8000 GitHub - odegnome/sb3_quad: Implementation of Deep Reinforcement Learning for Collision Prevention in Quadrotor - II
[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
Skip to content

odegnome/sb3_quad

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Implementation of Deep Reinforcement Learning for Collision Prevention in Quadrotor - II

This is a continuation of my dissertation project for building an autonomous quadrotor. Difference from my original implementation is that this uses sb3 algorithm to train and also, yaw axis acceleration is also a consideration in the reward. The yaw axis reward was necessary because otherwise, the quadrotor was spinning about that axis uncontrollably.

Performance after training for 500k time-steps

There is a glitch in the video below due to incorrect use of ffmpeg for converting images to mp4. Please ignore!

500k.mp4

About

Implementation of Deep Reinforcement Learning for Collision Prevention in Quadrotor - II

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

0