Towards vision guided manipulation

摘要

One of the major problems in the realisation of a really autonomous robotic system, has certainly been the impossibility of being provided with perfect a priori information. The authors propose an architecture integrating sensory skills both as an aid to the manipulation process and as a means to extract information from the scene. The vision based self-calibration is described together with the first steps in closing the control loop with vision.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>