Quantifying Uncertainties in Depth Estimation for Autonomous Driving in CARLA.

This bachelor thesis deals with the topic of distance estimation in the context of autonomous driving. There is a variety of techniques and sensors used in this field, often using multiple combinations and complex pipelines to estimate distances with high precision. In this thesis, I focus on a simple, straightforward architecture using only a monocular camera. For that, I have built a convolutional neural network that takes images from a single camera mounted on the hood of the vehicle as input and outputs a distance estimation. The model is trained with custom generated datasets in the CARLA simulator, which are used to train the model in a supervised manner. Furthermore, I analyze different model parameters and their impact on the performance, especially the resulting uncertainty of the model. The uncertainty aspects are analyzed using a deep ensemble, where multiple models with the same parameters are trained and compared on their performance, and a variety of image augmentations to test their influence on the model’s uncertainty.