This paper presents some results from investigating wheelchair driver learning when using different amounts of sensor support. A quantitative and qualitative empirical evaluation is made about the effect of using intelligent sensor support while a participant learns to operate a wheelchair. Reliance on teaching processes was recorded while a variety of levels of support were provided. Results are presented.

A system is presented that shares control between ultrasonic sensors, a tele-operator and a mobile robot. The mobile robot can be directed by the tele-operator, or by ultrasonic sensors, or they can share control. The mobile robot system can change direction if there are obstacles ahead or if it is helpful. Sharing control allows a human tele-operator to drive efficiently and safely. Controller gains are set automatically for a human tele-operator and the ultrasonic sensor system by calculating a confidence factor for the mobile robot tele-operator. The ultrasonic sensor system can assist a human tele-operator in driving the mobile robot to offset for shortcomings, for example the tele-operator may not be able to see the mobile robot or the human tele-operator may be tired. Finally, some testing is described to validate the proposed methods.