Objective This project quantifies operationally relevant measures of flight performance and workload in a high-fidelity long-duration spaceflight analog, longitudinally across mission duration, using a portable simulation platform. Background Real-time performance measures allow for the objective assessment of task performance and the timely identification of performance degradations. Methods Measures of flight performance on a piloted lunar lander task were collected on 32 total crewmembers across 8 simulated space missions of 45 days each (623 total sessions). Results Mission duration demonstrated a significant effect on measures of flight performance across all campaigns. Flight measures showed a general pattern of peaking in accuracy during the middle-late quartiles of overall mission time, then degrading again towards baseline. On the workload measure, however, a general linear decrease in workload consistent with progressive task learning was observed in both campaigns. Conclusion This investigation demonstrated the disruptive effect of time in mission on some, but not all, aspects of task performance. While mission interval differentially impacted measures of flight accuracy, workload, by contrast, seemed to steadily decrease with in-mission time. Application While more work is needed, the observed discrepancy between progression of flight performance and workload assessment highlights the importance of sensitive and specific measurement tools for the tracking of distinct performance metrics.

There are many robotic scenarios that require real-time function in large or unconstrained environments, for example, the robotic arm on the International Space Station (ISS). Use of fully-wearable gesture control systems are well-suited to human-robot interaction scenarios where users are mobile and must have hands free. A human study examined operation of a simulated ISS robotic arm using three different gesture input mappings compared to the traditional joystick interface. Two gesture mappings permitted multiple simultaneous inputs (multi-input), while the third was a single-input method. Experimental results support performance advantages of multi-input gesture methods over single input. Differences between the two multi-input methods in task completion and workload display an effect of user-directed attention on interface success. Mappings based on natural human arm movement are promising for gesture interfaces in mobile robotic applications. This study also highlights challenges in gesture mapping, including how users align gestures with their body and environment.