I derived a non linear differential equation system for a mechanical system. However I doubt the correctness of it. So I compared the model's dynamics with a VREP simulation. I did this by by Euler integrating the model with the same step size as in the simulation, and then comparing the state variable from the simulation and the model.

I wonder if this is a valid approach, and if there are other ways of doing it

  • $\begingroup$ Depending on how good your model is and your time step it could be that Euler integration is the biggest inaccuracy. So keep that in mind as well. $\endgroup$ – fibonatic Jun 27 '19 at 15:48

Verify your model against actual physical hardware. If you're sure of parts of the model, and unsure of others, then just verify the parts you're worried about. If the physical hardware is big and expensive, then if possible build a simplified physical model that still captures the behavior you're concerned about.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.