Real time control of the ASBGo walker through a physical human-robot interface

In this work it is presented the development of the conceptual design, implementation and validation of a Smart walker with an inexpensive integrated interface. This interface is based on a joystick and it intends to extract the user's command intentions. Preliminary sets of experiments were performed which showed the sensibility of the joystick to extract navigation commands from the user. These signals presented a higher frequency component that was attenuated by a Benedict-Bordner g-h filter. The resulting interaction signals are then classified and converted into motor commands through a fuzzy logic controller. Additionally, the detection of possible falls and instability of the user was also one of the aims integrated onto the overall system architecture. Results have shown that the resultant movement of the walker was constant and safe without bumps. Short enquiries to the users have provided positive feedback about the device maneuverability in terms of easiness to use.