Expose state estimation interface

[Stefan Kohlbrecher]

We talked about this in SC, but as with other issues, it probably makes sense to track this here also. I believe quite a few teams have already integrated MIT´s pronto state estimator and would like to continue to do so. This raises the question how that integration would work with the IHMC controller.

One option is to make the data available on the ROS interface. As there already is a ROS-LCM bridge available with Pronto, this is probably easiest integration-wise.

The other option I see is a IHMC-LCM bridge, which would cut down on conversion overhead and delay (no IHMC-ROS-LCM travel and vice versa) .

[btshrewsbury NA]

We are working on the java-ros-lcm pipeline. We haven't seen much delay in the java-ros-lcm conversion, but we are pushing our state estimator values to pronto which slows down the update rate. We will be pushing out the new msgs and launch file soon.

[DouglasS]

There seems to be enough differences between Atlas V3 and Atlas V5 that while we believe we are exposing all of the correct topics we are having a hard time doing verification with the real Pronto stack. We need to revisit this when we have more time.