Documente Academic
Documente Profesional
Documente Cultură
Particle Filtering
for Tracking and Localization
Data
z
d
x
Sensor Model P(z|x)
d
Most often
Tutorial : Monte analytic expression,
Carlo Methods Frank Dellaert can be learnedOctober 07
Recap: Bayes Law
P(x|z) ~ L(x;z)P(x)
data = z
model L(x;z)
Prior P(x)
Likelihood
L(x;z)
Posterior
P(x|z)
Tutorial : Monte Carlo Methods Frank Dellaert October 07
Problem: Large Open Spaces
P(z|x)
z
h(x)
A B C D
ABCDE
Very powerful
Gaussian, unimodal
sensor
motion motion
p(x)
x2
p(x)
x2
x1
Tutorial : Monte Carlo Methods Frank Dellaert October 07
Sampling Advantages
Arbitrary densities
Memory = O(#samples)
Only in Typical Set
Great visualization tool !
minus: Approximate
Mean
Variance (1D)
Expected angle = 30o
Tutorial : Monte Carlo Methods Frank Dellaert October 07
How to Sample ?
Image by MacKay
9% Rejection Rate
Tutorial : Monte Carlo Methods Frank Dellaert October 07
the Bad
Prior(x,y)
Posterior(x,y|measured_angle=20o)
Idea:
sample from P(x)
give each sample x(r) a importance weight equal
to P(Z|x (r))
Tutorial : Monte Carlo Methods Frank Dellaert October 07
Example Importance Sampling
{x(r),y(r)~Prior(x,y), wr=P(Z|x(r),y(r)) }
q(x)
(x)
Image by MacKay
B
G
B
G
y
R
P(x, y | color) " P(color | x, y)P(x, y)
P(color | x, y) = $ N( ,# c c ^2)
c= r,g,b
Tutorial : Monte Carlo Methods Frank Dellaert October 07
References
Prior P(x)
Likelihood
L(x;z)
Posterior
P(x|z)
Tutorial : Monte Carlo Methods Frank Dellaert October 07
1D Importance Sampling
Motion Model
Recursive Bayes Filter Equation:
Predictive Density
Monte Carlo Approximation:
(2)
(1) (3)
Sk-1 S k
weighted Sk Sk
Sequential Estimation
Iterates over:
Prediction with motion model
Importance Sampling for Inference
Resampling
P(xt| ,u)
Motion Model
P(Z|xt)
Sensor Model
O(N)
B
G
B
G
y
R
P(x, y | color) " P(color | x, y)P(x, y)
P(color | x, y) = $ N( ,# c c ^2)
c= r,g,b
Tutorial : Monte Carlo Methods Frank Dellaert October 07