search

robotology / human-dynamics-estimation

Software repository for estimating human dynamics
BSD 3-Clause "New" or "Revised" License
80 stars 28 forks source link

Add floor contact target with force-torque measurements #300

Closed lrapetti closed 2 years ago

lrapetti commented 2 years ago

Adding target type floorContact which takes as input a force-torque measurement, compare it with a threshold, and in case it is higher it fixies the x-y position of the contact and fixes the z to zero.

This allow floating base estimation with orientation+force-torque sensors only like:

https://user-images.githubusercontent.com/35487806/170206990-c09704e0-8ab1-4981-8338-d4ce8baa9dff.mov

lrapetti commented 2 years ago

The history of the commits need to be cleaned

lrapetti commented 2 years ago

The history of the commits need to be cleaned

Commit history is now cleaned

prashanthr05 commented 2 years ago

@lrapetti let us know if this PR can be reviewed already?

lrapetti commented 2 years ago

@lrapetti let us know if this PR can be reviewed already?

yes, it can be reviewed now, thanks!

lrapetti commented 2 years ago

Hi @prashanthr05, with some delays I have addressed your comments. Do you have other comments?

@RiccardoGrieco are you planning to review this PR?

RiccardoGrieco commented 2 years ago

For what concerns the check failures, I think that https://github.com/robotology/human-dynamics-estimation/issues/296 was addressed only for master.

RiccardoGrieco commented 2 years ago

For what concerns the check failures, I think that #296 was addressed only for master.

In 0729c6c I replaced the is/asDouble() methods with is/asFloat64().

Hopefully this will make all the checks pass and will fix the CI failures also on devel once the PR is merged.

lrapetti commented 2 years ago

For what concerns the check failures, I think that #296 was addressed only for master.

In 0729c6c I replaced the is/asDouble() methods with is/asFloat64().

Hopefully this will make all the checks pass and will fix the CI failures also on devel once the PR is merged.

Thanks a lot @RiccardoGrieco

lrapetti commented 2 years ago

@prashanthr05 do you have any additional comment? Otherwise we would proceed merging it in devel

prashanthr05 commented 2 years ago

Thanks @lrapetti ! No further comments from my side since @RiccardoGrieco also handled the YARP deprecations. We may proceed to merge.

RiccardoGrieco commented 2 years ago

With @lrapetti we noticed that the changes introduced in https://github.com/robotology/human-dynamics-estimation/pull/299 (commit 69d15cb ) caused a segmentation fault due to an uninitialized unique_ptr.

In 9cfaa99 this has been fixed.

RiccardoGrieco commented 2 years ago

@lrapetti the checks passed, I think you can proceed! 🚀

lrapetti commented 2 years ago

@lrapetti the checks passed, I think you can proceed! 🚀

Check passed, we can close it!

prashanthr05 commented 2 years ago

Awesome!!