TreeIRL: Safe Urban Driving with Tree Search and Inverse Reinforcement Learning

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Tomov, Momchil S., Lee, Sang Uk, Hendrago, Hansford, Huh, Jinwook, Han, Teawon, Howington, Forbes, da Silva, Rafael, Bernasconi, Gianmarco, Heim, Marc, Findler, Samuel, Ji, Xiaonan, Boule, Alexander, Napoli, Michael, Chen, Kuo, Miller, Jesse, Floor, Boaz, Hu, Yunqing
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911230745116672
author Tomov, Momchil S.
Lee, Sang Uk
Hendrago, Hansford
Huh, Jinwook
Han, Teawon
Howington, Forbes
da Silva, Rafael
Bernasconi, Gianmarco
Heim, Marc
Findler, Samuel
Ji, Xiaonan
Boule, Alexander
Napoli, Michael
Chen, Kuo
Miller, Jesse
Floor, Boaz
Hu, Yunqing
author_facet Tomov, Momchil S.
Lee, Sang Uk
Hendrago, Hansford
Huh, Jinwook
Han, Teawon
Howington, Forbes
da Silva, Rafael
Bernasconi, Gianmarco
Heim, Marc
Findler, Samuel
Ji, Xiaonan
Boule, Alexander
Napoli, Michael
Chen, Kuo
Miller, Jesse
Floor, Boaz
Hu, Yunqing
contents We present TreeIRL, a novel planner for autonomous driving that combines Monte Carlo tree search (MCTS) and inverse reinforcement learning (IRL) to achieve state-of-the-art performance in simulation and in real-world driving. The core idea is to use MCTS to find a promising set of safe candidate trajectories and a deep IRL scoring function to select the most human-like among them. We evaluate TreeIRL against both classical and state-of-the-art planners in large-scale simulations and on 500+ miles of real-world autonomous driving in the Las Vegas metropolitan area. Test scenarios include dense urban traffic, adaptive cruise control, cut-ins, and traffic lights. TreeIRL achieves the best overall performance, striking a balance between safety, progress, comfort, and human-likeness. To our knowledge, our work is the first demonstration of MCTS-based planning on public roads and underscores the importance of evaluating planners across a diverse set of metrics and in real-world environments. TreeIRL is highly extensible and could be further improved with reinforcement learning and imitation learning, providing a framework for exploring different combinations of classical and learning-based approaches to solve the planning bottleneck in autonomous driving.
format Preprint
id arxiv_https___arxiv_org_abs_2509_13579
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle TreeIRL: Safe Urban Driving with Tree Search and Inverse Reinforcement Learning
Tomov, Momchil S.
Lee, Sang Uk
Hendrago, Hansford
Huh, Jinwook
Han, Teawon
Howington, Forbes
da Silva, Rafael
Bernasconi, Gianmarco
Heim, Marc
Findler, Samuel
Ji, Xiaonan
Boule, Alexander
Napoli, Michael
Chen, Kuo
Miller, Jesse
Floor, Boaz
Hu, Yunqing
Robotics
Artificial Intelligence
Machine Learning
We present TreeIRL, a novel planner for autonomous driving that combines Monte Carlo tree search (MCTS) and inverse reinforcement learning (IRL) to achieve state-of-the-art performance in simulation and in real-world driving. The core idea is to use MCTS to find a promising set of safe candidate trajectories and a deep IRL scoring function to select the most human-like among them. We evaluate TreeIRL against both classical and state-of-the-art planners in large-scale simulations and on 500+ miles of real-world autonomous driving in the Las Vegas metropolitan area. Test scenarios include dense urban traffic, adaptive cruise control, cut-ins, and traffic lights. TreeIRL achieves the best overall performance, striking a balance between safety, progress, comfort, and human-likeness. To our knowledge, our work is the first demonstration of MCTS-based planning on public roads and underscores the importance of evaluating planners across a diverse set of metrics and in real-world environments. TreeIRL is highly extensible and could be further improved with reinforcement learning and imitation learning, providing a framework for exploring different combinations of classical and learning-based approaches to solve the planning bottleneck in autonomous driving.
title TreeIRL: Safe Urban Driving with Tree Search and Inverse Reinforcement Learning
topic Robotics
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2509.13579