Changes some frames from world to body conversion for NED to ENU.

[tonybaltovski]

For a desired position that is (1,5, 7) in ROS ENU, the must go forward 1 m which is X for both the ROS coordinate system and the PX4. For y-axis, it is -5 m on the PX4 and the same for the z-axis. I believe @mhkabir can verify this. The frame issue #49 with two different frame conversions is due to keeping X as forward on the physical vehicle in ROS's coordinate system. If we made Y forward, the IMU frames would be one conversion for body and inertial from NED to ENU thus being simpler to use. However, this does break the ROS standards for coordinates. @vooon Any insight?

[TSC21]

@vooon ping.
@mhkabir are you able to test this?

[TSC21]

Related: #225 (comment). Is this supposed to work?

[mhkabir]

This PR is actually not the way to go forward. This breaks ROS standards.

Please refer to #216 and add your comments

[tonybaltovski]

This does not break any ROS standards.

Closed #208 #208.

—
Reply to this email directly or view it on GitHub
#208 (comment).

[mhkabir]

@tonybaltovski Ok, in that case, we can get this back up after verification. It's gone stale anyway.
Thanks for your contributions.
Can you comment on the discussion please?

[vooon]

@mhkabir reopened just not to lose pr.

[tonybaltovski]

I believe that @congleetea confirmed this in #216 for the position frames.

[vooon]

Ok, so i do merge.

[TSC21]

@vooon I don't think this was supposed to merge yet. For all I know, @tonybaltovski was testing it today...

[vooon]

Ooops. I may revert if needed.

[TSC21]

Better you do it then :)

[vooon]

That generate new merge, so before revert i want to hear "thats bug, revert it with fire!" :)

[TSC21]

Ahahah sure! Let's see what @tonybaltovski tests say.

[TSC21]

@tonybaltovski vision_speed_estimate plugin also require this changes (if they are to be accepted). Would you mind add fix to those to? Thanks!