Skip to content

GKPr0/Autonomous-Robots-Localization

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Autonomous Robots: Localization

Repositary for Localization Course

Course link: https://www.udemy.com/course/autonomous-robots-localization/

Required Packages
python=3.7.4
numpy=1.16.4
matploblib=3.1.0

Assigment 1: Least Squares

-> Based on locations of poles and distance from robot´s position to each of the pole calculate robot´s location.
-> You don´t know which measuremnet corespond to which pole.
-> Based on cost function, you will find the location with lowest total cost.
-> Cost function: (Minimal total distance diference between guess location and pole measurements)**2

Assigment 2: Bayes rule

-> Bayes rule used to determinate location of robot in 1D world
-> Bayes rule is used to update probability of some phenomena based on new observation or in other words update your beliefs based on evidence
-> Bayes rule: https://en.wikipedia.org/wiki/Bayes%27_theorem or https://www.youtube.com/watch?v=HZGCoVF3YvM
-> 2-1 -> Robot moves every time exactly same distance with 0 uncertainty
-> 2-2 -> Robots moves with some move uncertainty

Assigment 3: 1D Particle filter

-> Using Particle filter to determinate location of robot in 1D world
-> Particle filter used multiple particles (virtual copy of robot) to determinate it position.
In every loop particles will:
1. Move -> movement of particles is identical to real movement of robot. Only difference is that every particle moves with some uncertainty which is based on gaussian distribution
2. Measure -> Robot and particles scan its environment. Measurement has also some uncertainty
3. Update -> Based on difference between what robot sees and particle sees weights are calculated.
4. Resample -> Some portion of particles with the highest weights will be choosen to be fundamental, and rest of the particles will be resampled around these fundamentals ones.

Assigment 4: 2D Particle filter

-> Using Particle filter to determinate location of robot in 2D world

About

Robot Localization with Bayes rule and Particle filter

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages