hironx init pose may not be set factory initial value

[130s]

I don't know if this applies to all Hiro but for some newer robots, the pose HIRONX.goInitial sets is different from its factory initial pose.

For NextageClient, it's adjusted.

Unless there's a need I want the same value for HIRONX so that I can use the same code between Hiro and NXO.

I can't find a commit log where the current value was set (log).

[emijah]

It may have been there from the beginning. The default value for GRX was [-0.6, 0, -100, 15.2, 9.4, 3.2], [ 0.6, 0, -100, -15.2, 9.4, -3.2], However, the new values should be the ones used to check joint alignment.

[k-okada]

what is factory initial pose ??? the pose which the robot is set after the calibration?

On Wed, Jun 4, 2014 at 2:44 PM, Hajime SAITO notifications@github.com wrote:

It may have been there from the beginning.

— Reply to this email directly or view it on GitHub https://github.com/start-jsk/rtmros_hironx/issues/107#issuecomment-45053530 .

[130s]

what is factory initial pose ??? the pose which the robot is set after the calibration?

I meant factory initial pose as the one manufacturer uses for calibration. As I said, initial pose of NextageClient is already set with the manufacturer's value.

If this suggestion is acceptable, I'll apply the change. Then we need to notify the current users since this might break their application program.

[k-okada]

Let me clear the situation

What is the reason of changing initial pose to factory calibration pose?(What's the dis-advantage of goInitial() is different from factory initial pose?)

Does factory initial pose happen only after checkEncoders() ? what's will happens if we put goInital in the end of checkEncoders() function?

Basically,

                   [-0.6, 0, -140, 15.2, 9.4, 3.2],
                   [0.6, 0, -140, -15.2, 9.4, -3.2],

is better than

                   [-0, 0, -130, 0, 0, 0],
                   [0, 0, -130, 0, 0, 0],

because, the angle of the end-effecter becomes 0. so that's seems better initial posture for users to start with, specially, they uses setTargetPoseRelative()

[130s]

What is the reason of changing initial pose to factory calibration pose?(What's the dis-advantage of goInitial() is different from factory initial pose?)

What I want is:

• (1st Priority) To have a common pose goInitial in the class HIRONX and NextageClient.
• To have factory init pose as a readily available option; For some hardware tests where manufacturer is involved (eg. acceptance test at their office, which most of all users basically have nothing to do with), they use factory initial pose.

Does factory initial pose happen only after checkEncoders() ? what's will happens if we put goInital in the end of checkEncoders() function?

I doesn't matter when the factory init pose happens, I think.

Basically,

               [-0.6, 0, -140, 15.2, 9.4, 3.2],
               [0.6, 0, -140, -15.2, 9.4, -3.2],

is better than

               [-0, 0, -130, 0, 0, 0],
               [0, 0, -130, 0, 0, 0],

because, the angle of the end-effecter becomes 0. so that's seems better initial posture for users to start with, specially, they uses setTargetPoseRelative()

I agree, and I'm sure @emijah does too.

So in conclusion, I guess I'll open a solution PReq that does:

• In NextageClient, set the same initial pose that HIRONX uses (ie. the one that sets the eef level up).
• In HIRONX, add an optional argument in goInital to set the factory initial pose as the initial pose.

[emijah]

Yes I agree too.