Lesson 3: Particle Filter - RobotWorld

Classes, Instance, and Inheritance (WebChurch)


;   - WebChurch code by PCM 2016/03/27 -
;   Landmarks and World Size of Particle World
;   (Thrun, AI for Robotics)
;   ....................

The world for localization by particle filtering is a simple 2-D square with edge length 100. It contains four landmarks to support the position finding process. From the programming language perspective the main problem is that Thrun used an object-oriented approach in Python. This implies mutation of local states of instance objects. Because WebChurch is purely functional we cannot use mutations. So we have to look for an object-based style with a functional state. So we take the old state as an input-argument and the new state as a functional output so that the methods remain pure functions.

Class Functions (e.g. Move) with Nested Defines and Command Sequences as Function Compositions (WebChurch)

Move and Sense with Nested Defines (WebChurch)

Move and Sense with Nested Lambdas (WebChurch)

; - WebChurch code by PCM 2016/04/08 -
; instruction is taken from S.Thrun's lecture "AI for Robotics"
;  "Make a robot called myrobot that starts at coordinates 30, 50 heading north (pi/2).
;   Have your robot turn clockwise by pi/2, move 15 m, and sense.
;   Then have it turn clockwise by pi/2 again, move 10 m, and sense again.
;   Your program should print out the result of your two sense measurements."
;   (S.Thrun, AI for Robotics)

A Markov Model for Actions, States, and Observations (WebChurch)

A Markov Model for Actions, States, and Observations in a Cyclic World (WebChurch)

Webmas8o2tertqm (manut2/ (Changed: 2020-01-23)