Skip to content

Instantly share code, notes, and snippets.

@tomas789

tomas789/ios_recording_101_fully_synced.txt

Created June 5, 2018 12:54
Show Gist options
  • Save tomas789/9566943508c68bf779df7411d8993d9d to your computer and use it in GitHub Desktop.
Save tomas789/9566943508c68bf779df7411d8993d9d to your computer and use it in GitHub Desktop.
Download ZIP
Raw
ios_recording_101_fully_synced.txt
(env) tomaskrejci @ Tomass-MBP-2.RD ➜ pymsckf git:(replay-from-app) ✗ docker run -v ~/ios_recordings:/data -v ~/Developer/pymsckf/examples:/examples -v `pwd`/calibration_output:/output --security-opt seccomp:unconfined -it -e DISPLAY=docker.for.mac.localhost:0 -w /output pymsckf_calibration kalibr_calibrate_imu_camera --bag /data/ios_recording_101.bag --cam /output/camchain-dataios_recording_96.yaml --imu /examples/imu_vins_mobile.yaml --target /examples/april_6x6_50x50cm.yaml
importing libraries
Initializing IMUs:
Update rate: 100.0
Accelerometer:
Noise density: 0.02
Noise density (discrete): 0.2
Random walk: 0.002
Gyroscope:
Noise density: 0.02
Noise density (discrete): 0.2
Random walk: 4e-05
Initializing imu rosbag dataset reader:
Dataset: /data/ios_recording_101.bag
Topic: /imu0
Number of messages: 5188
Reading IMU data (/imu0)
Read 5188 imu readings over 45.1 seconds
Initializing calibration target:
Type: aprilgrid
Tags:
Rows: 6
Cols: 6
Size: 0.055 [m]
Spacing 0.0165 [m]
Initializing camera chain:
Camera chain - cam0:
Camera model: pinhole
Focal length: [525.103347243366, 525.8822399972671]
Principal point: [325.9919034516905, 229.5969467482677]
Distortion model: radtan
Distortion coefficients: [0.03869682944949094, -0.05888424451155317, 0.0015779459175340209, 0.002600150278649132]
baseline: no data available
Initializing camera rosbag dataset reader:
Dataset: /data/ios_recording_101.bag
Topic: /cam0/image_raw
Number of images: 677
Extracting calibration target corners
Extracted corners for 677 images (of 677 images)
Building the problem
Spline order: 6
Pose knots per second: 70
Do pose motion regularization: False
xddot translation variance: 1000000.000000
xddot rotation variance: 100000.000000
Bias knots per second: 50
Do bias motion regularization: True
Blake-Zisserman on reprojection errors -1
Acceleration Huber width (m/s^2): -1.000000
Gyroscope Huber width (rad/s): -1.000000
Do time calibration: False
Max iterations: 30
Time offset padding: 0.020000
Estimating imu-camera rotation prior
Initializing a pose spline with 4507 knots (100.000000 knots per second over 45.065722 seconds)
Orientation prior camera-imu found as: (T_i_c)
[[-0.0040329 -0.99988416 0.01467689]
[-0.99998005 0.00396106 -0.00492009]
[ 0.00486139 -0.01469644 -0.99988018]]
Gyro bias prior found as: (b_gyro)
[ 0.00893672 -0.01446461 0.00538593]
Initializing a pose spline with 4515 knots (100.000000 knots per second over 45.145722 seconds)
Initializing the bias splines with 2257 knots
Adding camera error terms (/cam0/image_raw)
Added 677 camera error terms
Adding accelerometer error terms (/imu0)
Added 5187 of 5188 accelerometer error terms (skipped 1 out-of-bounds measurements)
Adding gyroscope error terms (/imu0)
Added 5187 of 5188 gyroscope error terms (skipped 1 out-of-bounds measurements)
Before Optimization
===================
Reprojection error squarred (cam0): mean 1.70529412366, median 0.912447298991, std: 2.5498727068
Gyro error squarred (imu0): mean 0.221551728312, median 0.115731120015, std: 0.372322028182
Accelerometer error squarred (imu0): mean 6.39305159256, median 4.95782526515, std: 5.29565971168
Transformation T_cam0_imu0 (imu0 to cam0, T_ci): [m]
[[-0.0040329 -0.99988416 0.01467689 0. ]
[-0.99998005 0.00396106 -0.00492009 0. ]
[ 0.00486139 -0.01469644 -0.99988018 0. ]
[ 0. 0. 0. 1. ]]
Optimizing...
No linear system solver set in the options. Defaulting to the sparse_cholesky solver
Using the sparse_cholesky linear system solver
Using the levenberg_marquardt trust region policy
No linear system solver set in the options. Defaulting to the sparse_cholesky solver
Using the sparse_cholesky linear system solver
Using the levenberg_marquardt trust region policy
Initializing
Optimization problem initialized with 9047 design variables and 101470 error terms
The Jacobian matrix is 285538 x 40700
[0.0]: J: 181958
[1]: J: 24010.1, dJ: 157948, deltaX: 0.0333428, LM - lambda:100 mu:2
[2]: J: 23796.1, dJ: 213.914, deltaX: 0.00372244, LM - lambda:33.3333 mu:2
[3]: J: 23239.1, dJ: 557.077, deltaX: 0.018222, LM - lambda:11.1111 mu:2
[4]: J: 22411.7, dJ: 827.329, deltaX: 0.0605426, LM - lambda:3.7037 mu:2
[5]: J: 22253.8, dJ: 157.932, deltaX: 0.0834936, LM - lambda:1.23457 mu:2
[6]: J: 22240.7, dJ: 13.0565, deltaX: 0.0674512, LM - lambda:0.419468 mu:2
[7]: J: 22240.1, dJ: 0.636288, deltaX: 0.0101223, LM - lambda:0.418653 mu:2
After Optimization (Results)
==================
Reprojection error squarred (cam0): mean 0.248825548423, median 0.150949634842, std: 0.306909362012
Gyro error squarred (imu0): mean 0.0160276449722, median 0.00808554677528, std: 0.027410174141
Accelerometer error squarred (imu0): mean 0.116370646854, median 0.0660367726834, std: 0.167425458322
Transformation T_cam0_imu0 (imu0 to cam0, T_ci): [m]
[[-0.00022895 -0.99992513 0.01223429 0.10441509]
[-0.99999959 0.00023969 0.00087673 0.02615085]
[-0.0008796 -0.01223408 -0.99992477 0.00300713]
[ 0. 0. 0. 1. ]]
Saving calibration to file: camchain-imucam-dataios_recording_101.yaml
Detailed results written to file: results-imucam-dataios_recording_101.txt
Raw
ios_recording_101_unsynced.txt
(env) tomaskrejci @ Tomass-MBP-2.RD ➜ pymsckf git:(replay-from-app) ✗ docker run -v ~/ios_recordings:/data -v ~/Developer/pymsckf/examples:/examples -v `pwd`/calibration_output:/output --security-opt seccomp:unconfined -it -e DISPLAY=docker.for.mac.localhost:0 -w /output pymsckf_calibration kalibr_calibrate_imu_camera --bag /data/ios_recording_101.bag --cam /output/camchain-dataios_recording_96.yaml --imu /examples/imu_vins_mobile.yaml --target /examples/april_6x6_50x50cm.yaml --time-calibration
importing libraries
Initializing IMUs:
Update rate: 100.0
Accelerometer:
Noise density: 0.02
Noise density (discrete): 0.2
Random walk: 0.002
Gyroscope:
Noise density: 0.02
Noise density (discrete): 0.2
Random walk: 4e-05
Initializing imu rosbag dataset reader:
Dataset: /data/ios_recording_101.bag
Topic: /imu0
Number of messages: 5187
Reading IMU data (/imu0)
Read 5187 imu readings over 45.1 seconds
Initializing calibration target:
Type: aprilgrid
Tags:
Rows: 6
Cols: 6
Size: 0.055 [m]
Spacing 0.0165 [m]
Initializing camera chain:
Camera chain - cam0:
Camera model: pinhole
Focal length: [525.103347243366, 525.8822399972671]
Principal point: [325.9919034516905, 229.5969467482677]
Distortion model: radtan
Distortion coefficients: [0.03869682944949094, -0.05888424451155317, 0.0015779459175340209, 0.002600150278649132]
baseline: no data available
Initializing camera rosbag dataset reader:
Dataset: /data/ios_recording_101.bag
Topic: /cam0/image_raw
Number of images: 676
Extracting calibration target corners
Extracted corners for 676 images (of 676 images)
Building the problem
Spline order: 6
Pose knots per second: 70
Do pose motion regularization: False
xddot translation variance: 1000000.000000
xddot rotation variance: 100000.000000
Bias knots per second: 50
Do bias motion regularization: True
Blake-Zisserman on reprojection errors -1
Acceleration Huber width (m/s^2): -1.000000
Gyroscope Huber width (rad/s): -1.000000
Do time calibration: True
Max iterations: 30
Time offset padding: 0.020000
Estimating time shift camera to imu:
Initializing a pose spline with 4500 knots (100.000000 knots per second over 44.999057 seconds)
Time shift camera to imu (t_imu = t_cam + shift):
0.0695765055256
Estimating imu-camera rotation prior
Initializing a pose spline with 4500 knots (100.000000 knots per second over 44.999057 seconds)
Orientation prior camera-imu found as: (T_i_c)
[[-0.00480984 -0.99987879 0.01480791]
[-0.99998105 0.00475237 -0.00391377]
[ 0.00384293 -0.01482645 -0.9998827 ]]
Gyro bias prior found as: (b_gyro)
[ 0.00899513 -0.01445784 0.00534239]
Initializing a pose spline with 4508 knots (100.000000 knots per second over 45.079057 seconds)
Initializing the bias splines with 2254 knots
Adding camera error terms (/cam0/image_raw)
Added 676 camera error terms
Adding accelerometer error terms (/imu0)
Added 5179 of 5187 accelerometer error terms (skipped 8 out-of-bounds measurements)
Adding gyroscope error terms (/imu0)
Added 5179 of 5187 gyroscope error terms (skipped 8 out-of-bounds measurements)
Before Optimization
===================
Reprojection error squarred (cam0): mean 1.70766155772, median 0.915104613398, std: 2.55123031374
Gyro error squarred (imu0): mean 0.219135950583, median 0.115051742038, std: 0.358662354413
Accelerometer error squarred (imu0): mean 6.68353460252, median 5.18642047839, std: 5.42136561445
Transformation T_cam0_imu0 (imu0 to cam0, T_ci): [m]
[[-0.00480984 -0.99987879 0.01480791 0. ]
[-0.99998105 0.00475237 -0.00391377 0. ]
[ 0.00384293 -0.01482645 -0.9998827 0. ]
[ 0. 0. 0. 1. ]]
cam0 to imu0 time: [s] (t_imu = t_cam + shift)
0.0695765055256
Optimizing...
No linear system solver set in the options. Defaulting to the sparse_cholesky solver
Using the sparse_cholesky linear system solver
Using the levenberg_marquardt trust region policy
No linear system solver set in the options. Defaulting to the sparse_cholesky solver
Using the sparse_cholesky linear system solver
Using the levenberg_marquardt trust region policy
Initializing
Optimization problem initialized with 9035 design variables and 101310 error terms
The Jacobian matrix is 285106 x 40641
[0.0]: J: 183366
[1]: J: 25139.8, dJ: 158226, deltaX: 0.0341231, LM - lambda:100 mu:2
[2]: J: 24961.6, dJ: 178.183, deltaX: 0.00633846, LM - lambda:33.3333 mu:2
[3]: J: 23717.4, dJ: 1244.21, deltaX: 0.0105701, LM - lambda:45.4531 mu:2
[4]: J: 23388.1, dJ: 329.368, deltaX: 0.0596146, LM - lambda:15.151 mu:2
[5]: J: 22557.8, dJ: 830.279, deltaX: 0.135323, LM - lambda:5.05034 mu:2
[6]: J: 22254.5, dJ: 303.306, deltaX: 0.105529, LM - lambda:1.68345 mu:2
[7]: J: 22196.7, dJ: 57.8113, deltaX: 0.0499514, LM - lambda:0.753922 mu:2
[8]: J: 22192.6, dJ: 4.0688, deltaX: 0.0590345, LM - lambda:0.582168 mu:2
[9]: J: 22187.2, dJ: 5.437, deltaX: 0.043353, LM - lambda:0.591306 mu:2
[10]: J: 22186.1, dJ: 1.0243, deltaX: 0.0302312, LM - lambda:0.571041 mu:2
Last step was a regression. Reverting
[11]: J: 22187, dJ: -0.844424, deltaX: 0.0279273, LM - lambda:0.750203 mu:2
After Optimization (Results)
==================
Reprojection error squarred (cam0): mean 0.248750981345, median 0.150705830879, std: 0.307010183646
Gyro error squarred (imu0): mean 0.0176584257005, median 0.00916696795625, std: 0.029155108086
Accelerometer error squarred (imu0): mean 0.112474968883, median 0.0598538555048, std: 0.168916849965
Transformation T_cam0_imu0 (imu0 to cam0, T_ci): [m]
[[ 0.00058265 -0.99992686 0.01208053 0.10450102]
[-0.9999998 -0.00057956 0.00025899 0.02608905]
[-0.00025197 -0.01208067 -0.99992699 0.0034942 ]
[ 0. 0. 0. 1. ]]
cam0 to imu0 time: [s] (t_imu = t_cam + shift)
0.0810945370552
Saving calibration to file: camchain-imucam-dataios_recording_101.yaml
Detailed results written to file: results-imucam-dataios_recording_101.txt
Raw
ios_recording_102_fully_synced.txt
(env) tomaskrejci @ Tomass-MBP-2.RD ➜ pymsckf git:(replay-from-app) ✗ docker run -v ~/ios_recordings:/data -v ~/Developer/pymsckf/examples:/examples -v `pwd`/calibration_output:/output --security-opt seccomp:unconfined -it -e DISPLAY=docker.for.mac.localhost:0 -w /output pymsckf_calibration kalibr_calibrate_imu_camera --bag /data/ios_recording_102.bag --cam /output/camchain-dataios_recording_96.yaml --imu /examples/imu_vins_mobile.yaml --target /examples/april_6x6_50x50cm.yaml
importing libraries
Initializing IMUs:
Update rate: 100.0
Accelerometer:
Noise density: 0.02
Noise density (discrete): 0.2
Random walk: 0.002
Gyroscope:
Noise density: 0.02
Noise density (discrete): 0.2
Random walk: 4e-05
Initializing imu rosbag dataset reader:
Dataset: /data/ios_recording_102.bag
Topic: /imu0
Number of messages: 4185
Reading IMU data (/imu0)
Read 4185 imu readings over 36.4 seconds
Initializing calibration target:
Type: aprilgrid
Tags:
Rows: 6
Cols: 6
Size: 0.055 [m]
Spacing 0.0165 [m]
Initializing camera chain:
Camera chain - cam0:
Camera model: pinhole
Focal length: [525.103347243366, 525.8822399972671]
Principal point: [325.9919034516905, 229.5969467482677]
Distortion model: radtan
Distortion coefficients: [0.03869682944949094, -0.05888424451155317, 0.0015779459175340209, 0.002600150278649132]
baseline: no data available
Initializing camera rosbag dataset reader:
Dataset: /data/ios_recording_102.bag
Topic: /cam0/image_raw
Number of images: 546
Extracting calibration target corners
Extracted corners for 546 images (of 546 images)
Building the problem
Spline order: 6
Pose knots per second: 70
Do pose motion regularization: False
xddot translation variance: 1000000.000000
xddot rotation variance: 100000.000000
Bias knots per second: 50
Do bias motion regularization: True
Blake-Zisserman on reprojection errors -1
Acceleration Huber width (m/s^2): -1.000000
Gyroscope Huber width (rad/s): -1.000000
Do time calibration: False
Max iterations: 30
Time offset padding: 0.020000
Estimating imu-camera rotation prior
Initializing a pose spline with 3633 knots (100.000000 knots per second over 36.332572 seconds)
Orientation prior camera-imu found as: (T_i_c)
[[ 0.00046006 -0.99992687 0.0120852 ]
[-0.99998242 -0.00053147 -0.00590591]
[ 0.0059119 -0.01208227 -0.99990953]]
Gyro bias prior found as: (b_gyro)
[ 0.0095835 -0.01495016 0.0056909 ]
Initializing a pose spline with 3641 knots (100.000000 knots per second over 36.412572 seconds)
Initializing the bias splines with 1821 knots
Adding camera error terms (/cam0/image_raw)
Added 546 camera error terms
Adding accelerometer error terms (/imu0)
Added 4182 of 4185 accelerometer error terms (skipped 3 out-of-bounds measurements)
Adding gyroscope error terms (/imu0)
Added 4182 of 4185 gyroscope error terms (skipped 3 out-of-bounds measurements)
Before Optimization
===================
Reprojection error squarred (cam0): mean 1.65924658965, median 0.921154917454, std: 2.45708164212
Gyro error squarred (imu0): mean 0.241320514383, median 0.127339801792, std: 0.414189309759
Accelerometer error squarred (imu0): mean 5.93976698719, median 4.42535842562, std: 5.91501462344
Transformation T_cam0_imu0 (imu0 to cam0, T_ci): [m]
[[ 0.00046006 -0.99992687 0.0120852 0. ]
[-0.99998242 -0.00053147 -0.00590591 0. ]
[ 0.0059119 -0.01208227 -0.99990953 0. ]
[ 0. 0. 0. 1. ]]
Optimizing...
No linear system solver set in the options. Defaulting to the sparse_cholesky solver
Using the sparse_cholesky linear system solver
Using the levenberg_marquardt trust region policy
No linear system solver set in the options. Defaulting to the sparse_cholesky solver
Using the sparse_cholesky linear system solver
Using the levenberg_marquardt trust region policy
Initializing
Optimization problem initialized with 7301 design variables and 86264 error terms
The Jacobian matrix is 239164 x 32840
[0.0]: J: 149062
[1]: J: 16346.2, dJ: 132715, deltaX: 0.0297786, LM - lambda:100 mu:2
[2]: J: 16208.4, dJ: 137.752, deltaX: 0.00624748, LM - lambda:33.3333 mu:2
[3]: J: 15845.6, dJ: 362.835, deltaX: 0.0458342, LM - lambda:11.1111 mu:2
[4]: J: 15239.3, dJ: 606.301, deltaX: 0.157083, LM - lambda:3.7037 mu:2
[5]: J: 15146.3, dJ: 92.9365, deltaX: 0.668176, LM - lambda:1.23457 mu:2
[6]: J: 15144.1, dJ: 2.1945, deltaX: 0.681681, LM - lambda:0.468851 mu:2
[7]: J: 15142.2, dJ: 1.9362, deltaX: 0.0945981, LM - lambda:0.588911 mu:2
[8]: J: 15135, dJ: 7.23624, deltaX: 0.0201986, LM - lambda:0.545686 mu:2
[9]: J: 15133.4, dJ: 1.52289, deltaX: 0.00537899, LM - lambda:0.181895 mu:2
Last step was a regression. Reverting
[10]: J: 15138.3, dJ: -4.89625, deltaX: 0.0023225, LM - lambda:0.0828562 mu:2
Last step was a regression. Reverting
[11]: J: 15138.3, dJ: -4.89583, deltaX: 0.00211387, LM - lambda:0.331425 mu:4
Last step was a regression. Reverting
[12]: J: 15138.4, dJ: -4.93313, deltaX: 0.00125178, LM - lambda:2.6514 mu:8
Last step was a regression. Reverting
[13]: J: 15139.1, dJ: -5.60636, deltaX: 0.000553604, LM - lambda:42.4224 mu:16
Last step was a regression. Reverting
[14]: J: 15139.1, dJ: -5.61388, deltaX: 9.41989e-06, LM - lambda:1357.52 mu:32
After Optimization (Results)
==================
Reprojection error squarred (cam0): mean 0.197765126371, median 0.129984080717, std: 0.220535061483
Gyro error squarred (imu0): mean 0.0145064758824, median 0.00659687544915, std: 0.0335628604514
Accelerometer error squarred (imu0): mean 0.0903179653003, median 0.0520418939953, std: 0.109019641716
Transformation T_cam0_imu0 (imu0 to cam0, T_ci): [m]
[[ 0.00156404 -0.99986058 0.01662455 0.09874269]
[-0.99998917 -0.00149095 0.00440814 0.02705961]
[-0.00438274 -0.01663126 -0.99985209 -0.00631493]
[ 0. 0. 0. 1. ]]
Saving calibration to file: camchain-imucam-dataios_recording_102.yaml
Detailed results written to file: results-imucam-dataios_recording_102.txt
Raw
ios_recording_102_unsynced.txt
(env) tomaskrejci @ Tomass-MBP-2.RD ➜ pymsckf git:(replay-from-app) ✗ docker run -v ~/ios_recordings:/data -v ~/Developer/pymsckf/examples:/examples -v `pwd`/calibration_output:/output --security-opt seccomp:unconfined -it -e DISPLAY=docker.for.mac.localhost:0 -w /output pymsckf_calibration kalibr_calibrate_imu_camera --bag /data/ios_recording_102.bag --cam /output/camchain-dataios_recording_96.yaml --imu /examples/imu_vins_mobile.yaml --target /examples/april_6x6_50x50cm.yaml --time-calibration
importing libraries
Initializing IMUs:
Update rate: 100.0
Accelerometer:
Noise density: 0.02
Noise density (discrete): 0.2
Random walk: 0.002
Gyroscope:
Noise density: 0.02
Noise density (discrete): 0.2
Random walk: 4e-05
Initializing imu rosbag dataset reader:
Dataset: /data/ios_recording_102.bag
Topic: /imu0
Number of messages: 4186
Reading IMU data (/imu0)
Read 4186 imu readings over 36.4 seconds
Initializing calibration target:
Type: aprilgrid
Tags:
Rows: 6
Cols: 6
Size: 0.055 [m]
Spacing 0.0165 [m]
Initializing camera chain:
Camera chain - cam0:
Camera model: pinhole
Focal length: [525.103347243366, 525.8822399972671]
Principal point: [325.9919034516905, 229.5969467482677]
Distortion model: radtan
Distortion coefficients: [0.03869682944949094, -0.05888424451155317, 0.0015779459175340209, 0.002600150278649132]
baseline: no data available
Initializing camera rosbag dataset reader:
Dataset: /data/ios_recording_102.bag
Topic: /cam0/image_raw
Number of images: 546
Extracting calibration target corners
Extracted corners for 546 images (of 546 images)
Building the problem
Spline order: 6
Pose knots per second: 70
Do pose motion regularization: False
xddot translation variance: 1000000.000000
xddot rotation variance: 100000.000000
Bias knots per second: 50
Do bias motion regularization: True
Blake-Zisserman on reprojection errors -1
Acceleration Huber width (m/s^2): -1.000000
Gyroscope Huber width (rad/s): -1.000000
Do time calibration: True
Max iterations: 30
Time offset padding: 0.020000
Estimating time shift camera to imu:
Initializing a pose spline with 3633 knots (100.000000 knots per second over 36.332572 seconds)
Time shift camera to imu (t_imu = t_cam + shift):
0.0782779288516
Estimating imu-camera rotation prior
Initializing a pose spline with 3633 knots (100.000000 knots per second over 36.332572 seconds)
Orientation prior camera-imu found as: (T_i_c)
[[-0.00109465 -0.99993847 0.01103865]
[-0.99997973 0.00102532 -0.00628425]
[ 0.00627254 -0.0110453 -0.99991933]]
Gyro bias prior found as: (b_gyro)
[ 0.00950291 -0.01491366 0.00564443]
Initializing a pose spline with 3641 knots (100.000000 knots per second over 36.412572 seconds)
Initializing the bias splines with 1821 knots
Adding camera error terms (/cam0/image_raw)
Added 546 camera error terms
Adding accelerometer error terms (/imu0)
Added 4181 of 4186 accelerometer error terms (skipped 5 out-of-bounds measurements)
Adding gyroscope error terms (/imu0)
Added 4181 of 4186 gyroscope error terms (skipped 5 out-of-bounds measurements)
Before Optimization
===================
Reprojection error squarred (cam0): mean 1.65924628048, median 0.921153456315, std: 2.4570824254
Gyro error squarred (imu0): mean 0.237577424056, median 0.12861625508, std: 0.394712498373
Accelerometer error squarred (imu0): mean 6.12980339624, median 4.65423239247, std: 5.85252682318
Transformation T_cam0_imu0 (imu0 to cam0, T_ci): [m]
[[-0.00109465 -0.99993847 0.01103865 0. ]
[-0.99997973 0.00102532 -0.00628425 0. ]
[ 0.00627254 -0.0110453 -0.99991933 0. ]
[ 0. 0. 0. 1. ]]
cam0 to imu0 time: [s] (t_imu = t_cam + shift)
0.0782779288516
Optimizing...
No linear system solver set in the options. Defaulting to the sparse_cholesky solver
Using the sparse_cholesky linear system solver
Using the levenberg_marquardt trust region policy
No linear system solver set in the options. Defaulting to the sparse_cholesky solver
Using the sparse_cholesky linear system solver
Using the levenberg_marquardt trust region policy
Initializing
Optimization problem initialized with 7302 design variables and 86262 error terms
The Jacobian matrix is 239158 x 32841
[0.0]: J: 149834
[1]: J: 16422.1, dJ: 133412, deltaX: 0.0312705, LM - lambda:100 mu:2
[2]: J: 16212.8, dJ: 209.324, deltaX: 0.0175721, LM - lambda:33.3333 mu:2
[3]: J: 15837.7, dJ: 375.117, deltaX: 0.149355, LM - lambda:11.1111 mu:2
[4]: J: 15234.5, dJ: 603.217, deltaX: 0.199769, LM - lambda:3.7037 mu:2
[5]: J: 15142.3, dJ: 92.1185, deltaX: 0.108999, LM - lambda:1.23457 mu:2
[6]: J: 15139.8, dJ: 2.52557, deltaX: 0.0504798, LM - lambda:0.576563 mu:2
[7]: J: 15137.7, dJ: 2.06968, deltaX: 0.030698, LM - lambda:0.685601 mu:2
[8]: J: 15130.3, dJ: 7.41588, deltaX: 0.0115357, LM - lambda:0.648712 mu:2
[9]: J: 15128.7, dJ: 1.58484, deltaX: 0.018595, LM - lambda:0.216237 mu:2
Last step was a regression. Reverting
[10]: J: 15133.6, dJ: -4.84494, deltaX: 0.133274, LM - lambda:0.0720791 mu:2
Last step was a regression. Reverting
[11]: J: 15133.6, dJ: -4.83709, deltaX: 0.0110752, LM - lambda:0.288316 mu:4
Last step was a regression. Reverting
[12]: J: 15133.6, dJ: -4.86729, deltaX: 0.00635325, LM - lambda:2.30653 mu:8
Last step was a regression. Reverting
[13]: J: 15134.3, dJ: -5.5453, deltaX: 0.000581093, LM - lambda:36.9045 mu:16
Last step was a regression. Reverting
[14]: J: 15134.3, dJ: -5.5885, deltaX: 1.21717e-05, LM - lambda:1180.94 mu:32
After Optimization (Results)
==================
Reprojection error squarred (cam0): mean 0.197612388404, median 0.129877879635, std: 0.220413415065
Gyro error squarred (imu0): mean 0.0167907582507, median 0.00725228030744, std: 0.0577435718152
Accelerometer error squarred (imu0): mean 0.0896374322406, median 0.0499228944401, std: 0.111785280951
Transformation T_cam0_imu0 (imu0 to cam0, T_ci): [m]
[[ 0.00109254 -0.9998685 0.01617978 0.09855709]
[-0.99999261 -0.00103277 0.00370201 0.02699315]
[-0.00368482 -0.0161837 -0.99986225 -0.00680206]
[ 0. 0. 0. 1. ]]
cam0 to imu0 time: [s] (t_imu = t_cam + shift)
0.0792052387331
Saving calibration to file: camchain-imucam-dataios_recording_102.yaml
Detailed results written to file: results-imucam-dataios_recording_102.txt
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment