A System of Systems Approach to Optimize a Realtime Risk Situational Awareness System

In 2017, about 37,151 fatalities resulted from motor vehicle traffic crashes. Crashes cost the U.S. over $800 billion annually in lives lost or injured, lost productivity, and property damage. Many interventions have been adopted to reduce fatalities and serious injuries. A real-time crash intervention can estimate the chance of crash occurrence and analyze risk factors of live video streams captured by the onboard camera of a vehicle, so as to notify the driver to take the appropriate response. This application paper is aimed to improve the prediction to achieve an optimal system by integrating existing risk factors, the algorithms to identify and analyze risk factors result in visualization, etc. Existing systems are integrated into a System of Systems (SoS), the overall objective of which is to maximize the Key Performance Attributes (KPA): Performance of the SoS predicted Time, Performance of the SoS predicted Decision, Affordability, Scalability and Adaptability. The meta-architecture is structured as a chromosome assessed and selected with the non gradient optimization approach based on the simple genetic algorithm integrated with a Fuzzy Inference System.