The objective of Stereoscopic Imaging for Slow-Moving Autonomous
Vehicle,
SISAV, is to develop a system that can provide an accurate terrain map
to be
used for navigating an autonomous vehicle. A well known example of
this is
the Mars Exploration Rovers engineered by NASA, which can be seen here. The
system will use two digital cameras to perform the stereoscopic imaging to obtain
a terrain map of the objects in front of the cameras. This map will be used by the
control algorithm running the autonomous vehicle to determine the direction in
which to move. The system will have two modes of operation: calibrate and run.
Calibration mode involves calculating the intrinsic and extrinsic parameters of each
camera by using a calibration rig with known geometry and easily detectable features.
Run
mode involves sending a signal to have the cameras take a set of
pictures and
processing the images to obtain the terrain map and distances to the
objects,
and then using this information to navigate an autonomous vehicle.