Unrealistic high knee moments and forces produced from ID
Posted: Mon Dec 04, 2017 6:39 am
Dear All:
I performed inverse dynamics using the lower limb model with implants (see the attached image) for normal gait, based on kinematic motion data and GRF data. The inverse kinematic is normal and correct, which produced good animation of gait motion.
The ankle moments in the output sto file from ID are correct and normal, too, while the knee moments and forces are unrealistic high (even >4000 Nm and > 20000N). The normal ankle moment data shows that it is not caused by wrong length unit. What's more, the implants are welded with the femur or tibia, respectively, forming only one "knee joint", which has the same effect as a "femur-tibia joint".
By the way, patella is also modeled with ligaments, I am not sure whether it may cause the "high moment and force" problem.
Moreover, if there are something wrong with the input files of ID, why the ankle moments are quite correct and normal ? Maybe it is also not the reason.
I also compared the ground reaction force data with the one from gait2392, both the directions of the three axes and the corresponding force values are similar during gait, thus it is sure that the GRF data here is also normal and correct.
The model is from the project data "Grand Challenge Competition to Predict In Vivo Knee Loads", and the output file inverse_dynamics.txt is also attached.
Thanks for anyone who can help to answer,
best regards
Jack
I performed inverse dynamics using the lower limb model with implants (see the attached image) for normal gait, based on kinematic motion data and GRF data. The inverse kinematic is normal and correct, which produced good animation of gait motion.
The ankle moments in the output sto file from ID are correct and normal, too, while the knee moments and forces are unrealistic high (even >4000 Nm and > 20000N). The normal ankle moment data shows that it is not caused by wrong length unit. What's more, the implants are welded with the femur or tibia, respectively, forming only one "knee joint", which has the same effect as a "femur-tibia joint".
By the way, patella is also modeled with ligaments, I am not sure whether it may cause the "high moment and force" problem.
Moreover, if there are something wrong with the input files of ID, why the ankle moments are quite correct and normal ? Maybe it is also not the reason.
I also compared the ground reaction force data with the one from gait2392, both the directions of the three axes and the corresponding force values are similar during gait, thus it is sure that the GRF data here is also normal and correct.
The model is from the project data "Grand Challenge Competition to Predict In Vivo Knee Loads", and the output file inverse_dynamics.txt is also attached.
Thanks for anyone who can help to answer,
best regards
Jack