Skip to content

Redfalcon5-ai/Capstone_AWS_DeepRacer

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

66 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Capstone Project for the Master of Science in Business Analytics at ESADE Business School

This repository contains the code that was used for the article "An Advanced Guide to AWS DeepRacer - Autonomous Formula 1 Racing using Reinforcement Learning". Feel free to check it out here.

  • The folder Compute_Speed_And_Actions contains a jupyter notebook, which takes the optimal racing line from this repo and computes the optimal speed. Additionally, it computes a custom action space with K-Means clustering. The folder also contains the K1999 racing line notebook from cdthompson, which I altered to be able to only use the inner 80% of the track.
  • The folder Reward_Function contains a .py file with the reward function that our team used to get to 12th place out of 1291 participants in the time trial category of the F1 event in May 2020
  • The folder Selenium_Automation contains a jupyter notebook, which allows you to submit a model to a race multiple times without using the AWS CLI. As a bonus, you can also automatically conducts experiments with hyperparameters. This can be used to conduct multiple experiments over night without having to manually set them up every couple of hours

GitHub Repositories that were used

License

Feel free to use, distribute, and alter the code as you like.

This is a finished university project. Therefore, we will not be maintaining the code any more.

About

Code that was used in the Article "An Advanced Guide to AWS DeepRacer"

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Jupyter Notebook 95.9%
  • Python 4.1%