Description: Autonomous, avionic and robotic systems are used in a variety of applications including launch vehicles, robotic precursor platforms, etc. Most avionic innovations are based on software-embedded systems, and this has resulted in an increase in the number of interactions (coupling) among heterogeneous subsystems. Avionic systems degrade in performance due to gradual development of anomalies and unanticipated failures ranging from issues affecting a single hardware or software subsystem to issues occurring as a result of coupling among multiple subsystems. In addition, system usage and operating conditions may lead to different failure modes necessating multiple recovery procedures possibly causing conflicts and deadlocks among recovery steps. QSI intends to address these challenges by leveraging the current capabilities of model-based fault management and supportability solutions of TEAMS to efficiently sequence individual steps within each procedure, including adding/deleting steps, and resolve conflicts and deadlocks in recovery procedures. TEAMS-RT, the real-time inference engine, has multiple fault diagnosis capability built-in. Additionally, TEAMS-RDS (TEAMS-remote diagnostic server) already exploits commonalities among test steps during guided troubleshooting, where each test is represented as a chain of pre-setup, post-setup and action nodes with Do and Undo steps interspersed. The proposed effort will extend this to more general digraphs of test and recovery/repair procedures and also embed this capability in a solution linked to enhanced TEAMS-RT for automated /crew-initiated recovery and resolution of conflicts and deadlocks in recovery procedures. This proposal aims to enhance QSI?s existing probabilistic inference engine to handle multiple, intermittent and coupled failure scenarios and developing an ISHM response engine module that dynamically assembles feasible and near-optimal recovery procedures to handle multiple failure scenarios.

Benefits: NASA is developing increasingly autonomous systems that can perform missions with a high degree of certainty with minimal human intervention. Examples of such mission include rovers operating in Mars, where the missions are extremely long, and therefore multiple components and subsystems will degrade and fail over the duration of the mission. However, due to the long communication delays between Mars and Earth, these systems cannot be monitored and diagnosed by mission control like any other near-earth mission. The proposed capability will be invaluable to NASA for such operations by (a) Predicting failures before they disrupt the mission, (b) Reducing false positives of such prediction with the proposed inference engine under multiple failure scenarios, and (c) merging multiple recovery procedures that may have conflict. This will enable NASA to fill the gap between the current flight avionics CW FDIR systems and desired ISHM systems handling multiple failure high stress situations.

Among the other agencies, DoD, US Air Force, US Navy, and commercial aviation (e.g., SpaceX) are the most potential customer for the resulting technologies. Large scale military systems (systems of systems) such as NORAD, Space Command ground segments, the Joint Strike Fighter fleet, the Navy shipboard platforms, Submarine Commands and ballistic missile defense (BMD) systems can be potential areas to field the proposed technology. In addition, UAVs, UMGs and other unmanned submersible vehicle markets could also be potential target for the proposed technology. The product is also expected to be of commercial value to the manufacturers of DoD and military?s remotely guided weapons and reconnaissance systems.