Page 1 of 1

Residual Actuators in Static Optimization

Posted: Mon Mar 24, 2014 2:39 am
by fantasymew
Hi,

I am performing Joint Reaction (JR) analysis and have some problems. I need some hints or help.

I use the tutorial model Gait2354_Simbody to perform Scaling and IK. Refer to the webinar 'Estimating Joint Loads in OpenSim' by Matt DeMers and the document about Static Optimization (SO), I use IK results as input kinematics. The following problem is that how to set a correct residual actuator file. I compare the differences between webinar example and tutorial. In webinar example, the residual actuator is weakened. There are only 6 actuators: 3 for PointActuator and 3 for TorqueActuator. Besides, the model and motion file seem to have been adjusted. (by RRA?) In tutorial, there is gait2354_RRA_Actuators.xml. Every DOF has an actuator.

My problem is
1) If I use the Gait2354_Simbody model, is it right to use gait2354_RRA_Actuators.xml setting files in SO and JR?
2) According to this article (https://simtk.org/forums/viewtopic.php? ... tion#p7539), the residuals can be estimated by ID results. That is, the parameters of residual actuators depend on IK motion results? If so, how can I estimate the residul actuators from ID results?
3) Which results used as input kinematics of SO is better, IK or RRA?

Thanks for your help

Best regards,
Yu-Wei

Re: Residual Actuators in Static Optimization

Posted: Tue Apr 01, 2014 11:04 am
by jimmy
Hi Yu-Wei,

The actuator files add additional actuators to the model during analysis. RRA is a torque driven analysis, so each coordinate of the model requires and actuator.

In the SO example you are looking at, muscles drive the coordinates and so actuators are not needed about the joints. The SO actuators file adds actuators between the ground and pelvis to account for the residual forces in the model.

Residuals act between the ground and a base segment. The residuals can be viewed by plotting the forces (Fx, Fy, Fz) and the moments (Mx, My, Mz) from ID. The actuators in the actuator file have to be strong enough to deal with whatever the residuals are in your simulation.

These residuals could be brought down by first running RRA and using the adjusted model and kinematics in your SO analysis. As RRA adjusts the kinematics, you will have to judge for yourself if the kinematics are acceptable.

Hope that helps,
-james

Re: Residual Actuators in Static Optimization

Posted: Tue Apr 01, 2014 7:46 pm
by fantasymew
Hi, James,

It helps me a lot. I think I know how to use the residuals and actuators in my analysis.

Thanks very much.

Best regards,
Yu-Wei