Enrico Mingo Hoffman

Hi all, I was starting to implement a new interface for a novel QP solver named [qpSWIFT](https://github.com/qpSWIFT/qpSWIFT) for CasADi. I was following the examples of [osqp_interface.hpp](https://github.com/casadi/casadi/blob/master/casadi/interfaces/osqp/osqp_interface.hpp) which derives from `Conic`....

[MJX] efc_pos missing in mjx.Data

2

Would it be possible to expose the ```efc_pos``` in ```mjx.Data``` as it is normally exposed in ```mjData```? This is a convenient way to check the position error in the constraints,...

This PullRequest contain as well the omniwheel4X

Added the possibility to select qpSWIFT as back-end. There is a strange small bug: when selecting qpSWIFT as back-end, the terminal output is "???" for the qp solver name, but...

Subconstraint implementation

2

I am trying to add this feature in CartesI/O based on the ```sub_constraint``` branch of OpenSoT, however, I am finding some problems understanding the flow during stack construction so I...

Task error not broadcates on rostopic when used as constraint.

1

As title.

As title, this would promote the implementation of addons inside the CartsI/O frameworks as well as not hiding the ones already implemented. I suppose we are not going to have...

In a general (bad) design, the problem description could be hardcoded inside a particular solver implementation. Despite this is not good, anyway it is possible therefore the needs of a...

[CartesianMarker] base_link of CartesianMarker should be fixed in RVIZ

1

This is not really a bug but we figured out that if the ```base_link``` of the task is not the ```fixed_frame``` in RVIZ, the Marker explodes. What we can do...

Local api fix

2

This pull request add the possiblity to set local velocity references for Cartesian task. While this was already possible using the topic by changing the `frame_id` entry of the message...