Sociology

Modalities

Social science

Pipeline (software)

Biology

Operating system

Machine learning

Paleontology

Artificial intelligence

Programming language

Interpretability

Context (archaeology)

Computer science

Artificial neural network

End-to-end principle

End user

IEEE Transactions on Neural Networks and Learning Systems

Autonomous driving is of great interest to industry and academia alike. The use of machine learning approaches for autonomous driving has long been studied, but mostly in the context of perception. In this paper we take a deeper look on the so called end-to-end approaches for autonomous driving, where the entire driving pipeline is replaced with a single neural network. We review the learning methods, input and output modalities, network architectures and evaluation schemes in end-to-end driving literature. Interpretability and safety are discussed separately, as they remain challenging for this approach. Beyond providing a comprehensive overview of existing methods, we conclude the review with an architecture that combines the most promising elements of the end-to-end autonomous driving systems.

A Survey of End-to-End Driving: Architectures and Training Methods