Closed gmfricke closed 3 years ago
@abbypribis not happy with the escape procedure on some maps. I am redesigning it.
I was?
@abbypribis lol. I am just tagging you in the comment not implying you are the source.
Is this the search strategy for round 2, since we have the volatile locations?
@abbypribis It is a search for a path to a point yes and yeah we would need it for round 2 to get to the vols and for round 3 to get back home(lander), because we don't have a planner this is the next best thing.
In Slack, you mentioned that this node could be fed waypoints from the qual 2 volatile locations service? How would I do that?
In Slack, you mentioned that this node could be fed waypoints from the qual 2 volatile locations service? How would I do that?
@abbypribis, If it had a scoot instance it could just call scoot.get_closest_vol_pose() and that would return the pose of the closest volatile but that should only work for round 2 Also it might be launch-able by a behavior and the behavior publishes the waypoint for it
First version. Reaches the goal location 70% of the time. Needs lots of tuning. Improvements: the main issue is that the robot can get trapped when climbing a slope while obstacle avoiding.
To run: start simulation
rosrun scoot FakeLocalization.py
rosrun navigation bug_nav.py bug0 0 0
Starts the bug_nav node and has it try to navigate to 100 randomly chosen waypoints. The first waypoint is the coordinates provided on the command line.