A cognitive architecture safety design for safety critical systems

Abstract

This research is presented as a safety analysis of a cognitive architecture with an intelligent decision support model (IDSM) that is embedded into an autonomous non-deterministic safety-critical system.Cognitive technology is currently simulated within safety-critical systems in order to highlight variables of interest, interface with intelligent technologies, and provide an environment that improves the system's cognitive performance. In this research, the safety of the architecture was analyzed on an actual safety-critical system, an unmanned surface vehicle (USV). The safety analysis was conducted in both a simulated and a real world nautical based environment. The objective was to define the safety design of a cognitive architecture. The input to the safety design was provided through an approach that identified and mitigated hazards associated with a USV controlled by a cognitive architecture. This analysis provided a structured, task-oriented approach for the dissemination of information concerning safety requirements. This approach was necessary to achieve a safe execution of the USV's capabilities through a design that reduces the potential for injury to personnel and damage to equipment.Other real time applications that would benefit from advancing the safety of cognitive technologies are unmanned platforms, transportation technologies, and service robotics. The results will provide cognitive science researchers with a reference for safety engineering of artificially intelligent safety-critical systems.