This article presents results from a study of the parameters of eye movement in human subjects using an oculographic interface to control a self-propelled chassis, i.e., an electric wheelchair. This work showed that from run to run, the time spent by the subjects on each run decreased and the number of errors and the number of subjects who did not successfully complete their runs decreased. Cluster analysis identified individual typological features of eye movement responses to the movement of the user’s head during execution of the main commands controlling the self-propelled chassis. When the self-propelled chassis is moving, generation of the “forward” and “backward” commands by the opto-oculographic interface is generally carried out in a manner common to users and does not require correction when controlling the self-propelled chassis. Generation of turn commands is a more individual process, associated both with calibration of the device and the individual user’s head and eye positions and therefore requires more careful monitoring when using oculographic interfaces.