Systematic Testing of Convolutional Neural Networks for Autonomous Driving
This work addresses safety and reliability issues for autonomous driving systems by providing a systematic testing method, though it is incremental as it builds on existing CNN analysis techniques.
The authors tackled the problem of analyzing convolutional neural networks (CNNs) for car classification in autonomous vehicles by developing a framework with a synthetic image generator and visualization tools to expose vulnerabilities and extract insights.
We present a framework to systematically analyze convolutional neural networks (CNNs) used in classification of cars in autonomous vehicles. Our analysis procedure comprises an image generator that produces synthetic pictures by sampling in a lower dimension image modification subspace and a suite of visualization tools. The image generator produces images which can be used to test the CNN and hence expose its vulnerabilities. The presented framework can be used to extract insights of the CNN classifier, compare across classification models, or generate training and validation datasets.