The data to train autonomous cars was not so abundant before a few years. After Waymo released their driving data, it is widely used in academic research. It consists of a huge amount of great quality images which is collected from various driving scenarios. Reliable and authenticated driving policies have crucial roles to develop efficient automated driving systems. This becomes one of the fundamental challenges for researchers to find the precise solution for the same. Academic researchers need to make several assumptions for the implementation of automated vehicle parts in their simulations or models. These assumptions may not be relevant to the real-time interactions as per the simulation-based research. Since the internal driving policy is under proprietary protection, researchers need to design robust and reliable policies to implement automated driving parts using deep learning models. This paper analyzes several deep learning systems to learn autonomous driving behavior using Waymo's dataset. In addition, this article provides an extensive overview of different aspects of designing automated driving simulators.
An Overview of Deep Learning Techniques for Autonomous Driving Vehicles
20.01.2022
431411 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Deep Reinforcement-Learning-based Driving Policy for Autonomous Road Vehicles
ArXiv | 2019
|Deep reinforcement-learning-based driving policy for autonomous road vehicles
IET | 2019
|A Deep Reinforcement Learning Driving Policy for Autonomous Road Vehicles
ArXiv | 2019
|Deep Learning & Autonomous Vehicles
TIBKAT | 2018
|Deep reinforcement‐learning‐based driving policy for autonomous road vehicles
Wiley | 2020
|