Jump Landing Inverse Dynamics: Pelvis Ty Issue During Pre-Landing
Posted: Mon Mar 23, 2020 11:51 am
Hello,
I am running Inverse Dynamics on stop-jump landing data as a checkpoint before running RRA. At the time range before the landing where the model has no ground contact, my inverse dynamics results (run with external GRF forces) come out problematic as shown in the image below, where the blue line (my ID result) is supposed to fluctuate about zero.
Red: GRF curve. Blue: ID results (with external forces)
I have previous archival simulation data of a different jump (vertical jump) give correct results despite there also being a time interval where the model is not making contact with the ground shown below. Here the ID results fluctuate about zero, which is the result I'm trying to achieve. However I am struggling to replicate this trend in my stop-jump data.
Hence I know that it is possible to achieve proper simulation results with a model in flight phase of the motion. However I'm not sure what the issue is with my inputs that causes the ID pelvis Ty curve to drift so far down. This will later create an unacceptably large mass change recommendations in RRA. Does anyone have some insight into how to fix this issue or if there is a "common mistakes" reason that causes this to happen for jumping simulations?
Thanks,
Jin
I am running Inverse Dynamics on stop-jump landing data as a checkpoint before running RRA. At the time range before the landing where the model has no ground contact, my inverse dynamics results (run with external GRF forces) come out problematic as shown in the image below, where the blue line (my ID result) is supposed to fluctuate about zero.
Red: GRF curve. Blue: ID results (with external forces)
I have previous archival simulation data of a different jump (vertical jump) give correct results despite there also being a time interval where the model is not making contact with the ground shown below. Here the ID results fluctuate about zero, which is the result I'm trying to achieve. However I am struggling to replicate this trend in my stop-jump data.
Hence I know that it is possible to achieve proper simulation results with a model in flight phase of the motion. However I'm not sure what the issue is with my inputs that causes the ID pelvis Ty curve to drift so far down. This will later create an unacceptably large mass change recommendations in RRA. Does anyone have some insight into how to fix this issue or if there is a "common mistakes" reason that causes this to happen for jumping simulations?
Thanks,
Jin