Just to recap, in the class we learned about localization. Localization

That's the localization question.

You learned about localization.

Here is a localization quiz.

We will talk today of localization.

Let's get back to Monte Carlo localization.

Here is the critical step for localization.

So congratulations. You understand probability and localization.

Let me ask you a localization question.

If you understand this step, you understand localization.

So localization, for example, it's clearly an approximate filter.

You should say to yourself, I just implemented localization.

What it really means is simultaneous localization and mapping.

It did localization similar to what I taught you before.

The next question we had was about internationalization and localization.

You wrote an algorithm that implements what's called Markov localization.

The very first problem I'm trying to solve is called localization.

In the Google self driving car, localization plays a key role.

If you understood this, you understand probability, and you understand localization.

You've programmed the core of Google's self driving car localization methods.

Sebastian Localization does depend on having a map of the environment.

Well, it was really easy to compute in our localization example.

Now, localization is nothing else but the iteration of sense and move.

Before we're done with localization, I'd like to talk about robot motion.

This representation is absolutely core to probability and to mobile robot localization.

Thrun Here is the application of HMM to a real robot localization example.

In Monte Carlo localization, we don't predict the exact outcome. We add noise.

SLAM is a method for mapping that's short for simultaneous localization and mapping.

You might remember in the localization classes we assumed the map was given.

If you built a localization system for robots with higher dimensional state spaces,

This is a supplemental system to the localization class I just taught you.

Driving a car is autonomously is primarily a question in perception and localization.

Here is our 1 dimensional localization example again, this time with particle filters.

But what you've implemented is the core of Google's self driving car localization idea.

So localization has a lot of math, but before I dive into mathematical detail,

I wonder how the memory used will scale for our histogram based localization method.

You can do localization on a hand held phone in a room using images.

Let's talk about localization of a car like our DARPA Urban Challenge car Junior.

So, it works really well for low dimensional problems like 3 dimensional robot localization problems.

Realize that now the state space is 3 dimensional, just like in our localization example.

One of our largest sources of community volunteer work is localization, which is basically translation.

I want to spend a few minutes and go over the formal definition of localization.

Just as before, begin with a uniform distribution as you always do in global localization.

You learned about Monte Carlo robot localization with a technique that I often call histogram filters.

Then localization cycles through these move, sense, move, sense, move, sense. move, sense, move, sense cycle.