An identification technique for adaptive shared control in human-machine systems

Abstract

developed to meet the needs of mulllpiy handicapped people who are unable to operate available wheelchair systems. The NavChair shares conuol with users by allowing them lo achieve desirable motion while overriding unsiit'e mineuvers. An outgrowth of this project is an investigation into methods of adaptive shared control. In particular. rnelhods of auionomous mode selection based upon observations of user behavior are being studied. This paper presents a modeling approach to monitoring human control behavior in real time; preliminary experimental results evaluate its ability to distinguish human control behaviors in a simulated system. The NavChair assistive navigation system [ I, 21 is being