Reasoning Cognition Of Human Behavior Based On Ontology Technology In Home Intelligent Space

Human behavior recognition and understanding is the key to realize the active personalized service of service robots.At this stage,there has been a lot of research work on the analysis and recognition of human behavior,but mainly for simple human behavior in fixed scenes.There are many kinds of objects in the home environment,the distribution of object is irregular and the obstacles change dynamically,which makes the understanding of the behavioral environment difficult.For understanding of the complex human behavior,the existing methods can not meet the requirements of accuracy and robustness.At the same time,when obtaining relevant behavioral information and environmental information,it is difficult for the service robot to make a correct identification due to the lack of representation and utilization of the knowledge information.To solve the above problems,this paper takes the home intelligence space as the research platform,using the ontology technology to fully express the human behavior.By combining the constructed knowledge base and behavioral inference rules,this paper studies the method based on ontology technology of reasoning cognition of the human behavior in the home intelligent space.The main work is as follows.(1)An environmental knowledge base model under the home intelligence space is constructed to realize the acquisition and utilization of user behavior environment knowledge.The behavioral environment is modeled by the ontology description language OWL.It represents the relevant information of the user domain and the environment domain.By comparing the advantages and disadvantages of memory storage,text storage and relational database storage,we choose to use relational database as the storage tool of ontology,which realizes the management and utilization of knowledge in ontology.It lays the foundation for reasoning cognition of human behavior based on ontology technology in home intelligent space.(2)A method of behavior time relation representation based on ontology technology is proposed.It uses ontology technology and interval algebra theory to realize the representation of complex behavior in the home environment.In order to study the simple action time relationship and complex behavior time relationship,the interval algebra theory is introduced into the representation of human behavior based on ontology technology.The OWL behavior time relationship ontology model is constructed by Protege,which makes up for the lack of time relationship representation of between human behavior.(3)A method of human behavior reasoning based on combined reasoning mechanism is proposed,which fully considers the problem of the reasoning of behavior rules at different behavior levels and realizes the mapping from low-level action elements to advanced behavior.Based on the research of semantic rules,the method based on the decision tree model is used to construct the basic action rules of human body.The reasoning cognition of human behavior is realized by combining the behavior rules based on decision trees with the semantic rules based on SWRL.(4)An overall framework of reasoning cognition realization of human behavior is proposed.It combines the environmental knowledge base model with the ontology model of human behavior time relationship in home intelligence space.The combines reasoning mechanism is introduced into the behavioral reasoning module.The corresponding behavior semantic rules are written to realize the reasoning from basic actions,simple actions to complex behavior.The paper validates the reasoning of human behavior in home environment and the experimental results show that the framework of reasoning cognition of human behavior based on ontology has certain tationality and timeliness.