DeepPicar is a low-cost autonomous RC car platform using a deep convolutional neural network (CNN). DeepPicar is a small scale replication of NVIDIA's real self-driving car called Dave-2, which drove on public roads using a CNN. DeepPicar uses the same CNN architecture of NVIDIA's Dave-2 and can drive itself in real-time locally on a Raspberry Pi 3.
Video:
Some other examples of the DeepPicar driving can be found at: https://photos.app.goo.gl/q40QFieD5iI9yXU42
If you wish to recreate the paper's findings, you can train one or more models using our dataset which can be found at: https://drive.google.com/open?id=1LjIcOVH7xmbxV58lx3BClRcZ2DACfSwh
Please refer to Embedded Platform Comparison for the steps needed to run the experiments conducted in the paper.
DeepPicar is comprised of the following components:
Please refer to Parts and Assembly for assembly steps, and Setup and Operation for in-depth installation and usage instructions.
The DeepPicar code utilizes MIT's DeepTesla (https://github.com/lexfridman/deeptesla), which provides a TensorFlow version of NVIDIA Dave-2's CNN.
NVIDIA Dave-2 (and its CNN) is described in the following paper. https://arxiv.org/pdf/1604.07316
The paper for DeepPicar can be found at https://arxiv.org/abs/1712.08644. It can be cited using the following BibTeX entry:
@inproceedings{bechtel2018picar,
title = {DeepPicar: A Low-cost Deep Neural Network-based Autonomous Car},
author = {Michael Garrett Bechtel and Elise McEllhiney and Minje Kim and Heechul Yun},
booktitle = {IEEE International Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA)},
year = {2018}
}