C S 3 5 4 R e p r e s e n t i n g Ma p s : T o - - PowerPoint PPT Presentation

SLIDE 1

C S 3 5 4

SLIDE 2

R e p r e s e n t i n g Ma p s : T

• p
• l
• g

i c a l

EngGeo2002 EngGeo_Hall_A EngGeo2003 Spine_A EngGeo_Hall_B EnGeo2007

• R

e p r e s e n t r e l a t i v e l

• c

a t i

• n

s u s i n g a g r a p h s t r u c t u r e . + G

• d

f

• r

h i g h l e v e l n a v i g a t i

• n
• D

i ffj c u l t t

• b

u i l d a u t

• n
• m
• u

s l y

• N
• t

g

• d

f

• r

l

• w
• l

e v e l l

• c

a l i z a t i

• n

a n d n a v i g a t i

• n

SLIDE 3

R e p r e s e n t i n g Ma p s : G e

• m

e t r i c L a n d m a r k B a s e d

• S

t

• r

e t h e g e

• m

e t r i c l

• c

a t i

• n
• f

r e c

• g

n i z a b l e l a n d m a r k s .

– Ma

y b e a r t i fj c i a l b e a c

• n

s

• r

m a r k e r s .

– Ma

y b e d i s t i n c t i v e e n v i r

• n

m e n t a l f e a t u r e s .

+ Me m

• r

y

• e

ffj c i e n t + A l l

• w

s p r e c i s e l

• c

a l i z a t i

• n
• L

a n d m a r k m i s

• i

d e n t i fj c a t i

• n

c a n c a u s e p r

• b

l e m s

• Ma

y n

• t

b e i d e a l f

• r

n a v i g a t i

• n

:

• n

l y l a n d m a r k p

• s

i t i

• n

s a r e s t

• r

e d , n

• t

n e c e s s a r i l y t h e p

• s

i t i

• n

s

• f

a l l

• b

s t a c l e s

SLIDE 4

R e p r e s e n t i n g Ma p s : O c c u p a n c y G r i d

• D

i v i d e t h e e n v i r

• n

m e n t i n t

• g

r i d c e l l s , m a i n t a i n a n “

• c

c u p i e d ” p r

• b

a b i l i t y f

• r

e a c h c e l l .

• Me

m

• r

y i n t e n s i v e ( p a r t i c u l a r l y i n 3 D ) + G

• d

f

• r

n a v i g a t i

• n

+ G

• d

f

• r

l

• c

a l i z a t i

• n

+ R e l a t i v e l y s i m p l e t

• c

r e a t e a u t

• n
• m
• u

s l y

SLIDE 5

Q u a d t r e e s / O c t r e e s

• L

a r g e

• c

c u p a n c y g r i d s c a n b e e x p e n s i v e t

• s

t

• r

e :

– 1

m x 1 m m a p , 1 c m r e s

• l

u t i

• n

– 1

, , c e l l s

• Q

u a d t r e e i s a m

• r

e s p a c e

• e

ffj c i e n t a l t e r n a t i v e …

SLIDE 6

Q u a d t r e e s / O c t r e e s

• L

a r g e

• c

c u p a n c y g r i d s c a n b e e x p e n s i v e t

• s

t

• r

e :

– 1

m x 1 m m a p , 1 c m r e s

• l

u t i

• n

– 1

, , c e l l s

• Q

u a d t r e e i s a m

• r

e s p a c e

• e

ffj c i e n t a l t e r n a t i v e …

• O

c t r e e i s t h e 3 d g e n e r a l i z a t i

• n

:

http://en.wikipedia.org/wiki/File:Octree2.svg, http://creativecommons.org/licenses/by-sa/3.0/

SLIDE 7

Ma p p i n g w / O c c u p a n c y G r i d s

• R

e l a t i v e l y e a s y i f w e k n

• w

t h e r

• b
• t

p

• s

e :

Increase occupied probability Decrease occupied probability H.P. Moravec. Sensor fusion in certainty grids for mobile robots. AIMagazine, pages 61–74, Summer 1988.

SLIDE 8

S L A M – S i m u l t a n e

• u

s L

• c

a l i z a t i

• n

a n d Ma p p i n g

• R

e c a l l t h e l

• c

a l i z a t i

• n

p r

• b

l e m :

• T

h e S L A M p r

• b

l e m i s r e a s s u r i n g l y f a m i l i a r :

• Wh

e r e r e p r e s e n t s t h e m a p .

• B

e f

• r

e w e w a n t e d a p r

• b

a b i l i t y d i s t r i b u t i

• n
• v

e r a l l p

• s

s i b l e r

• b
• t

p

• s

e s .

• N
• w

w e w a n t a j

• i

n t p r

• b

a b i l i t y d i s t r i b u t i

• n
• v

e r a l l p

• s

s i b l e r

• b
• t

p

• s

e s a n d a l l p

• s

s i b l e m a p s .

“ D i s t r i b u t i

• n
• v

e r p

• s

s i b l e m a p s ” i s n

• t

a s m a n a g e a b l e a s “ d i s t r i b u t i

• n
• v

e r p

• s

e s ”

SLIDE 9

S L A M “ S

• l

u t i

• n

”

• P

r e d i c t i

• n

:

• C
• r

r e c t i

• n

:

Adapted from Simultaneous Localisation and Mapping (SLAM): Part 1 The Essential Algorithms, Hugh Durrant-Whyte, 2006

SLIDE 10

S L A M S

• l

u t i

• n

s

• S
• l

u t i

• n

s f a l l i n t

• t

h r e e f a m i l i e s ( i n r

• u

g h l y h i s t

• r

i c a l

• r

d e r )

– E

F K S L A M

– P

a r t i c l e

• F

i l t e r S L A M

– G

r a p h S L A M

SLIDE 11

( E x t e n d e d ) K a l m a n F i l t e r S L A M

• Mo

s t a p p r

• p

r i a t e f

• r

l a n d m a r k

• b

a s e d m a p s .

• P

r

• b

l e m s :

– N

• t

c l e a r h

• w

t

• u

s e t h i s f

• r
• c

c u p a n c y g r i d s

– T

h e c

• v

a r i a n c e m a t r i x g e t s b i g a s t h e n u m b e r

• f

l a n d m a r k s g r

• w

s

– V

i d e

• E

x a m p l e

SLIDE 12

P a r t i c l e F i l t e r S L A M: P r

• b

l e m s

• C
• v

e r i n g t h e s p a c e

• f

p

• s

s i b l e p

• s

e s a n d m a p s w i t h p a r t i c l e s i s n

• t

p r a c t i c a l :

– “

P

• s

e p a r t i c l e ” : 3

• 6

d i m e n s i

• n

s

– “

Ma p p a r t i c l e ” f

• r

a ( t i n y ) 1 x 1 g r i d : 1 d i m e n s i

• n

s

– J

• i

n t m a p x p

• s

e p a r t i c l e : 3

• 6

d i m e n s i

• n

s

SLIDE 13

R a

• B

l a c k w e l l i z e d P a r t i c l e F i l t e r f

• r

S L A M

• S
• l

u t i

• n

/ A p p r

• x

i m a t i

• n

:

– E

a c h p

• s

e p a r t i c l e h a s a n a s s

• c

i a t e d m a p .

– E

a c h m a p i s u p d a t e d u n d e r t h e a s s u m p t i

• n

t h a t i t s p a r t i c l e r e p r e s e n t s t h e c

• r

r e c t p

• s

e .

– T

h e m a p m a y b e l a n d m a r k

• b

a s e d

• r
• c

c u p a n c y g r i d

• b

a s e d .

SLIDE 14

G r a p h S L A M

Thrun, Sebastian, and Michael Montemerlo. "The graph SLAM algorithm with applications to large-scale mapping of urban structures." The International Journal of Robotics Research 25.5-6 (2006): 403-429.

SLIDE 15

C h a l l e n g e s

• L

O O P C L O S U R E S ! ! ! !