One of the most desirable characteristics of a robotic manipulator is
its flexibility. Flexibility and adaptability can be achieved by incor
porating vision and generally, sensory information in the feedback loo
p. Our research introduces a framework called controlled active vision
for efficient integration of the vision sensor in the feedback loop.
This framework was applied to the problem of robotic visual tracking a
nd servoing, and the results were very promising. Full 3-D robotic vis
ual tracking was achieved at rates of 30 Hz. Most importantly, the tra
cking was successful even under the assumption of poor calibration of
the eye-in-hand system. This paper extends this framework to other pro
blems of sensor-based robotics, such as the derivation of depth maps f
rom controlled motion; the vision-assisted grasping; the active calibr
ation of the system robot-camera; and the computation of the relative
pose of the target with respect to the camera. We address these proble
ms by combining adaptive control techniques with computer vision algor
ithms. The paper concludes with a discussion on several relative issue
s such as the stability and robustness of the proposed algorithms and
the problem of incorporating stereo information in the existing algori
thms in order to increase the accuracy of the estimated depth.