Multiplatform phased mission reliability modelling for mission planning

Abstract

Autonomous systems are being increasingly used in many areas. A significant example is unmanned aerial vehicles (UAVs), regularly being called upon to perform tasks in the military theatre. Autonomous systems can work alone or be called upon to work collaboratively towards common mission objectives. In this case it will be necessary to ensure that the decisions enable the progression of the platform objectives and also the overall mission objectives.The motivation behind the work presented in this paper is the need to be able to predict the failure probability of missions performed by a number of autonomous systems working together. Such mission prognoses can assist the mission planning process in autonomous systems when conditions change, with reconfiguration taking place if the probability of mission failure becomes unacceptably high.In a multiplatform phased mission a number of platforms perform their own phased mission that contributes to an overall mission objective. Presented in this paper is a methodology for calculating the phase failure probabilities of a multiplatform phased mission. These probabilities are then used to find the total mission failure probability. Prior to the mission the failure probabilities are used to decide if the original mission structure is acceptable. Once underway, failure probabilities, updated as circumstances change, are used to decide whether a mission should continue. Circumstances can change owing to failures on a platform, changing environmental conditions (weather), or the occurrence of unforeseen external events (emerging threats). This diagnostics information should be used to ensure that the updated failure probabilities calculated take into account the most up-to-date system information possible. Since the speed of decision making and the accuracy of the information used are essential, binary decision diagrams (BDDs) are utilized to form the basis of a fast, accurate quantification process.

System design and maintenance modelling for safety in extended life operation

“…The system lifetime can be considered as a series of discrete time phases where different maintenance strategies are applied. This concept is similar to the use of phased mission analysis where the functional requirements change as a system mission progresses [8][9][10] Fault tree and event tree methods are not capable of modelling non-constant failure rates that increases over time due to wear out, dependencies between component states or complex maintenance processes. Alternative methods have features to overcome these limitations.…”

A reliability-based approach to mission planning in multi-platform phased missions

“…It is possible to calculate failure probabilities for individual platform phases and the entire missions of these platforms [2] as well as for the phases of the multi-platform phased missions and the overall mission [3]. Methods have been presented which allow the calculation of updated failure probabilities for platforms and overall mission given various pieces of diagnostic information.…”

“…Methods have been presented which allow the calculation of updated failure probabilities for platforms and overall mission given various pieces of diagnostic information. It is possible to take account of component failures, both single and multiple failure modes [2], events external to the systems in question, such as weather or other emerging threats [2], system subsystem or functional failures for which precisely which components have failed is unknown [2] and also the exact time that the diagnosis is provided and a mission prognosis is required [3].…”

Proxel-Based Performability Analysis of Non-Markovian Phased-Mission Systems

“…This includes failure rates of components, system configuration, as well as success criteria. In particular, autonomous systems are typically described and analyzed as phased-mission systems [4,5], where each phase can be expressed as, e.g., a fault tree.…”