15-494/694 Environment Exploration Project
Given an occupancy grid representation of the world, there will
be grid cells whose contents are unknown because we have
not imaged that area. We need an algorithm to plan our exploration of
the environment to efficiently reduce uncertainty.
The first step of this project is to develop heuristics for
deciding where to explore. For example, if there is a wall in
the world map and we don't know what's on the other side, it
makes more sense to explore that area than to wander off into
empty space.
For an example of an occupancy grid representation, look at the
cozmo-tools wavefront path planner in wavefront.py. You can see
the grid in operation by doing "show path_viewer" in simple_cli
and asking the robot to navigate to a cube,
e.g., GoToCube(cube1).now().
|