Visual servoing using fuzzy controllers on an unmanned aerial vehicle

This paper presents an implementa-
tion of three Fuzzy Logic controllers
working in parallel onboard a UAV,
two for a pan-tilt camera platform
and the third for control the yaw of
the helicopter. This implementation
uses a Lucas-Kanade tracker algo-
rithm with a pyramidal optical
ow
implementation, which gives infor-
mation to follow statics and moving
objects, besides the UAV vibrations
and movements. The platform con-
troller is helped by the heading con-
troller, in order to make smooth the
big movements to the platform, re-
ducing the risk of lost the warp selec-
tion of the object to track. Also, the
heading control remove the physic
limit of the platform at the yaw axis.
Some laboratory and UAV tests are
presented in order to show the differ-
ent behaviors and the good response
of the presented controllers.