DeepPicar: A Low-cost Deep Neural Network-based Autonomous Car.

We present DeepPicar, a low-cost deep neural network (DNN) based autonomous car platform. DeepPicar is a small scale replication of a real self-driving car called Dave-2 by NVIDIA, which drove on public roads using a deep convolutional neural network (CNN), that takes images from a front-facing camera as input and produces car steering angles as output. DeepPicar uses almost the exact same network architecture---9 layers, 27 million connections and 250K parameters---and can be trained to drive itself, in real-time, using a web camera and a modest Raspberry Pi 3 quad-core platform. Using DeepPicar, we analyze the Pi 3's computing capabilities to support end-to-end deep learning based real-time control of autonomous vehicles. We also systematically compare other contemporary embedded computing platforms using the DeepPicar's CNN based real-time control software as a workload. We find all tested platforms, including the Pi 3, are capable of supporting deep-learning based real-time control, from 20 Hz up to 100 Hz depending on hardware platform. However, shared resource contention remains an important issue that must be considered in applying deep-learning models on shared memory based embedded computing platforms.