Robot service framework based on big data technology

Autonomous robots play important roles in providing high quality services in pervasive computing environments in which huge amounts of data are invoked, measured, and analyzed. To enable accurate robot decisions, big data-based technologies (e.g., location-based service technologies to obtain the positions of objects and humans from measured data) are required to provide diverse types of services, and the approaches to generate and execute motor primitives for autonomous robots based on drawn information are also required. One avenue of related research involves the generation of motor primitives through demonstration-based learning, in which a human operator teaches motor primitives to autonomous robots through direct control. However, the problem of inaccurate comparison of motor primitives remains a weakness. In this paper, we propose a motor primitive generation method for autonomous robots in pervasive computing environments using Levenshtein-distance algorithm. As we were able to validate the usefulness of the algorithm in reducing the number of similar motor primitives, we can conclude that the use of the proposed method to improve the generation of motor primitives can enhance autonomous robot performance.