Dynamic across time autonomous-sensing, interpretation, model learning and maintenance theory
Description
A formal theory for the development of a generic model of an autonomous sensor is proposed and implemented. An autonomous sensor not only interprets the acquired data in accordance with an embedded expert system knowledge base but also uses it to modify and enhance the knowledge base. The main objective of the model is to combine the capabilities of the physical sensor and an expert operator monitoring the sensor in real-time. This gives the sensor the capabilities of not only sensing the measurand, but also interpreting the sensed data at a higher human-like level, maintaining its databases over time to account for bad or incomplete data, and learning about the measurand and sensor behaviors. The sensor's data base is defined as the quantitative data it senses as well as the qualitative data it interprets. Its knowledge base is defined as the rules that allow maintenance of the truth and integrity of the system and methodologies for model learning. Relevant aspects of the system are described by properties and their qualitative measure called states. A set of properties and their state values define a concept, and a set of consecutively occurring concepts, over time, describe behaviors that essentially provide a qualitative view of the different possible states of the system. These behaviors and associated concepts, called envisionments, are used to identify the measurand and sensor behaviors in real-time so as to take appropriate countermeasures for those behaviors that cause problems. The identification process is called interpretation and is similar to the pattern recognition problem. Dynamic Across Time Autonomous - Sensing, Interpretation, Model Learning And Maintenance Theory (DATA-SIMLAMT) is a novel theory in the field of robotics and artificial intelligence that attempts to model computer reasoning on human-like reasoning about system behaviors. It finds applications in any field that incorporates the human in the control system. Autonomous sensing is just one application of this theory