Semiautonomous wheelchair navigation using external sensing
Classical mobile robots incorporate a lot of sensors in order to achieve a high level of autonomous navigation. However this classical approach is not usually the best when we pretend to include navigation helps in an electric wheelchair. As an example it is a well-known fact that users usually reject sophisticated and intricate wheelchair with lots of sensors and devices (in particular when they are designed to carry out an automatically navigation).
Besides, new AmI (Ambient Intelligence) paradigm can offer us a useful infrastructure to help navigation. In fact video vigilance systems, domotic controllers, and in general wireless network sensors, may provide information about the structure and systems of a room (maps, lifts, doors, etc.) and even about the objects that are present in an “intelligent” building.
Using this information the aims of this project is to implement a help navigation system for wheelchairs (maybe useful for some others robots), satisfying three conditions: avoiding wheelchair sophistication, extracting environment information from building infrastructure, and being fully useful to final users.
Wheelchair should continuously emits its state information and receive selected or filtered environment information. The navigation system would combine building information with users commands and options in order to help the user to accomplish the different tasks that he/she needs. The help would be more or less accurate according to the provided information by the “ambient intelligence”.
More info: SemiwheelNav
In Press: ABC