ID from my own kinematics data or Patrick's

[avikde]

• load the yout from the target computer
• digitize and plot vid

[avikde]

Need a calibration stick

dC, dD for SDAB wings

[avikde]

video digitization worked!

As of 37d33d5

[avikde]

ID with Fex_pdep = true

param1 = 2.7679 20.4834 0.60008 4.56377 5.50074

test of traj1, param1:

 2.498e-16  -2.77556e-17   4.16334e-17   0.0        …  5.20417e-17  -3.46945e-18  -6.93889e-18  1.73472e-17
 0.0775701   0.0638868    -0.0167533    -0.0534821     0.0803172     0.0875317     0.0915455    0.0924821

7350

3.17063 20.561 0.289633 4.09398 3.18663

7360

3.29552 17.4416 0.274046 4.29043 3.23492

7370

2.88289 14.3045 0.228285 2.98886 2.31533

7380

0.546576 12.6358 0.23307 0.1 0.334528

Note:

• going away from resonance (drag, force not aligned)
• force at mid stroke exceeding max act force

[avikde]

Use components to debug

7320

7360

[avikde]

Adjusting the fixed params as well. Need to tune these more

• increased ka, lowered ko

This is probably the best result so far. 2.7098 29.39 0.579521 3.59876 3.9613

Components:

Adjust filtering

3.07278 28.8789 0.523188 4.40336 4.31495

Test: 2.22045e-16 0.0 2.77556e-17 0.0 … 1.02349e-16 6.93889e-18 -2.28983e-16 4.57967e-16 0.00338412 0.00135417 -0.00183369 -0.0050024 -0.00994683 -0.0193063 -0.0190643 -0.000671172

Components

Interesting conclusions:

• act = drag except for the second small jump
• looks like that is required for an inertial component in stroke
• mostly from non-coupling component
• ga ~= g (Noah said those have about equal stiffness contributions)

[avikde]

Lowered ko more

3.15685 31.4659 0.453312 4.25663 3.9111

param improvement plot