You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Default heuristics for ranking the frontiers uses plane Euclidean metric for evaluating the distance.
As we have a highly structured environment (indoor strongly resembles task of navigation inside the labyrinth) it would be better to use geodesical path distance: 'real distance we have to traverse to reach the point'. For this, we can reuse produced path from the ros_nav nodes (move_base group of packages)
TODO: add parametrisation for weight of this heuristic.
The text was updated successfully, but these errors were encountered:
Default heuristics for ranking the frontiers uses plane Euclidean metric for evaluating the distance.
As we have a highly structured environment (indoor strongly resembles task of navigation inside the labyrinth) it would be better to use geodesical path distance: 'real distance we have to traverse to reach the point'. For this, we can reuse produced path from the ros_nav nodes (move_base group of packages)
The text was updated successfully, but these errors were encountered: