motions have been recorded by Xsens MVN Awinda inertial motion capture system with 17 wireless motion trackers and an update rate set to 60 Hz.
I am interested in estimating the Ground Reaction Force out of the BVH files. Once the model analysis succeeded I have looked on the GRF predicted data (h5 file: Output/EnvironmentModel/ForcePlates/GRF_Prediction_/F) of the left and right foot in x,y and z-axis. I have found that there are not as many predicted data as expected.
I have set the FirstFrame=2, LastFrame= (.BVHFileData.Header.FramesNo - 1) and nStep = LastFrame - FirstFrame + 1. Therefore I have expected to see n = (FramesNo - 2) predicted data. However, I got n = (FramesNo - 8) many data.
I assumed there will be a GRF prediction for each frame which is included in the analysis. Thus, the number of predicted data should equal the number of nStep. Is this assumption correct?
The model will automatically cut away some frames in the beginning and end of the inverse analysis.
The reason is that the BSplines used to create the joint drivers, needs a few frames to “settle in”. It is possible to do the analysis in all frames but you may see unrealistic accelerations.
Thanks for your reply.
Can you tell me how many frames the joint drivers needs to settle in? The reason I am asking is that I am interested in extracting the GRF per each frame. For my analysis I can cut some frames from the beginning/end, but I like to determine the corresponding frame to the resulted GRF. Is that possible?
As i recall we subtract 4 frames beginning and end, it is done automatically in the inverse dynamic analysis, if you click the tStart and tEnd in the ModelTree for the inverse analysis you can find the definitions on this.
Note that it will not help to change anything in kinematic studies, the inverse analysis will need to always run with fewer frames than in kinematics.
When you run inverse analysis it will calculate GRF for all frames except the few frames which was cut out in start and end.