ergoCub 1.1 S/N:001 – Robot joints are miscalibrated

[GiulioRomualdi]

Robot Name 🤖

ergoCub 1.1 S/N:001

Request/Failure description

This morning I tried to run a classical walking experiment. The robot is not able to walk anymore since the joints of the legs seems completelly miscalibrated.

Detailed context

miscalibration.mp4

Additional context

No response

How does it affect you?

The robot does not walk anymore. The calibration is required to make the robot walk.

cc @isorrentino @LoreMoretti @DanielePucci @S-Dafarra

[github-actions]

⚠️ Detected missing mandatory information

Hi @GiulioRomualdi 👋🏻

Some of the following points need your attention.

You are required to:

• enter the Robot Name.
• give the issue a meaningful Title.
• provide a Detailed Context (at least 100 characters).

_{Please, mark the points above as solved once done.}

[Fabrizio69]

@DanielePucci told me that the hardware limits (pins) are bent and this is the cause of the miscalibration of the joints.
To avoid this problem, @isorrentino changed the offsets for hip pitch and ankle pitch, but for these joint the hardware limits are these (there aren't pins):

Another possibility are the knee joints:

but also in this case there isn't a pin. We can try to tighten the screws...

@isorrentino if I made a mistake in my description above, please report the corrections, thank you very much

cc: @fbiggi , @AntonioConsilvio

[GiulioRomualdi]

Recalibrating the entire robot might not be the optimal solution. In the past, we encountered a miscalibration caused by a broken pin (see #1838).

Since we don’t know which specific joint is miscalibrated, I would avoid recalibrating all of them. Instead, we might focus on identifying the problematic joint (though I'm not sure how to do this) and addressing that one.

Correct me if I’m mistaken, but if the robot is miscalibrated, it means that the encoder value when the robot reaches its limit differs from the previous value. If this is the case, it could be due to an encoder malfunction or relative motion between the hard stop and the mechanical part that contacts it.

[Fabrizio69]

Hi @GiulioRomualdi , we have verified that all hardware limits are correctly fixed (and there aren't bent dowel pins)

cc: @AntonioConsilvio

[AntonioConsilvio]

Hi @GiulioRomualdi! Were you able to recalibrate the robot properly?

[GiulioRomualdi]

We calibrated them at humanoids

[AntonioConsilvio]

Great, let's get on with closing the issue!