Robotic Localization Service Standard for Ubiquitous Network Robots

摘要

In this paper, we have presented the concept behind a new framework for treating and exchanging location information in network robotic systems. One of the issues regarding frameworks of this kind is how vendor-specific functionalities or raw data can be represented. Although in this paper we have focused on describing the representation of localization data, our framework includes elements such as data format or vendor-specific parameter specification that can be utilized for treating vendor specific data or interfaces. However note that most of the existing outputs (including raw sensor data) follows some rule. This means these data is defined over some kind of system, and this allows the data to be represented by means of CSs in most of the realistic cases. Although the proposed framework has no ability to treat grammar-based complex data flows, in realistic cases this limitation can be avoided by defining symbolic axes based on codebooks such as the symbolic information described above. Although the formal standard document has not been published yet, several companies and research projects have started to implement and use this specification. One such system is the Kansai Environmental Platform (14). The Kansai Environment is a field-trial environment for robotic services where sensor network is installed in a real world location such as shopping mall. Various types of sensors are installed in the environment to produce information on the environment such as where people is, how people are behaving or how a certain area (space) is used. This is done by performing people tracking and real-time pattern recognition based on statistical processing of the measured trajectories. These results are sent to servicing robots in the environment on request. By utilizing these information, robots can easily provide useful services to people in the environment. Here, RLS specification is extended for passing these