Quantify the effect of fewer or degraded IMUs on accuracy of joint torque trajectories

Step 1: Generate the most accurate joint torque trajectories

• mocap treadmill data
• create virtual markers in the symbrim model
• scale symbrim model to fit the geometric size and inertia of the cyclist
• solve the open loop tracking controls problem with opty
• use joint torque trajectories as the "gold standard"

Step 2: Generate joint torque trajectories from IMU tracking

• create virtual IMUs on the body segments and bicycle in the symbrim model
• calculate the fake IMU measurements from the mocap data
• now solve the open loop tracking control problem with opty
• compare the new joint torques against the joint torques from step 1

Step 3: Explore degrading the IMUs and the effect on joint torque trajectories

Step 4: Find optimal location of IMUs

Start by tracing how the mocap treadmill data, symbrim model, virtual markers, and opty open-loop tracking controls problem are represented in the repository. Reproduce the gold-standard and IMU-tracking trajectories first; the work is done when degraded IMU configurations are compared quantitatively and an optimal IMU location is identified.