Sign of torque values

Questions about our Python API
Post Reply
AdityaNair
Posts: 6
Joined: Fri Jul 15, 2022 3:49 am
Location: Pilani, Rajasthan

Sign of torque values

Post by AdityaNair » Wed Oct 26, 2022 11:36 pm

Hello again!

We're try to do something with the daisy hexapod's torque feedback. What I'm unable to understand is for the same motion, the torques in the leg (obtained using the getJointState or the get_next_feedback function) change signs for the left and right legs. For example, if the middle right leg is lifted in the air the torque in the shoulder will be some positive value but if the middle left leg is lifted, the torque in the respective shoulder will be the same value but negative.

This problem becomes worse because it leads to an incompatibility with the get_grav_comp_efforts function which gives the external torques due to gravity. For the same case, this function gives a positive valued shoulder torque for both the middle legs.

Thanks in advance for all your support :)
Aditya Nair
kamal_carter
Posts: 5
Joined: Fri Jul 09, 2021 1:45 pm

Re: Sign of torque values

Post by kamal_carter » Thu Oct 27, 2022 4:32 pm

Hello Aditya,

From what you described, that is the desired/correct feedback you are getting for the hexapods' torques at the legs.
hexpodLeftRightLabel.png
hexpodLeftRightLabel.png (173.68 KiB) Viewed 5785 times
Using the image above you can see actuators are oriented differently between the right and left leg for easier mounting. This mounting makes it so the shoulder and elbow actuators on the legs on the left and right will actually have opposite efforts signs with each other.

Let us know if this helps, for further questions you can use the forums or send us an email to setup a video call.
Attachments
hexapodLabel.png
hexapodLabel.png (173.62 KiB) Viewed 5785 times
AdityaNair
Posts: 6
Joined: Fri Jul 15, 2022 3:49 am
Location: Pilani, Rajasthan

Re: Sign of torque values

Post by AdityaNair » Fri Oct 28, 2022 3:38 am

Hi Kamal!

Thanks again for the response. Yeah this does make a lot of sense. Although, I still don't understand why the get_grav_comp_efforts function is giving the same positive value for the left and right legs. This is the trouble I'm having with get_grav_comp_efforts because it doesn't seem to respect these frames. I'd love to have a video chat too to discuss this more. I've sent you a private message to say hi.

PS: Loved your aimbot video :)
Post Reply