Giters

lucianzhong / DeepDriving

Driving in TORCS using Caffe following Direct Perception approach

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Deep Driving

Autonomous driving using deep learning (Caffe) in TORCS-- open source car racing game.

This code is a modified version Princeton's Deep Driving http://deepdriving.cs.princeton.edu/, which was published to back the results claimed in http://deepdriving.cs.princeton.edu/paper.pdf.

The project report for the code in this repo can be found in saiprabhakar.github.io/projects.html.

Results for the original implementation:

  1. https://www.youtube.com/watch?v=5Yf1V5-MggE

  2. https://www.youtube.com/watch?v=Ffyi9W23Fl0

About

Driving in TORCS using Caffe following Direct Perception approach

Languages

Language:Jupyter Notebook 48.2%Language:C++ 42.4%Language:Python 4.8%Language:Cuda 2.3%Language:CMake 0.6%Language:Protocol Buffer 0.6%Language:Makefile 0.4%Language:Shell 0.3%Language:MATLAB 0.2%Language:CSS 0.2%Language:HTML 0.1%