Page 1 of 1

Failure for forward dynamics during gait simulation

Posted: Wed Oct 14, 2020 9:26 am
by xhua
Dear OpenSim experts:

I am performing a gait simulation for normal walking. I have finished the scaling, inverse kinematics, RRA and CMC successfully. But when I plan to perform forward dynamics simulation based on CMC, I got a problem, the forward simulation was diverged from the expected trajectory at the end of gait (the simulation is normal initially), as shown in the attached figures (Figure 1 &2). Could you please tell me why this could happen and how to solve this problem? I have attached the CMC setup file and Forward setup file for your references as well. If you need more information, please let me know. Thanks.

Re: Failure for forward dynamics during gait simulation

Posted: Wed Oct 14, 2020 2:50 pm
by tkuchida
This behaviour is expected: you cannot compute control forces using CMC and then expect the model to follow exactly the same trajectory when these forces are applied in a forward dynamic simulation. First of all, you would also need to apply the forces/torques generated by the residual and reserve actuators. Even if you did this, however, any small numerical differences (e.g., due to different time step sizes when integrating) would cause the model to deviate from the trajectory that CMC tracked.

Re: Failure for forward dynamics during gait simulation

Posted: Wed Oct 14, 2020 3:49 pm
by xhua
Hi, Thomas

Thank you very much for your kindly reply.

(1) First of all, could you please tell me more details about how to apply the forces/torques generated by the residual and reserve actuators? e.g. for the residual forces/torques, are they the ones generated in the RRA and should I add them as external load together with the ground reaction forces? For the forces/torques generated by the reserve actuators, where should I find them? are they generated in the RRA step or CMC step? Should I add them through the "control" in the Input panel?

(2) As you mentioned if this behavior is expected, then how should I make sure the Forward results are reasonable and correct?

(3) I think gait simulation is a very common general simulation in Biomechanics, but I didn't find a manual or guidance that specify the methodology/pipeline of this kind of simulation. I am wondering could you please provide me any suggestions about where I can learn the whole processes of the gait simulation, particularly including the forward simulation (add forces/torques generated by the residual and reserve actuators)? Any examples in the online manual or YouTube?

Thank you very much in advance. Your kindly reply will be highly appreciated.

Xijin

Re: Failure for forward dynamics during gait simulation

Posted: Sat Sep 18, 2021 4:22 am
by hesamb1997
Dear Xijin
Did you succeed in performing your simulation? if you did, i was wondering if you tell me how did you use reserve actuators or doing something which helped you