Cynthia Vaskis
SLM521 Spring
2004
Web Quest Assignment
File: wqrover1.htm
Summary of Rover Autonomous Navigation video’s audio
The rover needs to avoid navigational
hazards as large rocks and steep slopes while traversing the Martian
terrain. The rover uses paired cameras
on its head to take stereo vision pictures that it compares to determine how
far objects are from the rover. It also
uses this data to create a three dimensional patch grid surrounding the rover’s
current position and in the direction of the rover’s goal or target
location. The rover’s traversing program
finds several safe driving pathways to try and selects one based on how close
it will get the rover to its target location in the safest and shortest
distance.
The rover selects one of
these paths and moves a short distance along it looking downward with another
pair of cameras on its front underside to avoid going over rocks that could
block its wheels or hang up under its underside. When the rover’s traversing program sees
obstacles in the pictures and while making the terrain grid database, it
creates a model for the obstacles on the terrain grid and expands the size of
these obstacles so that the rover does not get close to them when selecting a
pathway to travel. After the rover has
moved a short distance, the process repeats over taking pictures again and
creating a newer terrain grid.
The rover does not move
ahead any more if it reaches its goal location or if no safe pathway is found
or if Mission Control redirects the rover to stop and do something else. Recently, one of the rovers got stuck with no
safe pathway so Mission Control devised a way to have the rover spin around in
place by toeing its front and back wheels inward to form a circle pattern. Then the rover could turn in place and try a
new directional heading.