Context Modeling And Reasoning For Home Service Robot In Intelligent Environment

Along with social progress and technological development, people are no longer satisfied with teaching and performance of the robot. How to make robots more autonomous and intelligent to provide daily service becomes a research hotspot. Intelligent space technology provides new ideas to solve the limitations of traditional robot. However, service robots intelligent space is a heterogeneous, dynamic and complex computing environment, involving types of information. It becomes a very important research topic that how to effectively organize various information, and achieve the interconnection of information, in order to provide a favorable support for environmental sensing, service decision making of service robots.This thesis reviews the status of context-aware system. The context-aware system for home service robot under intelligent environment is designed, OWL based context modeling and Jena based context reasoning is implemented, and it is successfully applied to the intelligent home space structured by Service Robotics Lab of Shandong University. The main contributions of this thesis are shown as follows:Firstly, this thesis analyzes the functional requirements of intelligent space for home service robot and the context-aware system, discusses the implementation problems, and establishes the system architecture which consists of sensors/execution network, access and control to devices layer, context processing engine, mechanism of context storage and interface of application.Secondly, context related in the intelligent space is classified and extended using top-down approach. Intelligent space ontology is constructed by Protege, concepts and properties about information in the workspace are organized in hierarchical structure and described by OWL. Relationships between different information are illustrated by properties, and specific information is represented by instance.Thirdly, the interpretation from low-level context to high-level context is realized via the rule-based inference engine provided by the Jena reasoning subsystem. The reasoning rules consist of two parts:axioms and implicit rules provided by intelligent space ontology, inference rules developed by the user. According to service type, the user rules is built using concepts and properties defined in intelligent space ontology, then high level context is generated by fusing single context atoms in the process of context-aware.Finally, the application of context-aware based ontology and rules-based reasoning in home environment for service robot is described. Intelligent space context reasoning and monitoring system is developed by Java, which responsible for interpreting context and monitoring users, the robot and devices. The results show that the methods of context modeling and reasoning effectively describe the environment information, which realizes the information sharing, so as to providing decision support for service robot.With the development of related technologies, the function of context-aware system based on ontology and rule-based reasoning will be more perfect, and it will improve the service quality of service robot intelligent space. This paper studies the context-aware technology for service robot in home intelligent space, and it will be valuable to context modeling and reasoning in other service environment.