How to Support Robotic Manipulation Simulation

[matheecs]

Thank you for your open-source contribution to IPC simulation.

I would like to apply IPC to the field of robotic manipulation and add actuator constraints, such as a PD controller. The RotatingMotor example provided in the repository is great — may I ask how it can be modified to support a torque-driven motor?

[MuGdxy]

As you can see in the tutorial-animation.
The RotatingMotor in libuipc is a specified SoftTransformConstraint, which converts a "constant angular velocity with constant power" motor to a general constraint.

If you want to implement a torque-driven motor, you need some knowledge to convert such a motor into a general constraint. (I will add the information of the SoftTransformConstraint to the document later).

BUT, I think at this time, the SoftTransformConstraint may not fully meet your demand, and we are trying to add handy constraints to simulate the motor more accurately

Please leave some formulas you wanna use to simulate the torque-driven motor here, and see if we can put it into the simulator.

[matheecs]

Currently, the commonly used actuators in robots adopt PD control, and the torque output of the motor is

this formulas coming from https://xbpeng.github.io/projects/ActionSpace/ActionSpace_2017.pdf

[MuGdxy]

Thank you for your attention. We have implemented an ExternalArticulationConstraint constitution to incorporate any possible joint drivers. You can use your favorite articulation solver as input to get a strong coupling with IPC. More detailed documentation is available here, and a playable GUI example is here.