SBIR-STTR Award

Our innovation will detect, in near real-time, NAS operational anomalies by uniquely combing with analytical methods our existing Microsoft Azure based TFMData flight information warehouse, live Air Traffic Control (ATC)-Pilot voice communication records, and IBM Watson capabilities such as natural language processing.Implementation of our proposed capability will fill one of the gaps for monitoring and predictive safety tools in the terminal area. In the enroute domain, predictive metrics such as the Monitor Alert Parameter (MAP) and "going red" forecasts help traffic flow managers balance traffic and workloads, thereby increasing safety. However, this relies on the assumption that ATC-pilot communication is of superior quality, unambiguous, and strictly procedural. Also, pilots reacting to controller resolutions by changing the trajectory of the aircraft (either using lateral or vertical maneuvers) may react late, react wrongly, or not react at all. We aim to find these anomalies by correlating actual flight trajectory data and ATC voice communication data. While these anomalies could be precursors to unsafe events, we view them as indicators of inefficiencies in flight operations. Identifying these inefficiencies through innovative data mining methods can uncover unique and recurring problems that otherwise go undetected. Our concept will also provide better insight into the frequency and content of controller instructions and interventions.

Potential NASA Commercial Applications:
(Limit 1500 characters, approximately 150 words) RSSA: We offer an innovative approach to detecting anomalies that will benefit the RSSA milestone for deploying real-time safety monitoring tools.TBO: Our concept offers a method for identifying recurring operational inefficiencies that reduce capacity and increase flight time and costs. Our method complements traditional airspace analyses by providing for real-time monitoring that compares what should happen to what does happen.

Potential NON-NASA Commercial Applications:
(Limit 1500 characters, approximately 150 words) Operators and controllers report on recurring congestion in subsectors that cause inefficient deviations from planned routes. These inefficiencies do not typically get reported to the ATSCC and get de-conflicted in the planning process. Both the FAA and airlines would benefit from our system.

Technology Taxonomy Mapping:
(NASA's technology taxonomy has been developed by the SBIR-STTR program to disseminate awareness of proposed and awarded R/R&D in the agency. It is a listing of over 100 technologies, sorted into broad categories, of interest to NASA.) Air Transportation & Safety Data Processing

Robust Analytics team will add real-time speech-to-text (STT) monitoring of controller-pilot radio communications to the risk state analytical framework previously demonstrated by Robust Analytics. The combination of our successful Phase I application of STT to ATC communications with the risk state analysis data fusion and analytical framework will provide a real-time risk and safety margin monitoring capability for the NAS.Adding voice communications monitoring remains crucial for achieving NASA’s system-wide safety monitoring goals. Voice is still the main source of ATC-pilot communications and a large body of airspace situational awareness and operationally-related information is contained therein that never makes it effectively back into broader National Airspace Systems (NAS) use. Our innovation provides access to that information, in real-time, and with the built-in analytics to use that information to identify anomalies and provide alerts. When combined with the other factors included in our risk state assessment framework, our innovation would have identified recent crashes such as Asiana 214 and UPS 1354 as high risk flights, before the events occurred.Robust Analytics offers a vision for NAS-wide, real-time safety monitoring based on analyzing controller-pilot voice communications for anomalies and clearance deviations, and combining that insight with information from other data sources (flight plans, position reports, weather, infrastructure status, and traffic density) through advanced analytics with cloud computing for a scalable, reliable 24/7 solution.

Potential NASA Commercial Applications:
(Limit 1500 characters, approximately 150 words) Our ILSA innovation offers a direct contribution to meeting NASA's strategic objective for real-time safety assurance. We provide a robust, extensible approach to achieve the technical challenge for terminal area safety margin monitoring. With speech-to-text capability, we add another tool to identify operational anomalies in real-time, and a continuous monitoring system that generates a new source of data for NASA safety analyses. Our risk state analytical framework offers the SWS Project a platform for testing and deploying new anomaly detection algorithms that will be developed with the new voice transcription database we will generate.



Potential NON-NASA Commercial Applications:
:
(Limit 1500 characters, approximately 150 words) In the near term, we envision our innovation used by FAA facility shift supervisors to provide insight into potential risks. After additional validation, our innovation can provide alerts to controllers for flights deviating from clearance instructions.Preliminary discussions with airline decision support tool providers suggest interest in a version of ILSA that would operate in the AOC to provide real-time and prognostic support to dispatchers. The concept is to provide dispatchers with current and predicted information on airspace and aircraft risk status. This information, combined with data on crew and aircraft assignments from AOC information systems, can alert the dispatcher that, for example, a crew operating at the end of a long flight may be entering a region of medium or high risk. The value of this combined information is indicated by the two most aircraft rashes that occurred when the crews were flying non-normal times (0600 hours or at the end of a long trans-Pacific flight) with the ILS out, and in the UPS case with degraded weather. Operators and controllers report on recurring congestion in subsectors that cause inefficient deviations from planned routes. These inefficiencies do not typically get reported to the ATSCC and thus do not get de-conflicted in the planning process. Both the FAA and airlines would benefit from our system.

Technology Taxonomy Mapping:
(NASA's technology taxonomy has been developed by the SBIR-STTR program to disseminate awareness of proposed and awarded R/R&D in the agency. It is a listing of over 100 technologies, sorted into broad categories, of interest to NASA.) Air Transportation & Safety Data Processing Simulation & Modeling