Analyzing Closed-loop Training Techniques for Realistic Traffic Agent Models in Autonomous Highway Driving Simulations

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Bitzer, Matthias, Cimurs, Reinis, Coors, Benjamin, Goth, Johannes, Ziesche, Sebastian, Geiger, Philipp, Naumann, Maximilian
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914981978570752
author Bitzer, Matthias
Cimurs, Reinis
Coors, Benjamin
Goth, Johannes
Ziesche, Sebastian
Geiger, Philipp
Naumann, Maximilian
author_facet Bitzer, Matthias
Cimurs, Reinis
Coors, Benjamin
Goth, Johannes
Ziesche, Sebastian
Geiger, Philipp
Naumann, Maximilian
contents Simulation plays a crucial role in the rapid development and safe deployment of autonomous vehicles. Realistic traffic agent models are indispensable for bridging the gap between simulation and the real world. Many existing approaches for imitating human behavior are based on learning from demonstration. However, these approaches are often constrained by focusing on individual training strategies. Therefore, to foster a broader understanding of realistic traffic agent modeling, in this paper, we provide an extensive comparative analysis of different training principles, with a focus on closed-loop methods for highway driving simulation. We experimentally compare (i) open-loop vs. closed-loop multi-agent training, (ii) adversarial vs. deterministic supervised training, (iii) the impact of reinforcement losses, and (iv) the impact of training alongside log-replayed agents to identify suitable training techniques for realistic agent modeling. Furthermore, we identify promising combinations of different closed-loop training methods.
format Preprint
id arxiv_https___arxiv_org_abs_2410_15987
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Analyzing Closed-loop Training Techniques for Realistic Traffic Agent Models in Autonomous Highway Driving Simulations
Bitzer, Matthias
Cimurs, Reinis
Coors, Benjamin
Goth, Johannes
Ziesche, Sebastian
Geiger, Philipp
Naumann, Maximilian
Robotics
Artificial Intelligence
Machine Learning
Multiagent Systems
68T07
I.2.6; I.6.5
Simulation plays a crucial role in the rapid development and safe deployment of autonomous vehicles. Realistic traffic agent models are indispensable for bridging the gap between simulation and the real world. Many existing approaches for imitating human behavior are based on learning from demonstration. However, these approaches are often constrained by focusing on individual training strategies. Therefore, to foster a broader understanding of realistic traffic agent modeling, in this paper, we provide an extensive comparative analysis of different training principles, with a focus on closed-loop methods for highway driving simulation. We experimentally compare (i) open-loop vs. closed-loop multi-agent training, (ii) adversarial vs. deterministic supervised training, (iii) the impact of reinforcement losses, and (iv) the impact of training alongside log-replayed agents to identify suitable training techniques for realistic agent modeling. Furthermore, we identify promising combinations of different closed-loop training methods.
title Analyzing Closed-loop Training Techniques for Realistic Traffic Agent Models in Autonomous Highway Driving Simulations
topic Robotics
Artificial Intelligence
Machine Learning
Multiagent Systems
68T07
I.2.6; I.6.5
url https://arxiv.org/abs/2410.15987