OPTIMIZATION FAILED

Provide easy-to-use, extensible software for modeling, simulating, controlling, and analyzing the neuromusculoskeletal system.
POST REPLY
User avatar
Max Rusch
Posts: 18
Joined: Tue Dec 15, 2009 11:22 am

OPTIMIZATION FAILED

Post by Max Rusch » Thu Nov 25, 2010 8:06 am

Hello,

I'm having problems with static optimization tool. I do not understand the significance of these errors.

Can anyone help me with this problem?

Regards, Max


Example for one time(this problem occur in several time):

time = 0.233333 Performance =9.23433 Constraint violation = 635.744
SimTK Exception thrown at InteriorPointOptimizer.cpp:249:
Optimizer failed: Ipopt: Infeasible problem detected (status 2)
OPTIMIZATION FAILED...

StaticOptimization.record: WARN- The optimizer could not find a solution at time = 0.25

The model appears unsuitable for static optimization.
Try appending the model with additional force(s) or locking joint(s) to reduce the following acceleration constraint violation(s):
elv_angle: constraint violation = 0.0532864
shoulder_elv: constraint violation = -0.00109953
shoulder_rot: constraint violation = 0.0332366
elbow_flexion: constraint violation = 0.575879
pro_sup: constraint violation = -95.9118
deviation: constraint violation = -9.1774
flexion: constraint violation = 0.173407
cmc_flexion: constraint violation = 8.5887
cmc_abduction: constraint violation = -51.8523
mp_flexion: constraint violation = -0.168305
ip_flexion: constraint violation = 7.49353
2mcp_flexion: constraint violation = 55.9373
2mcp_abduction: constraint violation = 2.69856
2pm_flexion: constraint violation = -5.37656
2md_flexion: constraint violation = -1338.59

User avatar
Nur Adila Faruk Senan
Posts: 75
Joined: Tue Apr 06, 2010 8:20 pm

RE: OPTIMIZATION FAILED

Post by Nur Adila Faruk Senan » Thu Nov 25, 2010 10:52 am

Hey Max,

I haven't done any work on the arm model, but in terms of the StaticOptimization errors, I *thinnk* this is what "Constraint Violation" and "Performance" values mean:

F - ma = Constraint Violation (where the 'F' and 'ma' are coordinate specific, and 'a' is the acceleration)
and
sum_{all muscles}(activation^p) = Performance.

So, it doesn't really matter how large or small the 'Performance' value is, but the 'Constraint Violation' number should be as close to 0 as possible for the results to be valid.

I don't know what the rest of your error messages mean though, but either adding more muscles/making them stronger or slowing down the motion might work.

hope this helps!
adila

User avatar
Max Rusch
Posts: 18
Joined: Tue Dec 15, 2009 11:22 am

RE: OPTIMIZATION FAILED

Post by Max Rusch » Thu Nov 25, 2010 11:51 am

Hi Adila,

Thanks for the advice! But I'm using the upper limb model and I can't add more muscles.

And, when I tried to make the muscles stronger (modifying the max isometric force) the results that I found were erroneous, varying without a plausible behavior.

Any more suggestions?

Thx

User avatar
Nur Adila Faruk Senan
Posts: 75
Joined: Tue Apr 06, 2010 8:20 pm

RE: OPTIMIZATION FAILED

Post by Nur Adila Faruk Senan » Fri Nov 26, 2010 7:50 pm

You can add more muscles if you want to by editing the model in Notepad++ (or some such text-editing software), although I'm not sure if you need to for this model.

Otherwise, sorry, but I don't really have any other suggestions. I just figured that it might help for you to know what the "Performance" and "Constraint Violation" terms meant. Ayman would probably know more about how to help you I'm sure.

Good luck!
adila

POST REPLY