Spatial ontologies and models for safety-critical cyber-physical systems

Our research is concerned with the modeling and design of safety-critical cyber-physical systems (CPS), where decision making procedures are required to take the right action at the right time, and in the right place. This paper examines the role that spatial ontologies and models can play in formally representing and reasoning with spatial domain entities occurring in CPS. We develop and introduce a new multi-dimensional spatial modeling hierarchy and reasoning framework supported by region connected calculus (RCC-8). The proposed approach is demonstrated on a problem that considers collision of two vehicles at a traffic intersection. We show that the dimensional fidelity of spatial entities (e.g., one-dimensional representations versus two-dimensional representations) strongly affects the accuracy of decision making outcomes.