Skip to content

Visualizations of algorithms covered in Sebastian Thrun's excellent Artificial Intelligence for Robotics course on Udacity.

Notifications You must be signed in to change notification settings

daniel-s-ingram/ai_for_robotics

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

75 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Artificial Intelligence for Robotics by Sebastian Thrun

Lesson 1

Histogram Localization in a Cylic 1D World

Using only dead reckoning, the robot's position quickly becomes uncertain even when it starts with 100% certainty of where it is. Dead reckoning uses only information about how the robot tries to move to estimate its position, but there is a possibility that the robot will move by some amount less than or more than it tried to move. Eventually, the robot's belief of where it is approaches a state of maximum uncertainty, known as the limit distribution, in which the robot believes it could be anywhere.

The only way to overcome this uncertainty is to use the known positions of landmarks to get a better estimate of the robot's position. Landmarks even help to localize when we start with no idea of where we are.

Histogram Localization in a Cyclic 2D World

Lesson 2

1D Kalman Filter

2D Kalman Filter

Lesson 3

Particle Filter

Lesson 4

A*

D*

Lesson 5

Path Smoothing

P Controller

PID Controller

Lesson 6

SLAM