Repeated stopping of PR2 Motors (doesn't appear to be a caster slip ring issue)

[Arvind Car]

Dear PR2 users,

I am hoping for some help debugging an issue I am facing with the PR2, where it stops a few seconds after starting a trajectory (whether it’s the base, arms, head, torso, etc.), almost as if the stop button on the controller module was pressed. However, the controller module has new batteries.

After reviewing similar posts, it appears to me that our issue is not related to the slip ring of the casters, based on the rqt_pr2_dashboard (screenshot attached). I also get "Safety Lockout: UNDERVOLTAGE" for the joints of the right arm whenever this behaviour occurs (screenshot attached).

(Additional context (not sure if this is relevant): This problem is fairly recent, and the sequence of events leading up to it is as follows: I was attempting to move the base of the PR2 using the script provided here: Using the base controller with odometry and transform information. I was able to move it forward and backward with no issues. When I tried to command an angular velocity, the casters rotated and then stopped suddenly. I couldn't send any further commands to the robot, so I restarted it. Since then, I have been facing the above issue.)

Has anyone encountered a similar problem or have any ideas or suggestions on what might be causing this issue and how to resolve it?

Any help would be greatly appreciated!

Regards,
Arvind

[Michael Görner]

Hi Arvind,

sorry for not digging through the material in depth and providing more
links, but here's my first impression. Let me start out saying that our
lab's robot did not have these problems with either arm (yet).

In general there's a lot of errors in your dashboard. By the looks of it
most of them are regulars on your setup and point out broken
configurations (since the indigo upgrade?) Sadly many PR2s have a long
list of such errors in everyday operation, because fixing them adds
quite a bit of work to the local maintenance overhead. Still, an error
that states the number of CPUs is wrong always strikes me as bizarre..
The right arm ethercat errors aside, the main thing I don't know about
is the missing IBPS status messages. Do your batteries still exist/work?
Might either be a simple misconfiguration issue on the Linux side
(e.g., wrong usb devices configured or invalid permissions) or a
relevant error in the battery power management (board not responding).

You show UNDERVOLTAGE errors only for the right arm, so I expect only
the right arm power system is effected. the two arms and the rest of
the robot have in total three different power circuits maintained by the
power board, so the power issue might be related to the right arm
circuit only. Supported by your screenshot of the dashboard, top left,
where two question marks (missing icons?) are green (high-power mode?)
and the third one yellow (low-power mode?).

Maybe disable the right arm entirely, possibly through software
(disabling the right arm in the dashboard, replacing the right arm
joints by fixed joints in the URDF to stop the driver from talking to
the MCBs - not sure if either works for the arm) or for testing maybe
also in hardware by unplucking it from the power board or explicitly
disconnecting power for the MCBs in the right arm?

Here's the first documentation I can find on the power board in
general, but you can find more relevant things by digging a bit deeper.
http://www.clearpathrobotics.com/wp-content/uploads/2014/11/pr2_power_board_rr.pdf

Also, the power board has status LEDs which might blink to indicate
error types when it switches away from high-power mode. I don't have
that documentation at hand, but there is a loose list of blinking codes
and explanations somewhere.

If you are unlucky the power board might slowly die, in which case you
either have a spare one or you don't.
I just checked JSK's list of spare components - Mail on Nov 15, 2022 on
this list - and am not sure whether the "Power board circuit for
computer" entry is the whole (or relevant) part you would need.

If you are lucky it might be that one of the slip rings in the right
arm has a similar connectivity issue as the one the casters' tend to
develop. In-between is a breaking Motor Control Board for a motor in
the arm. Easier to replace than the power board, but you might still
need to get one sent from JSK if you don't have a stash, and will need
to program it correctly. The gripper MCB is also a different beast I
think and JSK has much fewer of these stashed.

Best of luck and greetings from Hamburg Germany,

Michael Görner / v4hn

On Wed, 17 Jul 2024 14:09:40 -0700 (PDT)
Arvind Car <ac...@andrew.cmu.edu> wrote:
> Dear PR2 users,
> I am hoping for some help debugging an issue I am facing with the
> PR2, where it stops a few seconds after starting a trajectory
> (whether it’s the base, arms, head, torso, etc.), almost as if the
> stop button on the controller module was pressed. However, the
> controller module has new batteries.
>
> After reviewing similar posts, it appears to me that our issue is not
> related to the slip ring of the casters, based on the
> rqt_pr2_dashboard (screenshot attached). I also get "Safety Lockout:
> UNDERVOLTAGE" for the joints of the right arm whenever this behaviour
> occurs (screenshot attached).
>
> (Additional context (not sure if this is relevant): This problem is
> fairly recent, and the sequence of events leading up to it is as
> follows: I was attempting to move the base of the PR2 using the
> script provided here: Using the base controller with odometry and
> transform information

> <http://wiki.ros.org/pr2_controllers/Tutorials/Using%20the%20base%20controller%20with%20odometry%20and%20transform%20information>.

> I was able to move it forward and backward with no issues. When I
> tried to command an angular velocity, the casters rotated and then
> stopped suddenly. I couldn't send any further commands to the robot,
> so I restarted it. Since then, I have been facing the above issue.)
>
> Has anyone encountered a similar problem or have any ideas or
> suggestions on what might be causing this issue and how to resolve
> it?
>
> Any help would be greatly appreciated!
>
> Regards,
> Arvind
>

--
Michael 'v4hn' Görner, M.Sc. Cognitive Science, Research Associate
Universität Hamburg
Faculty of Mathematics, Informatics and Natural Sciences
Department of Informatics
Group Technical Aspects of Multimodal Systems

Vogt-Kölln-Straße 30
D-22527 Hamburg

Room: F-315
Phone: +49 40 42883-2432
Website: https://tams.informatik.uni-hamburg.de/people/goerner/