Ball on Plate Control is not like expected

I'm implementing a ball on plate system with simscape. i had some "normal" troubles modelling the system but after searching i could finally bring the joints in a good position where there DoF are not violating each other. When i simulate the motor control in a simple model it works fine, but now when i simulate the whole system the control mechanism works to slow and the simulating time is very slow. Changing the parameters of the PID seems not to be helpful. I thought this could be the DoF of the Joints but when i measure a misplacement its close to zero (~1e-15).

Does anyone have a different idea for a solution?

File is attached

Thank you!

[Sam Chak]

Not sure what went wrong. Most likely it is due to the Simscape model and your previous Mathematical model are different. In other words, even though the PID law is succussfully designed to stabilize the previous model, it does not guarantee the stability of the Simscape model. Can you double-check if both models are mathematically the same?

[Jonas Winkler]

thank you for your input, i had adapted the parameters for the model. maybe i dont get your idea.

of course they are different mathematically. One is one dimensional and the other is two dimensional, but the control for the axis is adapted on the dimensions of the axis.

Answer 1

[Jonas Winkler]

The main challenge is a physical challenge for the simulation. the joints had struggle with the camp forces. switching the revolute joint to a spherical joint is the solution.