Abstract

In situations where generating in internal representation of an object's shape can not be carried out via visual methods, such as in low light conditions or due to a hardware fault, an internal representation of the object must be built from tactile information alone. Existing methods for doing so rely on the availability of a complete force sensitive skin. The work described in this paper shows a method that requires no such equipment and in fact requires no external sensors at all. The method demonstrates that an accurate representation of the object can be built with the accuracy depending on the number of attempts at contacting the object. This internal representation of object can then be used for other tasks such as planning grasps.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call