Evaluation of Physiologically-Based Artificial Neural Network Models to Detect Operator Workload in Remotely Piloted Aircraft Operations
Abstract
The current research focuses on preventing performance decrements associated with mental overload during remotely piloted aircraft (RPA) operations. This can be accomplished using physiological signals to sense moments of high cognitive workload and providing augmentation to reduce workload and improve performance. Two RPA operators were interviewed to identify factors that impact workload in RPA, surveillance and target tracking missions. Performance, subjective workload, cortical, cardiac, respiration, voice stress, and ocular data were collected. Several physiological measures were sensitive to changes in workload as evidenced by performance and subjective workload data. In addition, several real-time workload models were evaluated. Potential future applications of this research include closed loop systems that employ advanced augmentation strategies, such as adaptive automation. By identifying physiological measures well suited for monitoring workload in a realistic simulation, this research advances the literature toward real-time workload mitigation in RPA field operations.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 13, 2016
- Accession Number
- AD1021748
Entities
People
- Chelsey Credlebaugh
- Christina Gruenwald
- Jonathan Mead
- Matt Middendorf
- Michael Hoepf
- Paul Middendorf
- Scott Galster
Organizations
- Oak Ridge Institute for Science and Education