System Health Awareness in Total-Ionizing Dose Environments

Understanding the relationship between the impact of radiation at the component and system levels is challenging. This paper discusses a hierarchical approach, based on Bayesian theory, to establish a mechanism for determining system health based on the status of, and interactions between, the radiation response of component parts. When the Bayesian network is trained with a combination of experimental data, data from similar parts, simulations, and expert estimates, a quantitative estimate of the Total-Ionizing Dose (TID) response of a system can be obtained. Bayesian networks enable inference about system-level functional performance, the dose exposure, and the sensitivity of different components to TID, thus providing a framework for TID awareness in design and operation of systems. A case study of a robotic system consisting of commercial components is presented.