What is it about?
Vehicles can follow roads based on a forward-looking camera, but this has to be done reliably in all circumstances. In daily traffic, they can encounter many unforeseen situations. Training for those situations in simulations should prepare them for such encounters, but this requires simulated worlds with enough complexity. In this paper we test the learned behavior of an autonomous vehicle for the first time in a recently introduced high-definition simulation map.
Featured Image
Photo by Maxim Hopman on Unsplash
Why is it important?
It is good to have a number of standard simulation benchmarks, but once reliable behavior is demonstrated on such benchmark, one should go further and explore new challenges. The high-definition simulation map used in this paper is such a challenge.
Perspectives
I would have expected that the realism of the map would make driving much harder. Yet, we found out that the type of situations encountered in such simulated world (for instance the layout of the crossings) are the real challenge, instead of how realistic the background is.
Dr. Arnoud Visser
Universiteit van Amsterdam
Read the Original
This page is a summary of: End-to-End Imitation Learning for Autonomous Vehicle Steering on a Single-Camera Stream, January 2022, Springer Science + Business Media,
DOI: 10.1007/978-3-030-95892-3_16.
You can read the full text:
Contributors
The following have contributed to this page
