Kinematics base update #57
Conversation
…ositionIK(...) method that returns multiple solutions
…amples to generate
|
@gavanderhoorn ping |
|
Yeah I know. Might be able to do it this week. |
|
@gavanderhoorn I know you're super busy, but this still needs a fair shake. |
| ss<<sol[i]<<" "; | ||
| } | ||
| ss<<"]"; | ||
| ROS_DEBUG_NAMED("ikfast","Sol %d: %s", s, ss.str().c_str()); |
There was a problem hiding this comment.
this is that inefficient section that was removed in another PR, right? we don't need this stringstream...
|
+1 |
|
@jrgnicho Ping |
|
@davetcoleman I added the requested explanation comments and removed the unnecessary console debug message printing. |
|
@davetcoleman These minor changes make the indigo-devel branch slightly different from the jade-devel branch. Not sure that this is a problem but if it is then I could create PR's for jade-devel and kinetic that will push these changes into those branches. |
|
Mirroring these changes in the J/K branches sounds like a good idea to me. |
|
@davetcoleman then I'll create PR's into those branches to mirror the changes here. |
Mirrors changes in PR #57
This PR addresses the issue discussed in #56
In a nutshell, this PR updates the ikfast template so that it overloads the getPositionIk(...) method in the moveit_core::KinematicsBase class that returns multiple solutions.
@gavanderhoorn please review