Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Increase the damping of iCubGazebov3 and the PID gains #197

Closed
S-Dafarra opened this issue Mar 24, 2021 · 9 comments · Fixed by #199
Closed

Increase the damping of iCubGazebov3 and the PID gains #197

S-Dafarra opened this issue Mar 24, 2021 · 9 comments · Fixed by #199
Assignees

Comments

@S-Dafarra
Copy link
Contributor

The damping is currently set ot 0.06 for all the joints. This value is very small, also considering the real robot. After some tests, 2.0 seems a more reasonable value.

Regarding the PID, I used the following values P 40.0 I 0.35 D 0.35 on all the joints, and seemed to be working fine, at least for
the walking.

I can also prepare a PR if you point me to the correct files to modify.

cc @Nicogene @traversaro

@Nicogene
Copy link
Member

Hi @S-Dafarra !

The damping can be specified in the csv (e.g. https://github.com/robotology/icub-models-generator/blob/master/simmechanics/data/icub3/ICUB_3_joint_all_parameters.csv)
And the PID can be changed in all the control board ini files (see

kp 1.745 5.745 15.745
kd 0.174 0.174 0.174
ki 0.174 4.174 3.174
for example)

A PR is more than welcome, just remember to change all the csv files here https://github.com/robotology/icub-models-generator/tree/master/simmechanics/data/icub3, for now we are using the one with all the joints in one, but in the future we would follow the modular approach

@traversaro
Copy link
Member

A PR is more than welcome, just remember to change all the csv files here https://github.com/robotology/icub-models-generator/tree/master/simmechanics/data/icub3, for now we are using the one with all the joints in one, but in the future we would follow the modular approach

Can't we just delete the redundant one, and once we want to go for the modular approach we take the all joint one and decompose it in the separate one? Trying to maintain two different "non"-single sources of true it tricky.

@Nicogene
Copy link
Member

Can't we just delete the redundant one, and once we want to go for the modular approach we take the all joint one and decompose it in the separate one? Trying to maintain two different "non"-single sources of true it tricky.

I am afraid that it is not yet possible because the csv it taken by the simmechanics-to-urdf script with the simmechanics XML and for now we have not all the modular simmechanics XMLs and we do not know how to assemble the resulting urdfs.

Or it is possible to pass to the script more than one csv??

@traversaro
Copy link
Member

traversaro commented Mar 25, 2021

Can't we just delete the redundant one, and once we want to go for the modular approach we take the all joint one and decompose it in the separate one? Trying to maintain two different "non"-single sources of true it tricky.

I am afraid that it is not yet possible because the csv it taken by the simmechanics-to-urdf script with the simmechanics XML and for now we have not all the modular simmechanics XMLs and we do not know how to assemble the resulting urdfs.

Or it is possible to pass to the script more than one csv??

Sorry, I meant to delete the modular one, as those are the one that are the moment are not used. In any case, this is not a big priority.

@Nicogene
Copy link
Member

Sorry, I meant to delete the modular one, as those are the one that are the moment are not used. In any case, this is not a big priority.

Yeah we can restore them later if necessary

@S-Dafarra
Copy link
Contributor Author

So the final verdict? I can remove them if you want, less files to edit 😁

@traversaro
Copy link
Member

So the final verdict? I can remove them if you want, less files to edit 😁

Let's remove them!

@traversaro
Copy link
Member

Because it seems that ICUB3_L_LEG_SIM_MODEL.xml, ICUB3_R_LEG_SIM_MODEL.xml and ICUB3_UPPERBODY_SIM_MODEL.xml are already lagging behind w.r.t. to ICUB3_ALL_SIM_MODEL.xml, so better to remove them before creating more confusion.

@S-Dafarra
Copy link
Contributor Author

I have noticed those files are reference here:

simmechanics_to_urdf ICUB3_UPPERBODY_SIM_MODEL.xml --csv ICUB_3_joint_upperbody_parameters.csv --yaml ICUB_3_upperbody_options.yaml --outputfile model.urdf
gz sdf -p model.urdf > iCub3Upperbody.sdf
simmechanics_to_urdf ICUB3_L_LEG_SIM_MODEL.xml --csv ICUB_3_joint_leftleg_parameters.csv --yaml ICUB_3_leftleg_options.yaml --outputfile model.urdf
gz sdf -p model.urdf > iCub3LeftLeg.sdf
simmechanics_to_urdf ICUB3_R_LEG_SIM_MODEL.xml --csv ICUB_3_joint_rightleg_parameters.csv --yaml ICUB_3_rightleg_options.yaml --outputfile model.urdf
gz sdf -p model.urdf > iCub3RightLeg.sdf
# simmechanics_to_urdf ICUB3_HEAD_SIM_MODEL.xml --csv ICUB_3_joint_head_parameters.csv --yaml ICUB_3_head_options.yaml --outputfile model.urdf
# gz sdf -p model.urdf > iCub3RightLeg.sdf

Should I also comment out those lines?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

Successfully merging a pull request may close this issue.

3 participants