Brian Hou

We only need to check the formatting on one build.

I think the current way that we've been using this in `libherb` is to have a sequence of `[ConfigToConfig, ConfigToTSR]` where `ConfigToTSR` is an explicit adaptation of the `ConfigToConfig`. It...

I haven't thought about this very much, but I initially thought it might make sense to add trajectory constraints as a part of all `Problem`s rather than as a specific...

Hmm, I could be wrong but I was under the impression that @sniyaz tried to preserve the locking in the lowest level `planTo` functions (e.g. https://github.com/personalrobotics/aikido/blob/master/src/planner/vectorfield/VectorFieldPlanner.cpp#L151). Regardless, I think this...

We should use DART's `MetaSkeleton::getLockableReference()` method rather than locking `Skeleton`s (available since DART 6.5). @jslee02's comment from #417: > Previously, AIKIDO had to use `Skeleton` rather than `MetaSkeleton` due to...

Now that the planner API is mostly complete, we should consider revisiting this issue.

Maybe we should be setting this at the `Manipulator` level?

If I remember correctly, there are still alignment issues but @dqyi11 believes that they're in DART rather than AIKIDO at this point. If that's the case, maybe we should try...

Do you have a small code snippet that we can use to replicate this? If it's working properly, that error is thrown when the `TrajectoryMarker`'s skeleton doesn't match the skeleton...

Our current version of clang-format (3.8, the latest version available on Trusty) doesn't allow us to automatically handle this. Upgrading to Xenial should allow us to use clang-format-4.0. We can...