Behavioural Cloning
The goals of this project were the following:
- Generate and augment a behavioural cloning dataset by driving in a simulator.
- Build a deep-learning end-to-end driving model that predicts driving actions from camera data.
- Test, train and validate the model using the simulator driving data.
- Apply the model in the simulator, recording a video of the completion of one lap of the track.
Requirements
numpy
scipy
pandas
sklearn
matplotlib
seaborn
opencv
jupyterlab
tensorflow
In addition, this project requires the Term 1 Udacity simulator: https://github.com/udacity/self-driving-car-sim
Usage
To train the model:
- Download the relevant release of the Udacity simulator for your platform.
- Use the simulator to drive vehicle in manual mode and record data.
python preproc.py
python model.py
To run the trained model:
- Run the driving model:
python drive.py models/model.h5 run1
. - Run the simulator in autonomous mode.