, are more suitable for humanoid robots.
Although in recent years there have been quite a few studies aimed at the navigation of robots in cluttered environments, few of these have addressed the problem of robots navigating while moving a large or heavy objects. This is especially useful when transporting loads with variable weights and shapes without having to change the robot hardware. Inspired by the wide use of makeshift carts by humans, we tackle, in this work, the problem of a humanoid robot navigating in a cluttered environment while displacing a heavy load that lies on a cart-like object. We present a complete navigation scheme, from the incremental construction of a map of the environment and the computation of collision-free trajectories to the control to execute these trajectories. Our contributions are as follows: (1) a whole-body control scheme that makes the humanoid use its hands and arms to control the motions of the cart–load system (e.g. tight turns) (2) a sensorless approach to automatically select the appropriate primitive set according to the load weight (3) a motion planning algorithm to find an obstacle-free trajectory using the appropriate primitive set and the constructed map of the environment as input (4) an efficient filtering technique to remove the cart from the field of view of the robot while improving the general performances of the SLAM algorithms and (5) a continuous and consistent odometry data formed by fusing the visual and the robot odometry information. We present experiments conducted on a real Nao robot, equipped with an RGB-D sensor mounted on its head, pushing a cart with different loads. Our experiments show that the payload can be significantly increased without changing the robot’s main hardware, and therefore enacting the capacity of humanoid robots in real-life situations.
Static force capability optimization of humanoids robots based on modified self-adaptive differential evolution
2017, Computers and Operations Research
Citation Excerpt :
Therefore, the topic about humanoid robots is one of the most important and challenging in the field of robotics. Worldwide, there are a number of humanoid robots that have been recently built such as the prototypes HRP-2W [5], Honda [6], WABIAN [7], Bonten–Maru [8], Johnnie [9], and HanSaRam [10]. However, there are few studies such as [11] that deal with the static force capability of these machines, which is necessary to estimate the limits supported by the mechanism and ensure the accomplishment of a particular task, specifically in high load situations.
The current society requires solutions for many problems in safety, economy, and health. The social concerns on the high rate of repetitive strain injury, work-related osteomuscular disturbances, and domestic issues involving the elderly and handicapped are some examples. Therefore, studies on complex machines with structures similar to humans, known as humanoids robots, as well as emerging optimization metaheuristics have been increasing. The combination of these technologies may result in robust, safe, reliable, and flexible machines that can substitute humans in multiple tasks. In order to contribute to this topic, the static modeling of a humanoid robot and the optimization of its static force capability through a modified self-adaptive differential evolution (MSaDE) approach is proposed and evaluated in this study. Unlike the original SaDE, MSaDE employs a new combination of strategies and an adaptive scaling factor mechanism. In order to verify the effectiveness of the proposed MSaDE, a series of controlled experiments are performed. Moreover, some statistical tests are applied, an analysis of the results is carried out, and a comparative study of the MSaDE performance with other metaheuristics is presented. The results show that the proposed MSaDE is robust, and its performance is better than other powerful algorithms in the literature when applied to a humanoid robot model for the pushing and pulling tasks.
MAHRU-M: A mobile humanoid robot platform based on a dual-network control system and coordinated task execution
2011, Robotics and Autonomous Systems
Citation Excerpt :
In recent years, there has been a great deal of interest in humanoid robots. Particularly, intelligence service robots are the main type of the humanoid robot that has been studied [1–3]. To service a human being, it is necessary to develop a coordinated robot system that can execute various tasks in everyday environments.
This paper introduces a mobile humanoid robot platform able to execute various services for humans in their everyday environments. For service in more intelligent and varied environments, the control system of a robot must operate efficiently to ensure a coordinated robot system. We enhanced the efficiency of the control system by developing a dual-network control system. The network system consists of two communication protocols: high-speed IEEE 1394, and a highly stable Controller Area Network (CAN). A service framework is also introduced for the coordinated task execution by a humanoid robot. To execute given tasks, various sub-systems of the robot were coordinated effectively by this system. Performance assessments of the presented framework and the proposed control system are experimentally conducted. MAHRU-M, as a platform for a mobile humanoid robot, recognizes the designated object. The object’s pose is calculated by performing model-based object tracking using a particle filter with back projection-based sampling. A unique approach is used to solve the human-like arm inverse kinematics, allowing the control system to generate smooth trajectories for each joint of the humanoid robot. A mean-shift algorithm using bilateral filtering is also used for real-time and robust object tracking. The results of the experiment show that a robot can execute its services efficiently in human workspaces such as an office or a home.
Applications of electro-hydraulics actuators
2016, Conference Proceedings - IEEE SOUTHEASTCON
On the improvement of static force capacity of humanoid robots based on plants behavior
2016, ESANN 2016 - 24th European Symposium on Artificial Neural Networks
Adaptive asymptotic stable biped locomotion
2014, Proceedings of the 33rd Chinese Control Conference, CCC 2014

View all citing articles on Scopus

Tetsunari Inamura is an Associate Professor in the Principles of Informatics Research Division in the National Institute of Informatics, and an Associate Professor in the department of Informatics, School of Multidisciplinary Sciences, the Graduate University for Advanced Studies (SOKENDAI). He received BE in 1995 from the Department of Mechano-Informatics at the University of Tokyo, and MS and PhD in the Department of Informatics Engineering, School of Engineering, at the University of Tokyo in 1997 and 2000. He conducted research at the CREST program, Japanese Science and Technology Cooperation from 2000 to 2003, and then joined the Department of Mechano-Informatics at School of Information Science and Technology at the University of Tokyo as a Lecturer from 2003 to 2006. His research interests include imitation learning and symbol emergence on humanoid robots, development of interactive humanoid robot, stochastic information processing and so on. He received a Funai encouraging prize in 2003, Best paper award from Japanese Society for Artificial Intelligence in 2003 and 2006, and Young researcher encouraging award from Robotics Society of Japan in 2008. He is a member of IEEE Robotics and Automation Society, Robotics Society of Japan, Japanese Society for Artificial Intelligence, and so on. Currently he is also a Project Associate Professor in the Information and Robot Technology Research Initiative, The University of Tokyo.

Kei Okada received the BE in Computer Science from Kyoto University in 1997. He received the M.S. and the Ph.D. in Information Engineering from The University of Tokyo in 1999 and 2002 respectively. From 2002 to 2006, he joined the Professional Programme for Strategic Software Project in The University Tokyo. Since March 2006, he has been a project assistant professor in Creative Informatics at The University of Tokyo. His research interests include humanoid robots, real-time 3D computer vision, and recognition-action integrated system. He is a member of IEEE, Information Processing Society of Japan and the Robotics Society of Japan.

Satoru Tokutsu is a Ph.D. candidate student in the Department of Creative Informatics, Graduate School of Information Science and Technology, the University of Tokyo. He is a Research Fellow of the Japan Society for the Promotion of Science (JSPS). He graduated from the Department of Mechano-Informatics, Faculty of Engineering, The University of Tokyo in 2006 and he finished his masters in the Department of Mechano-Informatics, Graduate School of Informatics Science and Technology in 2008. His research is on daily life assistive humanoid behaviors with audio-visual perception He is a member of The Robotics Society of Japan and of The Institute of Electronics, Information and Communication Engineers (IEICE).

Naotaka Hatao received the B.E and M.E from the University of Tokyo in 2005 and 2007 respectively. Since 2007, he has been doing a doctoral course of the University of Tokyo. His current research interests include SLAM, path planning of a mobile robot, and environment recognition. He is a member of Information Processing Society of Japan and the Robotics Society of Japan.

Masayuki Inaba is a Professor in the Department of Mechano-Informatics, School of Information Science and Technology, at the University of Tokyo. He graduated from the Department of Mechanical Engineering at the University of Tokyo in 1981, and received MS and PhD degrees from the Graduate School of Information Engineering at the University of Tokyo in 1983 and 1986. He was appointed as a Lecturer in the Department of Mechanical Engineering at the University of Tokyo in 1986, an Associate Professor in 1989 and a Professor in the Department of Mechano-Informatics in 2000. His research interests include key technologies of robotic systems and software architectures to advance robotics research. His research projects have included hand–eye coordination in rope handling, vision-based robotic server systems, remote-brained robot approach, whole-body behaviors in humanoids, robot sensor suits with electrically conductive fabric and developmental software systems for evolving a robot body with a spine. He received several awards including outstanding Paper Awards in 1987, 1998 and 1999 from the Robotics Society of Japan, JIRA Awards in 1994, ROBOMECH Awards in 1994 and 1996 from the division of Robotics and Mechatronics of Japan Society of Mechanical Engineers, and Fellow of the Robotics Society of Japan in 2008. Currently, he is a chair of the Department of IRT Systems Study, Information and Robot Technology Research Initiative, The University of Tokyo.

Hirochika Inoue is a honorary professor in the Department of Mechano-Informatics, School of Information Science and Technology, at the University of Tokyo. He received BE, ME and Dr.Eng. degrees in 1965, 1967 and 1970, respectively, all from the University of Tokyo, in Mechanical Engineering. He conducted research at the Robotics Division of the Electrotechnical Laboratory, MITI Japan from 1970 to 1978. He joined the faculty at the University of Tokyo, Department of Mechanical Engineering as an Associate Professor in 1978 and became a Professor in 1984. Over the past four decades he has conducted research at the core of robotics, including bilateral control of robot arms, visually guided robot motion, development of parallel manipulators, development of high-speed robot vision, learning by seeing, real-time tracking vision systems and integrated robot systems such as humanoids. His research interests cover wide aspects of robotics: mechanical design, force control, vision-based robotics, language, planning and system integration. He has earned numerous awards and has served as the project leader for large-scale Inter-University research programs on intelligent robotics sponsored by the Ministry of Education from 1995 to 1998, a JSPS Research for the Future Project on soft robotics, and the Japanese National Project for Humanoid Robot R&D. He is also serving as President of the Board of the International Foundation of Robotics Research. Currently, he is a inspector general of Japan Society for the Promotion of Science.

View full text