Embedded Brain Reading in a Complex Virtual Environment

Abstract :

High mental workload can overwhelm a human operator. This can in certain setups (e.g. pilots, robot control) result in potentially dangerous situations. Overly complex and unintelligent human machine interfaces (HMIs) can have a negative effect, e.g., loss of situation awareness, in critical situations. Here, we propose the usage of adaptive online single-trial analysis of the electroencephalogram (EEG) to estimate the task engagement of the operator and adjust the HMI of a virtual robot control environment accordingly. By doing this, we will investigate if the amount of errors that are caused by an excessive demand of the human operator can be decreased or the effectiveness can be improved by reducing the amount of time that is required to finish a mission in the scenario.