Final Scientific/Technical Report - Sensor Indoor Location Network for Smart Airport Terminal Management
- Skymantics, LLC
The purpose of the research performed in the Phase I project is to develop and validate a Proof of Concept for a distributed smart sensor platform (called PAXSense) for tracking sensor deployments in airport terminal spaces. This platform provides capabilities of smart data processing deriving information categorization and prediction of future events in the terminal, without hardware modifications in the sensors. In addition, the platform is intended to be fully distributed and provide direct connectivity between sensors in a mesh network that can be infinitely scalable. The research has been carried out following two parallel lines of work to address separate problems: 1. Quality of the processed data by sensors to allow accurate tracking end-to-end in an indoor facility of 100% of the passengers, categorization according to movement patterns, and correlation with flight information collected from timestamped boarding pass scans to provide predictability calculations. 2. Capability of Artificial Intelligence algorithms running in low-power microprocessors behind sensors, or sets of sensors, covering subspaces (rooms) of the terminal to derive stable passenger patterns in terms of movement, relation to their flight and use of concessions, and prediction of time arrivals at critical milestones of the transit process. For each line of work, a literature or technology state-of-the art has been performed, together with a definition of the operational environment and scenarios, and associated requirements. This is followed by a design and implementation of testbeds using real data (for problem dimension 1) or large volumes of simulated data (for problem dimension 2). Tests have been executed to collect, store, process and display information and validate the results against the estimated design requirements derived. The results have been overwhelmingly positive. It has been found that a correct combination of sensor technologies deployed in the indoor facilities can yield sub-meter and sub-second accuracy levels which are enough for the PAXSense application. In addition, AI algorithms have proven to be highly accurate (figures of 80-95% estimation accuracy have been achieved) and can run on low-power microprocessors interfacing groups of sensors and distributed in the building facility. These algorithms provide passenger pattern categorization, in-room estimation of transit times and exit door, and collaborative calculation of total transit times using real-time data from neighboring sensors. Applications of the research results include distributed High-Performance Computing, AI-powered Internet of Things (IoT) for predictable indoor status monitoring, transportation operations research and a potential of integration into Blockchain, indoor mapping and navigation for first responders and National Security and building energy efficiency by Demand-Response (DR) electricity consumption monitoring.
- Research Organization:
- Skymantics, LLC
- Sponsoring Organization:
- USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR)
- DOE Contract Number:
- SC0018444
- OSTI ID:
- 1495687
- Type / Phase:
- SBIR (Phase I)
- Report Number(s):
- DE-SC0018444-1
- Country of Publication:
- United States
- Language:
- English
Similar Records
Development of Smart Ventilation Control Algorithms for Humidity Control in High-Performance Homes in Humid U.S. Climates
Coastal Ohio Wind Project