Help with Probe for power

[Frédéric Lamon]

Hello,

I am trying to use Probe in order to get the actuators power as well as the joints power. However the actuatorspowerprobe is giving me incoherent results and the internaljointpowerprobe is not giving any outputs. I do not know what to do to solve this. Has anyone already used probes and could give me some advice on this. It would be very much appreciated.

Cheers,

Frederic

[Dario Cazzola]

Hi Frederic,

I did use it and it worked for me. Did you check the value of 'initial condition for integration'?

Cheers

Dario

[jimmy d]

I assume you are using probes with CMC since they won't work with SO.

[Frédéric Lamon]

Hello Dario and James,

Thanks for your answers. I do not use integrate but only value as I am interested in the power and not the work. As I do not use integrate I don't need to run it through CMC right ? But it is not working. I have put some files here if you have time to have a look at it. Thanks a lot.

Fred

[Dario Cazzola]

Hi Fred,

From your probe reporter it seems you calculate the work done (you do use 'integrate') by the actuator (Probe 0), and the actuator force (Probe 2). Probe 1 should calculate the joint internal power, and in your case (knee joint) results 0 ...which seems not to be correct.

I think we need to clarify whether the Joint internal Power is:
-) Joint power computed as the product of net joint torque and corresponding joint angular velocity. This sounds to me as joint EXTERNAL power though.

or

-) The power calculate through the calcPower() function, and therefore representing the dot product of the applied body spatial forces and body spatial velocities, plus the dot product of the applied mobility forces and corresponding mobilities (generalized speeds) u

I have carried out a forward dynamic simulation of a humeral (negative) elevation due to gravity, and plotted the result of a jointinternalpower probe of glenohumeral joint (see attachment). It seems working as the power is increasing linearly as the only force acting on the system is the gravity force.

Any thoughts?

Cheers

Dario

[Frédéric Lamon]

Hello Dario,

Thanks a lot for helping me. So In the files I added before yeah I did let the integrate but it should be changed as value as I do not want to have the work. Sorry about that. If you do,you will see that not all actuators get a power value and I don't understand why, especially some of them are getting too high values as well...

As for the internal power probe, I don't know how it is supposed to be working actually but I would love to be able to get the external joint power and it would be interesting to compare it with the method you described with calcPower function but I don't know how to do that. Would you be able to help me out?

cheers,

Frederic

[Dario Cazzola]

Hi Fred,

Apparently the JointInternalPower Probe with 'integrate' flagged on, is doingexactly wat calcPower does (Have alook to this: https://simtk.org/forums/viewtopic.php? ... wer#p12932).

However calcPower() can be used if you do use API in C++, as I have not tried yet if it is working with Matlab API.

You can always work out the external power as the product of the joint moment from ID and angular velocity from IK (?).

Also, I'd try to calculate the joint power using a probe when the applied force are all known (e.g. only gravity).

Dario

[Frédéric Lamon]

Hello Dario,

Jointinternal power doesn't work with my data and I can't figure out why. As working with the API I have never done it and don't really know how to do it. I have already tried to take the IK results and use the derivative and multiply them with ID results but I wanted to use these probes in order to confirm if my results are coherent or not.

But the most important for me would be to get the actuatorpower probe to work because I do not know any other way of getting muscle power. Do you have any idea as to why it is not working properly in my case ?

Thank you again for taking the time to look at this and for your previous suggestions.

Fred

[jimmy d]

Its not working because the probes use states as inputs. States come from simulations, like CMC, not from static optimization.

[Frédéric Lamon]

Hello James,

Ok, but even if I use the value and not integrate ?

Cheers,

Fred