InducedAcceleration Analysis Tool

[Dimitrios Tsaopoulos]

Hello to everyone,

I was wondering if there is any Opensim documentation regarding the InducedAcceleration Analysis Tool that has been added at the latest version of Opensim! I haven't seen anything about this in the latest Opensim user guide.

Many Thanks,
Dimitris

[Rudolf Klemetti]

See chapter Perturbation!

Rudi

[Ajay Seth]

Rudolf, the Induced Acceleration Analysis is only similar to Perturbation in that it also computes contributions to system accelerations by individual actuators. IAA uses a constraint at the center-of-pressure to compute instantaneous contributions of the ground reaction force due to muscle force. It does not integrate forward in time to estimate accelerations, it solves accelerations and reactions simultaneously. Thus the interface/inputs are different.

[Ayman Habib]

Hi Dimitris,

Induced Acceleration Analysis is not yet publicized/documented because it is still undergoing some user testing. We apologize for creating the confusion. If you would like to be one of our testers and give us feedback we would be glad to provide you with more details. As the functionality gets more mature/robust we'll add documentation to our user guide.

Sorry again,
-Ayman

[Glen Lichtwark]

Ayman and Ajay,

Is there a reason why the perturbation analysis has been taken out of OpenSim 2.2.0? I used to run this from the command line, but no longer can in the latest version.

Is Induced Acceleration analysis replacing Perturb? I am having problems using the Induced Acceleration analysis, as there seems to be a problem in reading the MOT file for the kinetics. Is this because of a change in the requirements of the file types for this analysis (i.e. using the XML file instead of the MOT file)?

I realise this is still being tested, but any help would be appreciated.

Thanks,
Glen

[Tim Dorn]

Hi Glen,

I thought I might let you know that I have released my own implementation of a novel induced acceleration algorithm as an OpenSim plugin. It uses multiple contact locations and a set of "rules" based on the measured center of pressure location to determine the kinematic constraints present between the foot and the ground at each time instant. I have performed it on all my running and walking data and it reproduces results from the literature nicely. You can try it out if you want at https://simtk.org/home/tims_plugins
It is very easy to setup (thanks to the extension capabilities of OpenSim) and comes with a full user manual and a fully worked example of walking.

Tim

[Glen Lichtwark]

Thanks Tim,

I'll have a look and see if I can get it up and running.

I'd still be interested to find out about the future of perturb vs induced acceleration analysis tools in OpenSim.

Glen

[Sam Hamner]

Hi Glen,

Sorry for the inconvenience, but some recent changes to the underlying simTK code, which made improvements to the integrators in OpenSim, also "broke" the Perturb tool in 2.2.0. These improvements make sure the integrators are accurate and numerically sound, and we are working on the Perturb code to make it compliant with these changes. We hope to have Perturb fixed soon, but for now if you want to use the Perturbation analysis you will have to stick with an older version of OpenSim.

However, after OpenSim 2.0, we have been shifting toward more testing and support of the Induced Acceleration Analysis (IAA) instead of Perturb. While, IAA is still technically in "beta testing", as we have no formal documentation or support, it has been tested with published results. IAA has been giving us better results (i.e., better superposition) for determining muscle contributions to mass center and joint accelerations for both walking and running, and is more than 50X faster than Perturb.

If you'd like to use IAA as a beta tester, we can give you some support.

Hope this clears things up.

Best,
Sam