Behavioral Cloning

Project 3 in Udacity self-driving car nanodegree: train neural network to steer a wheel in a racing track

Description of our implementation is available here.

Setup

• Follow instructions here to set up Anaconda environment of Term 1 . You can verify that your environment is set up correctly using the instructions here.
• Download driving simulator from here.

Files

• model.py - code for training neural network and generating a model
• model.h5 - model generated by model.py
• drive.py - code for driving a car using generated model
• video.mp4 - video of driving using the model model.h5
• writeup_report.md - description of the process of data gathering and building a model
• images - folder of images used in writeup_report.md

Usage

• Activate environment of Term 1 by running

  activate carnd-term1

• Run driving simulator, choose autonomous mode and the left track

• Go to directory of drive.py and run

  python drive.py model.h5

• Alternatively, download and play video.mp4