loading
Papers Papers/2022 Papers Papers/2022

Research.Publish.Connect.

Paper

Authors: Yann Koeberle 1 ; 2 ; Stefano Sabatini 1 ; Dzmitry Tsishkou 1 and Christophe Sabourin 2

Affiliations: 1 IoV team, Paris Research Center, Huawei Technologies, France ; 2 Univ. Paris Est Creteil, LISSI, F-77567 Lieusaint, France

Keyword(s): Driving Simulation, Learning from Demonstrations, Adversarial Imitation Learning.

Abstract: Traffic simulation has gained a lot of interest for autonomous driving companies for qualitative safety evaluation of self driving vehicles. In order to improve self driving systems from synthetic simulated experiences, traffic agents need to adapt to various situations while behaving as a human driver would do. However, simulating realistic traffic agents is still challenging because human driving style cannot easily be encoded in a driving policy. Adversarial Imitation learning (AIL) already proved that realistic driving policies could be learnt from demonstration but mainly on highways (NGSIM Dataset). Nevertheless, traffic interactions are very restricted on straight lanes and practical use cases of traffic simulation requires driving agents that can handle more various road topologies like roundabouts, complex intersections or merging. In this work, we analyse how to learn realistic driving policies on real and highly interactive driving scenes of Interaction Dataset based on AI L algorithms. We introduce a new driving policy architecture built upon the Lanelet2 map format which combines a path planner and an action space in curvilinear coordinates to reduce exploration complexity during learning. We leverage benefits of reward engineering and variational information bottleneck to propose an algorithm that outperforms all AIL baselines. We show that our learning agent is not only able to imitate humane like drivers but can also adapts safely to situations unseen during training. (More)

CC BY-NC-ND 4.0

Sign In Guest: Register as new SciTePress user now for free.

Sign In SciTePress user: please login.

PDF ImageMy Papers

You are not signed in, therefore limits apply to your IP address 35.175.201.191

In the current month:
Recent papers: 100 available of 100 total
2+ years older papers: 200 available of 200 total

Paper citation in several formats:
Koeberle, Y.; Sabatini, S.; Tsishkou, D. and Sabourin, C. (2022). Learning Human-like Driving Policies from Real Interactive Driving Scenes. In Proceedings of the 19th International Conference on Informatics in Control, Automation and Robotics - ICINCO; ISBN 978-989-758-585-2; ISSN 2184-2809, SciTePress, pages 419-426. DOI: 10.5220/0011268400003271

@conference{icinco22,
author={Yann Koeberle. and Stefano Sabatini. and Dzmitry Tsishkou. and Christophe Sabourin.},
title={Learning Human-like Driving Policies from Real Interactive Driving Scenes},
booktitle={Proceedings of the 19th International Conference on Informatics in Control, Automation and Robotics - ICINCO},
year={2022},
pages={419-426},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011268400003271},
isbn={978-989-758-585-2},
issn={2184-2809},
}

TY - CONF

JO - Proceedings of the 19th International Conference on Informatics in Control, Automation and Robotics - ICINCO
TI - Learning Human-like Driving Policies from Real Interactive Driving Scenes
SN - 978-989-758-585-2
IS - 2184-2809
AU - Koeberle, Y.
AU - Sabatini, S.
AU - Tsishkou, D.
AU - Sabourin, C.
PY - 2022
SP - 419
EP - 426
DO - 10.5220/0011268400003271
PB - SciTePress