0
$\begingroup$

Rosanswers logo

Hello,

The tutorial for Navigation Stack states, since the Global Planner assumes circular robots, it produces waypoints that are optimistic for the actual robot footprint, and therefore the generated path may be infeasible to follow.

I was wondering if it is still the case if we consider the circumscribed circle of the robot's footprint. In this case, is it not even more conservative?

Thanks.


Originally posted by ROSCMBOT on ROS Answers with karma: 651 on 2014-06-04

Post score: 0

$\endgroup$

1 Answer 1

0
$\begingroup$

Rosanswers logo

If you pass in the circumscribed circle as the inner radius it will be conservative, i.e. plans should be feasible. On the other hand you might loose plans that would work, i.e. it's incomplete.

All this relates to global planners such as NavFn. There are global planners that plan actual trajectories in the configurations space and thus are correct (but usually slower).


Originally posted by dornhege with karma: 31395 on 2014-06-05

This answer was ACCEPTED on the original site

Post score: 2


Original comments

Comment by ROSCMBOT on 2014-06-05:
Thanks.

1- Does NavFn consider the radius of the inscribed circle of the robot's footprint?

2- Could you name Global Planners in ROS that plan actual trajectories in the configuration space?

Comment by dornhege on 2014-06-05:\

  1. yes, 2. One example that can do that: http://wiki.ros.org/sbpl
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.