Is it possible to drive a forward dynamics simulation without muscles for a long time?
I want to drive an arm model only using torques but there is already a previous similar question:
https://simtk.org/forums/viewtopic.php?f=91&t=3624
where the conclusion seems to be that it's not possible to run longer simulations based on torques. I don't really understand how this can be the case in something as simple as the Arm26 model (which I'm currently using). Can someone explain to me what the underlying problem is?
Long forward dynamic simulations without muscles
-
suzunosuke nagaoka
- Posts: 4
- Joined: Tue Dec 09, 2014 4:49 am
Re: Long forward dynamic simulations without muscles
Hi Suzunosuke-
The short answer is yes, in theory you can. The issue with any type of forward simulation is that small errors accumulate over time and will cause the simulation to diverge. We have seen simulations run for 10-20 gait strides but the time to make these was high.
Alot of it comes down to how complicated your system is and how exact the RRA and forward analysis are to one another.
Cheers,
-James
The short answer is yes, in theory you can. The issue with any type of forward simulation is that small errors accumulate over time and will cause the simulation to diverge. We have seen simulations run for 10-20 gait strides but the time to make these was high.
Alot of it comes down to how complicated your system is and how exact the RRA and forward analysis are to one another.
Cheers,
-James