supplemental material for eneralizing locomotion style to...

7
Supplemental Material for Generalizing Locomotion Style to New Animals With Inverse Optimal Regression Kevin Wampler * Adobe Research Zoran Popovi´ c University of Washington Jovan Popovi´ c Adobe Research 1 Parameter Plots The following plots compare the quadruped inverse parameters found by independent and joint inverse optimization for all of the parameters listed in table 1. 2 4 6 8 10 0 10 1 10 2 10 3 10 4 2 4 6 8 independent joint m i h g h g 0 1 2 3 4 5 6 7 10 0 10 1 10 2 10 3 10 4 0 1 2 3 4 5 6 7 independent joint m i w g w g * e-mail:[email protected] e-mail:[email protected] e-mail:[email protected] 0 1 2 3 4 5 10 0 10 1 10 2 10 3 10 4 0 1 2 3 4 5 independent joint m i w v w v 0.0 0.1 0.2 0.3 0.4 0.5 10 0 10 1 10 2 10 3 10 4 0.0 0.1 0.2 0.3 0.4 0.5 independent joint m i h w h w 0 2 4 6 8 10 10 0 10 1 10 2 10 3 10 4 0 2 4 6 8 10 independent joint m i d w d w 0 1 2 3 4 5 10 0 10 1 10 2 10 3 10 4 0 1 2 3 4 5 independent joint m i c l c l

Upload: others

Post on 01-Jun-2020

3 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Supplemental Material for eneralizing Locomotion Style to ...grail.cs.washington.edu/projects/inverse... · Generalizing Locomotion Style to New Animals With Inverse Optimal Regression

Supplemental Material forGeneralizing Locomotion Style to New Animals With Inverse Optimal Regression

Kevin Wampler∗

Adobe ResearchZoran Popovic†

University of WashingtonJovan Popovic‡

Adobe Research

1 Parameter Plots

The following plots compare the quadruped inverse parametersfound by independent and joint inverse optimization for all of theparameters listed in table 1.

2

4

6

8

100 101 102 103 104

2

4

6

8

independent

joint

mi

hg

hg

01234567

100 101 102 103 104

01234567

independent

joint

mi

wg

wg

∗e-mail:[email protected]†e-mail:[email protected]‡e-mail:[email protected]

012345

100 101 102 103 104

012345

independent

joint

mi

wv

wv

0.00.10.20.30.40.5

100 101 102 103 104

0.00.10.20.30.40.5

independent

joint

mi

hw

hw

0246810

100 101 102 103 104

0246810

independent

joint

mi

dw

dw

012345

100 101 102 103 104

012345

independent

joint

mi

cl

cl

Page 2: Supplemental Material for eneralizing Locomotion Style to ...grail.cs.washington.edu/projects/inverse... · Generalizing Locomotion Style to New Animals With Inverse Optimal Regression

1.01.52.02.53.03.54.0

100 101 102 103 104

1.01.52.02.53.03.54.0

independent

joint

mi

rl

rl

012345

100 101 102 103 104

012345

independent

joint

mi

ca

ca

0.00.51.01.52.02.5

100 101 102 103 104

0.00.51.01.52.02.5

independent

joint

mi

ra

ra

1234567

100 101 102 103 104

1234567

independent

joint

mi

th

th

234567

100 101 102 103 104

234567

independent

joint

mi

tk

tk

123456

100 101 102 103 104

123456

independent

joint

mi

ta

ta

0123456

100 101 102 103 104

0123456

independent

joint

mi

ts

ts

012345

100 101 102 103 104

012345

independent

joint

mi

te

te

Page 3: Supplemental Material for eneralizing Locomotion Style to ...grail.cs.washington.edu/projects/inverse... · Generalizing Locomotion Style to New Animals With Inverse Optimal Regression

01234567

100 101 102 103 104

01234567

independent

joint

mi

tw

tw

0

2

4

6

8

100 101 102 103 104

0

2

4

6

8

independent

joint

mi

wf

wf

0.00.10.20.30.40.5

100 101 102 103 104

0.00.10.20.30.40.5

independent

joint

mi

wa

wa

3.54.04.55.05.56.06.57.0

100 101 102 103 104

3.54.04.55.05.56.06.57.0

independent

joint

mi

wh

wh

0123456

100 101 102 103 104

0123456

independent

joint

mi

kh

kh

−1.5−1.0−0.50.00.51.01.5

100 101 102 103 104

−1.5−1.0−0.50.00.51.01.5

independent

joint

mi

qh

qh

1234567

100 101 102 103 104

1234567

independent

joint

mi

dh

dh

0.51.01.52.02.53.03.54.0

100 101 102 103 104

0

1

2

3

4

independent

joint

mi

kk

kk

Page 4: Supplemental Material for eneralizing Locomotion Style to ...grail.cs.washington.edu/projects/inverse... · Generalizing Locomotion Style to New Animals With Inverse Optimal Regression

−1.5−1.0−0.50.00.51.0

100 101 102 103 104

−1.5−1.0−0.50.00.51.0

independent

joint

mi

qk

qk

01234567

100 101 102 103 104

01234567

independent

joint

mi

dk

dk

0

2

4

6

8

100 101 102 103 1040

2

4

6

8

independent

joint

mi

ka

ka

−1.0

−0.5

0.0

0.5

1.0

100 101 102 103 104

−1.5−1.0−0.50.00.51.0

independent

joint

mi

qa

qa

23456789

100 101 102 103 104

23456789

independent

joint

mi

da

da

1

2

3

4

5

100 101 102 103 104

1

2

3

4

5

independent

joint

mi

ks

ks

−1.5−1.0−0.50.00.51.01.5

100 101 102 103 104

−1.5−1.0−0.50.00.51.01.5

independent

joint

mi

qs

qs

0.51.01.52.02.53.03.5

100 101 102 103 104

0.51.01.52.02.53.03.5

independent

joint

mi

ds

ds

Page 5: Supplemental Material for eneralizing Locomotion Style to ...grail.cs.washington.edu/projects/inverse... · Generalizing Locomotion Style to New Animals With Inverse Optimal Regression

123456

100 101 102 103 104

123456

independent

joint

mi

ke

ke

0.00.20.40.60.81.01.21.4

100 101 102 103 104

0.00.20.40.60.81.01.21.4

independent

joint

mi

qe

qe

1234567

100 101 102 103 104

1234567

independent

joint

mi

de

de

2345678

100 101 102 103 104

12345678

independent

joint

mi

kw

kw

−1.0−0.50.00.51.01.5

100 101 102 103 104

−1.0−0.50.00.51.01.5

independent

joint

mi

qw

qw

12345678

100 101 102 103 104

12345678

independent

joint

mi

dw

dw

Page 6: Supplemental Material for eneralizing Locomotion Style to ...grail.cs.washington.edu/projects/inverse... · Generalizing Locomotion Style to New Animals With Inverse Optimal Regression

Figure 1: From top left to bottom right: plots of the hip angle, knee angle, ankle angle, and foot height over the frames in gaits for an ostrich.

Figure 2: From top left to bottom right: plots of the hip angle, knee angle, ankle angle, and foot height over the frames in gaits for aThomson’s gazelle.

Figure 3: From top left to bottom right: plots of the hip angle, knee angle, ankle angle, and foot height over the frames in gaits for anelephant.

Page 7: Supplemental Material for eneralizing Locomotion Style to ...grail.cs.washington.edu/projects/inverse... · Generalizing Locomotion Style to New Animals With Inverse Optimal Regression

Figure 4: From top left to bottom right: plots of the hip angle, knee angle, ankle angle, and foot height over the frames in gaits for a giraffe.