UNILOAD - TWO
Abstract
To build Neuroadaptive Technology - technology that adapts to the current state of user and environment - detecting and interpreting covert aspects of user state (e.g. workload) through Passive Brain-Computer Interfaces has proven to be a powerful approach. A pBCI generates a classifier that can distinguish between different characteristics of user state (e.g. high and low workload) by evaluating the user’s brain activity in its context, in real-time. To make Neuroadaptive Technology applicable in realistic scenarios, the currently time-consuming calibration process is necessary need to be optimized. One approach towards this goal is the implementation of universal classifiers, i.e. classifiers that do not need special calibration when transferred from one user to another, or when the user switches between tasks. The focus of this project is set on EEG-based pBCIs that are calibrated for mental workload detection. Specifically, an approach for generating universal classifiers will be investigated in a series of experiments. The project has two stages: Stage 1 is set for the first year while Stage w adds two more year. In total, three experiments will be completed by 51 subjects starting out in a laboratory context with high-density conventional gel-based EEG recording and concluding in a real-world context with an optimized subset of novel dry EEG system independent of gel-application.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Aug 11, 2021
- Source ID
- FA86552017007
Entities
People
- Thorsten Zander
Organizations
- Air Force Office of Scientific Research
- United States Air Force