End-to-End Steering for Autonomous Vehicles via Conditional Imitation Co-Learning

Mahmoud Kishky, Hesham Eraqi, Khaled Elsayed

2024

Abstract

Autonomous driving involves complex tasks such as data fusion, object and lane detection, behavior prediction, and path planning. As opposed to the modular approach which dedicates individual subsystems to tackle each of those tasks, the end-to-end approach treats the problem as a single learnable task using deep neural networks, reducing system complexity and minimizing dependency on heuristics. Conditional imitation learning (CIL) trains the end-to-end model to mimic a human expert considering the navigational commands guiding the vehicle to reach its destination, CIL adopts specialist network branches dedicated to learn the driving task for each navigational command. Nevertheless, the CIL model lacked generalization when deployed to unseen environments. This work introduces the conditional imitation co-learning (CIC) approach to address this issue by enabling the model to learn the relationships between CIL specialist branches via a co-learning matrix generated by gated hyperbolic tangent units (GTUs). Additionally, we propose posing the steering regression problem as classification, we use a classification-regression hybrid loss to bridge the gap between regression and classification, we also propose using co-existence probability to consider the spatial tendency between the steering classes. Our model is demonstrated to improve autonomous driving success rate in unseen environment by 62% on average compared to the CIL method.

Paper Citation

in Harvard Style

Kishky M., Eraqi H. and Elsayed K. (2024). End-to-End Steering for Autonomous Vehicles via Conditional Imitation Co-Learning. In Proceedings of the 16th International Joint Conference on Computational Intelligence - Volume 1: NCTA; ISBN 978-989-758-721-4, SciTePress, pages 629-636. DOI: 10.5220/0013069900003837

in Bibtex Style

@conference{ncta24,
author={Mahmoud Kishky and Hesham Eraqi and Khaled Elsayed},
title={End-to-End Steering for Autonomous Vehicles via Conditional Imitation Co-Learning},
booktitle={Proceedings of the 16th International Joint Conference on Computational Intelligence - Volume 1: NCTA},
year={2024},
pages={629-636},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013069900003837},
isbn={978-989-758-721-4},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 16th International Joint Conference on Computational Intelligence - Volume 1: NCTA
TI - End-to-End Steering for Autonomous Vehicles via Conditional Imitation Co-Learning
SN - 978-989-758-721-4
AU - Kishky M.
AU - Eraqi H.
AU - Elsayed K.
PY - 2024
SP - 629
EP - 636
DO - 10.5220/0013069900003837
PB - SciTePress