Chapter 16 - Simulating a Self-Driving Car using End-to-End Deep Learning with Keras

Note: All images in this directory, unless specified otherwise, are licensed under CC BY-NC 4.0.

Figure List

Figure number	Description
16-1	The SAE Levels of Driving Automation
16-2	The NavLab 1 in all its glory
16-3	AirSim’s photo realistic simulation versus Apollo postperception simulation
16-4	The Landscape map in AirSim
16-5	Plot showing the contents of the file normal_1/images/img_0.png
16-6	The three parts of the image as seen by the self-driving car
16-7	The ROI for our car to focus on during training
16-8	Flipping image on the y-axis
16-9	Reducing image brightness by 40%
16-10	Plot showing steering angles from the two driving strategies
16-11	Dataset split for the two driving strategies
16-12	Steering angle distribution for the two driving strategies
16-13	Drawing steering angles on images
16-14	Network architecture
16-15	Drawing actual and predicted steering angles on images
16-16	Trained model driving the car
16-17	Architecture of NVIDIA’s DAVE-2 system