Robots Concientes Del Contexto

Developing a Context-Aware System for Providing Intelligent Robot Services*
Chung-Seong Hong1, Joonmyun Cho1, Kang-Woo Lee1, Young-Ho Suh1, Hyun Kim1, and Hyun-Chan Lee2
Electronics and Telecommunication Research Institute Intelligent Robot Research Division, 161 Gajeongdong, Yuseonggu Daejeon, Republic of Korea {cshong, jmcho, kwlee, yhsuh, hyunkim}@etri.re.kr 2 Hongik Univ., Dept. ofInformation and Industrial Eng., Sangsudong, Mapogu Seoul, Republic of Korea hclee@hongik.ac.kr
1

Abstract. To realize the vision of ubiquitous computing, it is important to develop a context-aware system which can help ubiquitous agents, services, and devices become aware of their contexts. Context-aware system is required to be capable of configuring appropriate context model and managing it throughthe context manipulation. In this paper, we propose the context space and the conceptual structure for context acquisition, representation, and utilization. We also describe how this conceptual structure is implemented in the service robot area. As the result of the research, we developed the CAMUS (Context-Aware Middleware for URC Systems) which is a context-aware server framework for anetwork-based intelligent robot system and applied it to the field test.

1 Introduction
Ubiquitous computing is an emerging paradigm of personal computing, characterized by the shift from dedicated computing machinery to ubiquitous computing capabilities embedded in our everyday environments. In the vision of ubiquitous computing environment, computer system understands their situational contexts andprovides appropriate services to users [1]. To realize this vision, it is important to develop a context-aware system which can help ubiquitous agents, services, and devices become aware of their contexts because such computational entities need to adapt themselves to changing situations [2, 3]. Context is an available, detectable, and relevant piece of information or attribute about thecircumstances, objects, or conditions surrounding a user at the time of interaction between the user and the computing environment. In contrast to the traditional computing paradigm, in which the computing environment is often static and welldefined, in Ubiquitous Computing the underlying environment is open and dynamic [4]. In an open and dynamic environment, contexts in question are often distributed inheterogeneous sources.
*

This work was supported in part by MIC & IITA through IT Leading R&D Support Project.

P. Havinga et al. (Eds.): EUROSSC 2006, LNCS 4272, pp. 174 – 189, 2006. © Springer-Verlag Berlin Heidelberg 2006

Developing a Context-Aware System for Providing Intelligent Robot Services

175

As context-aware computing gets more and more interests, attempts applying it todiverse kinds of research fields, such as service robots, have been tried. In order that service robots can be introduced in our daily lives as if they were digital appliances and information devices, they should be able to provide various services that a user wants with context information. However, the functionalities of current service robots are neither worth their costs, nor do they havekiller applications. Under this background, ETRI (Electronics and Telecommunications Research Institute) has developed a new concept of a network-based intelligent robot, which is called as URC (Ubiquitous Robotic Companion) [5]. A network-based robot proposed in the URC can sufficiently utilize robot’s own functionalities such as sensing, processing and acting by distributing its internal abilitiesto the external resources through the network as shown in Fig. 1. That is, it is enabled to utilize external sensors embedded in the environment to expand internal sensors of the robot itself. Also, a number of robots can use the high performance remote servers to increases the internal computing power. The improvement of sensing abilities brings enhancement to context awareness about external...