ATHENA - Appraisal of Task Health and Effort through Non-Intrusive Assessments

Based on a series of prior and ongoing research projects funded by NASA and the US Air Force (US AF), we have demonstrated viable methods to achieve accurate workload measures (70-100% accuracy) for an Unmanned Aerial Vehicle (UAV) work domain using linguistics and keystroke dynamics. We believe these methods can be reconfigured to use for spaceflight operations. These completely unobtrusive measures of cognitive load can be part of an overall workload assessment assay and are compatible with both team and individually performed tasks. Within Phase I, we will conduct an extensive literature review of the current state of the art in automated detectors, focusing on unobtrusive methods that can detect dimensions proven scientifically to predict workload. This review will reveal evidence-based features that show the most promise as both behavioral indicators of workload and as objective measurements that are amendable to machine learning and statistical techniques. The results of this review will serve as a risk reduction exercise to evaluate and down-select a set of workload detection methods that warrant further investigation. We will then use the set of candidate metrics, along with our own prior work with semantic analysis and keystroke dynamics to design machine learning algorithms, ultimately producing a workload sensor tailored for use with long duration mission relevant tasks. Phase I will include an exploratory investigation of a candidate measures, initial sensor designs, and produce experimental plans and IRB protocols for the overall system validation using a combination of data gathered from laboratory settings and ground-based analogs such as the Human Exploration Research Analog (HERA) at JSC, the UTMB Bedrest facility, and the Hawaii Space Exploration Analog Simulation (HISEAS). Phase II will focus on engineering efforts to implement the sensor algorithms, the support tools for the experiment, conducting the validation study, and data analysis.

ATHENA is a software-based completely unobtrusive cognitive load sensor. It will provide astronauts, mission planners, and system designers with a rapid, automatic, and systematic way to assess the demands of tasks, allowing them to schedule modifications or adapt interfaces in order to optimize the use of the crew's cognitive resources and reduce the risks associated with prolonged over-work or under-work conditions that can threaten mission success and crew performance and behavioral health. This directly addresses the HRP's identified risk of "Performance Errors Due to Fatigue Resulting from Sleep Loss, Circadian Desynchronization, Extended Wakefulness, and Work Overload". ATHENA will produce empirical evidence and the underlying technologies to help close the knowledge gaps "Sleep1: ...identify a set of validated and minimally obtrusive tools to monitor and measure sleep-wake activity and associated performance changes for spaceflight", and "Sleep2: We need to understand the contribution of [...] work overload, on individual and team behavioral health and performance, including operational performance, for spaceflight." The successful deployment of the ATHENA sensor can be used with NASA's current mission planning tool, playbook, for missions onboard ISS or future long duration missions. ATHENA can also be used by NASA's Flight Analogs Project team, where its evidence based workload measures can be used to facilitate the manipulation of experimental conditions.

While some aspects of the spaceflight work environment are extremely unique, the need for reliable and unobtrusive workload detection exists in a number of domains that contain high tempo high criticality tasks. For example, the US Air Force Research Laboratory is interested in sensors that do not require wearable devices. Methods for near real-time assessments of cognitive load of UAV operators can be used in concert with automation aides to inform and execute the re-routing of tasks. If successful, ATHENA represents the first step in the process of automatically optimizing human cognitive capacity, enhancing performance in the human-automation ecology. We believe a validated cognitive load sensor will be useful in any domain where humans often switch between tasks that require different levels of cognitive processing. The ability to accurately describe and characterize cognitive load requirements will impact human-automation interaction designs in wide ranging applications from military to industrial uses. Where possible, we will leverage existing relationships with collaborators to seek opportunities for ATHENA use such oil refinery and power plant control and management.

This is early/mid-stage (TRL 4) — the most realistic path in is NASA SBIR/STTR, which funds small businesses and research institutions to develop technology aligned with NASA's needs (equity-free, phased funding). Check whether a current SBIR/STTR solicitation topic overlaps with this project's technology area, or contact the project directly (above) to ask.

None of these are guaranteed paths for this specific project — TechPort itself doesn't have an "apply" button. Reaching out to the contact(s) above with a specific question is usually the fastest way to find out what's actually open.