servoOff command becomes unpredictable when pause button (H/W) used

[130s]

Pause button (black lever in this image) is recently introduced by the manufacturer, which stops the movement while servo is kept on.

HIRONX.servoOff turns off servo after certain time. When the pause feature is used during servoOff, arms stop first but in the worst case they fall after that period passes (confirmed with the real robot), which may be counter-intuitive.

[k-okada]

I'd like to know if there is a function that can tell whether the button is pressed or not in iob level, which inly manufacturer knows.

◉ Kei Okada

On Fri, Dec 5, 2014 at 3:56 PM, Isaac I.Y. Saito notifications@github.com wrote:

Pause button (black lever in this image http://wiki.ros.org/rtmros_nextage/Tutorials/Quick%20Start%20for%20Robot%20Users?action=AttachFile&do=get&target=relay_box.jpg) is recently introduced by the manufacturer, which stops the movement while servo is kept on.

HIRONX.servoOff https://github.com/start-jsk/rtmros_hironx/blob/f551037aa697c1cf8b0faf1291e4c3565eba0059/hironx_ros_bridge/src/hironx_ros_bridge/hironx_client.py#L513 turns off servo after certain time. When the pause feature is used during servoOff, arms stop first but in the worst case they fall after that period passes (confirmed with the real robot), which may be counter-intuitive.

— Reply to this email directly or view it on GitHub https://github.com/start-jsk/rtmros_hironx/issues/295.

[130s]

I'd like to know if there is a function that can tell whether the button is pressed or not in iob level, which only manufacturer knows.

@emijah Would you shed a light on this? Is the API between public and proprietary iob (planned to become) open for Hiro/NXO?

[130s]

Now that we have hrpiob opensourced, read_servo_state or something might do the work we want.

If someone who has an access to a robot can write a code and test that'd be a great pull request candidate.