Agency: Department of Defense | Branch: Air Force | Program: SBIR | Phase: Phase I | Award Amount: 150.00K | Year: 2014
A formal approach to the management of information involved in human-automation interactions is proposed. The mathematical formalism is enabled by employing the notion of hybrid systems. The information management involves ensuring correct information flow and shared cognition between human operators and machines. It is envisioned that the proposed algorithms will formalize the analysis of human-automation interactions and hence reduce the number of iterations that are common in the design and validation of automation solely based on human cognition perspective.
Agency: National Aeronautics and Space Administration | Branch: | Program: SBIR | Phase: Phase II | Award Amount: 750.00K | Year: 2014
Based upon the feasibility demonstrated in the Phase I research, Optimal Synthesis Inc.(OSI) proposes to develop a software tool that can be used validate aircraft flight deck user interfaces over the entire flight envelope. The approach is based on a mathematical formalism derived from hybrid systems theory. The correctness of information content in user interfaces is analyzed by a special observability test that takes into account of the limitations in human cognition and psychology. A possible mismatch between an operational mode perceived by the human operator and the one active in the aircraft is detected using an algorithm that compares the inferred intent of the human operator to that of the machine. Metrics-based performance evaluation will be carried out to demonstrate the benefits of the prototype software developed under the Phase II research. The feasibility of employing the software on the flight deck as a real-time pilot aid will also be analyzed in the Phase II research.
Agency: National Aeronautics and Space Administration | Branch: | Program: SBIR | Phase: Phase I | Award Amount: 125.00K | Year: 2014
A significant portion of the NextGen research is aimed at (i) developing ground-side automation systems to assist controllers in strategic planning operations such as scheduling flights, and (ii) developing tactical controller decision support tools to separate and space the traffic. Central to the success of these automation systems is the ability to predict the future trajectory of any aircraft in the National Airspace System (NAS). The research related to this area is referred to as Trajectory Prediction (TP) and sometimes Trajectory Synthesis. Notwithstanding past research, TP remains a very challenging exercise and the quest for improved TP accuracy continues. Any improvements in TP can benefit a wide array of NextGen concepts pursued by NASA. The objective of the current research is to seek a novel approach to TP specifically aimed at addressing some of the deficiencies of the past TP research. The approach involves: (i) machine learning algorithms, and (ii) big data computational platforms. Phase I research will demonstrate the benefits of supervised and unsupervised machine learning algorithms for TP. Phase II research seeks to develop real-time trajectory prediction algorithms that can be used for a wide variety of NASA NextGen concepts.
Agency: Department of Defense | Branch: Missile Defense Agency | Program: SBIR | Phase: Phase I | Award Amount: 100.00K | Year: 2015
The sensor models to be employed in the proposed research will improve tracking performance. An implementation of the technology on High-Performance Computers is proposed in order to address the computational complexity. Phase I research will demonstrate benchmark simulations on a GPU and MIC-equipped computer. (Approved for Public Release 15-MDA-8482 (17 November 15))
Agency: Department of Defense | Branch: Navy | Program: SBIR | Phase: Phase I | Award Amount: 79.95K | Year: 2015
Military operations and logistics are inextricably linked. The importance of In-Transit Visibility (ITV) capability for reliable tracking and delivery of logistical items has well been recognized across Department of Defense (DoD) logistics programs. The current state-of-the-art solution to ITV employs a Radio-Frequency Identification (RFID) network, which comprises a large number of read-and-write stations and hence requires high cost and intensive manpower. Motivated by past experiences on air traffic management and machine learning, Optimal Synthesis Inc. (OSI) proposes an alternative approach that exploits existing variety of data sources, a good example of which is the IDE/GTN Convergence (IGC) that is a unified data service across DoD cargo movement and tracking. The approach combines machine learning techniques with the estimation and prediction methodologies for tracking the cargo movement and predicting the time-of-arrival in each mission leg. By performing on-line risk monitoring, a decision support tool that generates automated alerts for human intervention is also developed using a stochastic decision framework. The proof-of-concept demonstration is planned in Phase I, and the software prototype is planned to be developed in Phase II for functional demonstration.