[ompl-users] Motion planning of quadruped robot

[Jan K. <squ...@gm...>]

Hi,

Unfortunately, due to my failure, I missed OMPL in research for my
university thesis.

I am trying to write an ROS software for quadruped robot (spider-style
mechanics).  After some basic ideas, I picked sample-based motion
algorithm. I am able to reasonably define the state of the robot, evaluate
simple linear changes of that state and make an efficient search using
Anytime D* Lite. (In theory)

After some implementation in progress. I rechecked actual state of the art
and realized that this is how OMPL works. Which for some reason I missed
previously. (And I consider myself dumb for that)

So, I am in a state when I have mechanics for moving the legs of the robot
based on state search.

Are you aware of anybody trying to do the same? This is not the standard
usage of OMPL and I prefer to ask here, rather than risk that my "search
skill" is going to fail again.

Sincerely,
John Koniarik
squ...@gm...