A Review of End-to-End Autonomous Driving in Urban Environments

Abstract

Autonomous driving in urban environments requires intelligent systems that are able to deal with complex and unpredictable scenarios. Traditional modular approaches focus on dividing the driving task into standard modules, and then use rule-based methods to connect those different modules. As such, these approaches require a significant effort to design architectures that combine all system components, and are often prone to error propagation throughout the pipeline. Recently, end-to-end autonomous driving systems have formulated the autonomous driving problem as an end-to-end learning process, with the goal of developing a policy that transforms sensory data into vehicle control commands. Despite promising results, the majority of end-to-end works in autonomous driving focus on simple driving tasks, such as lane-following, which do not fully capture the intricacies of driving in urban environments. The main contribution of this paper is to provide a detailed comparison between end-to-end autonomous driving systems that tackle urban environments. This analysis comprises two stages: a) a description of the main characteristics of the successful end-to-end approaches in urban environments; b) a quantitative comparison based on two CARLA simulator benchmarks (CoRL2017 and NoCrash). Beyond providing a detailed overview of the existent approaches, we conclude this work with the most promising aspects of end-to-end autonomous driving approaches suitable for urban environments.INDEX TERMS autonomous driving, end-to-end, imitation learning, reinforcement learning, urban environments.

Integrating the Inception V3 model in an end-to-end driving scenario

“…In contrast, end-to-end systems employ a large deep learning model to directly output control commands, significantly reducing the training cost and enhancing the model's ability to adapt to complex environments [2]. Despite these differences, end-to-end approach, like modular approach, requires tasks like image recognition and object perception.…”

Efficiency study of VGG networks in autonomous driving tasks

“…Its primary objective is to acquire a strategy to convert sensor-perceived data into control commands. Compared to the modular approach, the end-to-end approach is characterized by its simplicity and resemblance to human-driven behavior, as it encompasses both perception and action [4]. Imitation learning (IL) is the main paradigm in the end-to-end approach, which serves as a supervised learning method designed to train models to imitate expert behavior.…”