ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question

Stefan Stiene's profile - activity

2013-01-17 07:10:29 -0500 received badge  Famous Question (source)
2013-01-17 07:10:29 -0500 received badge  Popular Question (source)
2013-01-17 07:10:29 -0500 received badge  Notable Question (source)
2012-12-13 20:10:32 -0500 received badge  Famous Question (source)
2012-12-13 20:10:32 -0500 received badge  Popular Question (source)
2012-12-13 20:10:32 -0500 received badge  Notable Question (source)
2011-11-03 01:18:39 -0500 asked a question why are the wait motions ignored in sipp dynamic path planner

Hi, I have tested the sipp library of the sbpl_dynamic_planner. I used the EnvIntervalLat I get the following warning:

WARN: wait motion in mprim file ignored

And the resulting path also looks like the warning is right. The robot drives slowly backwards until the obstacle disappears that blocks the way to the goal. Afterwards it drives forwards to the goal. The desired path is that the robot drives slowly forwards and wait in front of the obstacle. What is the reason why the wait motion is ignored?

Best Stefan

2011-09-02 11:14:46 -0500 marked best answer goal time window for dynamic sbpl planner

Currently there isn't support for this type of constraint (directly). The planner tries to find a time-minimal collision free path to the goal. However, it would be possible to add this extension to the planner (and I'll do so in the next release).

Until then though an easy hack to get the planner to reach the goal on the interval (t1,t2) would be to put a "fake" dynamic obstacle at the goal location from time 0 to t1 and another fake dynamic obstacle at the goal location from time t2 to some large time. This is essentially how I will implement your feature in the next release anyway.

2011-06-19 21:49:09 -0500 asked a question goal time window for dynamic sbpl planner

Hello, I have a question concerning the new dynamic version of the sbpl planner http://www.ros.org/wiki/sbpl_dynamic_... . Is it possible to give the goal a permitted time window? I want the sbpl to give me only solutions where the robot reaches the goal at a specific time. If this is not possible, how complex would it be to extend the current solution in this direction? Best Stefan