An experimental study on usability of brain-computer interaction technology in human spaceflight

Abstract

Over the past few decades, the extensive development of human-computer interaction has greatly improved the life of human beings. As a novel form of human-computer interaction (HCI), brain-computer interface (BCI) has shown its potential application values in some special areas such as mental typing, rehabilitation engineering, etc. It is known that the astronauts in space, especially in long duration spaceflight, are always occupied to deal with many complicated tasks. More effective HCI devices are required to aid astronauts to fulfill their tasks with lower mental workload. The rising of new HCI technologies including BCI may provide promising solutions for this problem. We took the advantages of China Tiangong-2 Space Lab in Nov. 2016 to carry out on-orbit experiments to examine the usability of BCI in space and factors influencing BCI performance. The experiment design adopted three typical paradigms of BCI including event-related potential (ERP) based BCI, motor imagery (MI) based BCI, and steady-state visual evoked potential (SSVEP) based BCI. Besides, three different experiment environments: normal experiment, simulated on-orbit experiment and real on-orbit experiment were conducted to compare the changing of physiological responses and BCI performance between ground and space. Thirty-five health participants took part in the normal experiment and simulated on-orbit experiment. Furthermore, the real on-orbit experiment was carried out in Tiangong-2 space lab in November 2016, and two crewmembers completed the tests as scheduled. The experimental results indicated that machine noise had significant effect on performance of P300 based BCI and MI based BCI between normal experiment and simulated on-orbit experiment. Additionally, negative emotion had significant effect on the performance of MI based BCI. Besides, the difference between average accuracy of normal experiment and on-orbit experiment was not significant. From the aspect of brain response, few differences were observed over the three BCI paradigms in the three experimental conditions. The results suggest that BCI technology is a very competitive ways of HCI which could be used in future space missions, however, further improvements are needed in both BCI hardware and adaptive algorithms for better performance.