Risk modeling of autonomous underwater vehicle operation focusing on the human operator

Abstract

Autonomous Underwater Vehicles (AUVs) are used in scientific, commercial and military organizations to conduct surveys and follow a preprogrammed mission path through the oceans. If a fault or an unexpected deviation from the mission plan should occur, which the autonomous capability of the AUV does not discover as a fault, the operators have to detect these instead and act appropriately. This paper presents a Bayesian Belief Network to assess the probability of monitoring success of an AUV mission. Some of the influences that are incorporated in the model are; Trust in the System, Automation Etiquette, Additional Workload, Time Delay and Loss of Status Messages. It is believed that the model can help improve AUV operations by clarifying relationships between technical, human and organizational factors and their influence on mission risk.