Adaptive Automation for Anomaly Resolution

As ground operations personnel and crew manage the operations of orbiting and exploration spacecraft, inevitably anomalies will arise. Resolving anomalies in real-time and preventing their future occurrence can be difficult. Operations personnel must capture the conditions leading up to the anomaly from voluminous telemetry logs; the development of a workaround is a trial and error process; users must also determine if the anomaly affects a piece of equipment or a class of equipment; and the reconfiguration of the equipment under the conditions of the anomaly may preclude certain other operations. Using software from previous research, we propose to design and develop an automation framework that provides 1) the ability to capture the system configuration at the time of the anomaly, using processed telemetry and execution states from both plan and procedure execution; 2) a workaround authoring capability to produce first flight notes and eventually full workaround procedures, derived from the original procedures; 3) the ability to efficiently modify the preconditions and the effects of the workaround as well as the configuration of the underlying models and the affected operational constraints; and 4) an interactive ability to generate operations plans that use the workaround procedures to test the workarounds, new operational constraints and other affected procedures.

Mission planning and execution is at the core of all space missions in order to efficiently employ high cost space assets such as astronauts, equipment, vehicles and communication links. Since anomaly resolution is essential to executing both plans and procedures, our technologies will have applications across many NASA programs, from Mission Control to on-board NASA vehicles and outposts. Because this development will have been done in concert with MCC flight controllers, we believe its use will resonate with program managers in the mission operations directorate. We expect applications of our technology to impact ISS operations by streamlining and reducing errors in anomaly resolution, but also a variety of research programs, such as the new Mission Control Technologies (MCT) software being developed by Alan Crocker of NASA JSC DS. Our work will provide a connection to automated planning technology development ongoing at NASA Ames in support of automation for operations. We will work closely with Dr. Jeremy Frank at NASA ARC during Phase 1 to ensure our relevance to those projects.

The number of robotic entities becoming available for military operations is increasing dramatically and their capabilities are evolving at a rapid pace. Rather than being used as scarce, specialized resources, these entities will be relied upon to play an integral role on the future battlefield. As the military begins to move from teleoperations to semi-autonomous and autonomous operations, it will be plagued with the same anomaly resolution problems as NASA. Our technology will again serve as a unifying framework to streamline and reduce errors in problem resolution related to integrated human and robotic systems. A mirror of NASA space operations are operations in refineries, chemical plants, nuclear and other power plants and any installation that has established standard operating procedures that must be carefully followed under often stressful situations. As these industries move to electronic procedures tied to system telemetry and integrated with planning for more efficient and safe operations, they will require our anomaly resolution framework to streamline and reduce errors in problem resolution.

This is early/mid-stage (TRL 4) — the most realistic path in is NASA SBIR/STTR, which funds small businesses and research institutions to develop technology aligned with NASA's needs (equity-free, phased funding). Check whether a current SBIR/STTR solicitation topic overlaps with this project's technology area, or contact the project directly (above) to ask.

None of these are guaranteed paths for this specific project — TechPort itself doesn't have an "apply" button. Reaching out to the contact(s) above with a specific question is usually the fastest way to find out what's actually open.