Skip to content

A simple simulation of a self driving car in a highway traffic using feed forward neural networks.

Notifications You must be signed in to change notification settings

narenkarthicktp/self-driving-car

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

15 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Self-driving car 🚗

A simple simulation of a self driving car in a highway traffic using feed forward neural networks. The neural network is refined using genetic algorithm. The network is mutated everytime the simulation is reloaded resulting in many slightly altered networks, in which some of them might perform better than the parent. This way, the best car from each simulation is saved and mutated in the next simulation to improve the neural network.


Preview :


Steps to train the car :

  • Run the simulation until the car crashes [eventually]
  • Save the model and reload
  • Repeat the above steps until you get an optimal model that learns to reach a specific distance without crashing

Tip : Inorder to train the network faster, try increasing the mutation rate and no. of car instances


About

A simple simulation of a self driving car in a highway traffic using feed forward neural networks.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages