Affiliations: Graduate School of Science and Technology, Keio University, Hiyoshi 3-14-1, Kohoku-ku, Yokohama, Japan | Faculty of Science and Technology, Keio University, Hiyoshi 3-14-1, Kohoku-ku, Yokohama, Japan E-mail: [email protected]
Abstract: This paper proposes an environmental sensor bridge system named COSPI for autonomous communication robots. In the near future, various sensors will get installed in day-to-day environments and robots will use information from these sensors to recognize their surroundings. To this end, different sensors can be installed in different environments. COSPI enables the robots to work with different sensor configuration environments without requiring any robot software reconfiguration. The basic idea of COSPI is the abstraction of recognition types. We investigate 278 recognition routines in behavior modules of a communication robot that works in practical situations and classify them based on communication cues of the communication robot. COSPI facilitates robot behavior development as robot behavior and sensor processing procedures can be independently developed. Furthermore, COSPI encourages the reuse of environmental sensor modules for other robot tasks or applications within the same environment. We have conducted an experiment to confirm that a robot can work in a visual sensor environment, in an infrared ray and pyroelectric sensor environment, and in an optical motion capture system environment. The results showed that robots can work using sensor information from environmental sensors in all these environments.
Keywords: Human-robot interaction, communication robot, sensor network, robot behavior development
