Below are some of the innovative enabling technologies that were used to power some of our product solutions.
Evidential Reasoning Network® – ERN®
Collaborative Reasoning with multiple data-support to back up those opinions
- Bring relevant information and opinion into one holistic data set using a semantic algebra
- Human and non-Human opinions can be aggregated and evaluated producing a fused opinion accounting for uncertainty
- Trust and uncertainty are retained and used to produce an effective result
- Algorithms can further search for relevant information to support or dissuade opinions
- All the information is combined and provided to give the best results based on opinion and fact. We made it scalable, reliable, and easy to extend. Now you have ERN.
PinPST – Paths in Perforated Space Time
PinPST solves path finding problems by expressing the search space as a space-time volume
- Helps keep self-controlled (or man-controlled) moving objects from colliding
- Quickly updates trajectory possibilities based on time, space, and tracked object constraints
- Developed for submarine route planning to perform collision avoidance in a 3D environment
- Used to perform automatic optimal route planning for all maritime objects in the area of interest
Adaptive Resonance Theory Engine (ART-E)
Performing reasoning like a human would perform it…
- Developed in conjunction with world-class researchers at Missouri University of Science and Technology
- Neural network models to perform learning, pattern recognition and prediction (based on Carpenter and Grossberg’s Adaptive Resonance Theory)
- Clusters data according to user preference and similar data patterns
- Used in several R&D projects to “fill-in-the-blanks” where data was sparse or disparate
FUsion Through Uncertainty REduction (FUTURE)
All data sources are not created equal
- FUTURE handles complex target resolution by working from the known to the unknown (like a Sudoku puzzle)
- FUTURE recognizes and uses to its advantage the fact that not all sensor detection measurements are created equal
- Mathematics of FUTURE uses an ambiguity measure that helps prioritize results putting the “easiest to explain” result at the top
- FUTURE is a credible and superior technique when compared to contemporary statistical search/optimization methods
21CSi 