Return to Video

04-18 Shifting the Mean

  • 0:00 - 0:05
    In Kalman filters we iterate measurement and motion.
  • 0:05 - 0:08
    This is often called a "measurement update,"
  • 0:08 - 0:10
    and this is often called "prediction."
  • 0:10 - 0:17
    In this update we'll use Bayes rule, which is nothing else but a product or a multiplication.
  • 0:17 - 0:24
    In this update we'll use total probability, which is a convolution,
  • 0:24 - 0:27
    or simply an addition.
  • 0:27 - 0:35
    Let's talk first about the measurement cycle and then the prediction cycle,
  • 0:35 - 0:44
    using our great, great, great Gaussians for implementing those steps.
  • 0:44 - 0:47
    Suppose you're localizing another vehicle,
  • 0:47 - 0:50
    and you have a prior distribution that looks as follows.
  • 0:50 - 0:55
    It's a very wide Gaussian with the mean over here.
  • 0:55 - 0:58
    Now, say we get a measurement that tells us something about
  • 0:58 - 1:03
    the localization of the vehicle, and it comes in like this.
  • 1:03 - 1:07
    It has a mean over here called "mu,"
  • 1:07 - 1:11
    and this example has a much smaller covariance for the measurement.
  • 1:11 - 1:16
    This is an example where in our prior we were fairly uncertain about a location,
  • 1:16 - 1:21
    but the measurement told us quite a bit as to where the vehicle is.
  • 1:21 - 1:23
    Here's a quiz for you.
  • 1:23 - 1:36
    Will the new mean of the subsequent Gaussian be over here, over here, or over here?
  • 1:36 -
    Check one of these three boxes.
Title:
04-18 Shifting the Mean
Description:

04-18 Shifting the Mean

more » « less
Video Language:
English
Team:
Udacity
Project:
CS373 - Artificial Intelligence
Duration:
01:39
Udacity Robot edited English subtitles for 04-18 Shifting the Mean
Amara Bot added a translation

English subtitles

Incomplete

Revisions Compare revisions