Assessment of Astronauts’ Workload with Task-Irrelevant Auditory Probes In Manually Controlled Spacecraft Rendezvous and Docking

Abstract

Introduction. The rendezvous and docking (RVD) operation is a challenging task of high risk for astronauts. It is an essential part of missions involving more than one spacecraft such as transportation missions for the International Space Station supply. While intensive automation is an indispensable ingredient of the modern RVD operations, manual docking remains one of the critical skills for an astronaut. Manual RVD is a cognitively demanding task that requires a constant sustained attention. The astronaut has to perform a precise docking by minimizing the linear and rotational speeds of the spacecraft using minimum control actions, thus, minimizing the fuel consumption. The performance in RVD operation can be modulated by the astronauts’ workload. Efficient assessment techniques can be used on the ground both for human factor evaluation of the system design and training purposes, as well as in space for adaptive astronaut-spacecraft interaction. In this study, we evaluate whether task-irrelevant auditory probes can be an effective assessment method of astronauts’ workload. Materials and Methods. The docking operation was simulated using the simulator of the Moscow Aviation Institute, which uses two stereographic projectors and 3D-glasses reproduces the immersive experience of the SOYOUZ module docking to the International Space Station (ISS). An astronaut can operate two levers: left lever controls the translational speed (Vx, Vy, Vz), while the right one controls the rotational speed (Wx, Wy, Wz) of the spacecraft. The numerical values of both speeds are displayed on the screen. The levers can be operated with or without feedback. With the feedback, when a lever is released, the corresponding speed is automatically reduced to 0. Visual guiding cues can also be displayed on the screen: a dotted circle indicating the direction of the speed vector, a cross corresponding to the projection of the docking in the current configuration, and a corridor represented with four trajectory lines perpendicular to the final contact plan of the docking. Three scenarios were used to modulate the participants’ workload. In the first scenario, both levers feedback and visual cues were present. In the second scenario, only feedback was present. In the third one, only visual cues were present. Three participants with docking simulator experience performed three times each docking scenario. The workload was measured using task-irrelevant auditory probes. The participants were presented with 200 ms high (2 kHz) and low (1 kHz) tones in 20/80% proportion with a variable inter-stimuli interval from 2 to 3 seconds. They were asked to verbally report the rare high tones and their response times were measured. The participants evaluated each scenario using NASA-TLX questionnaire dimensions. Results. The participants performed fewer input actions for the scenario n°2 with visual cues only. While most efficient for fuel consumption, the second scenario induced the least accurate final docking positions, and was considered by participants as the most demanding both mentally and physically. The response times were consistent with the subjective and objective difficulty of the scenarios: 475.3 ms on average for the scenario n°2 vs 412.4 ms for the scenario n°1 and 387.6 ms for the scenario n°3. Conclusion. The presented work showed that a simple technique of task-irrelevant auditory probes can be used to assess the astronauts’ workload during manual controlled spacecraft rendezvous and docking. The participants of this experiment in SOYOUZ module simulator showed less accurate docking in a scenario without lever feedback. They also evaluated this scenario as particularly cognitively and physically demanding. The response times to irrelevant high-pitch tones were in line with the objective (performance) and subjective (questionnaire) measurement of the scenarios’ difficulty. Therefore, this simple technique can be used in such situations as manual RVD in order to evaluate the astronauts' workload.