Informed Reinforcement Learning for Situation-Aware Traffic Rule Exceptions

Daniel Bogdoll, Jing Qin, Moritz Nekolla, Ahmed Abouelazm, Tim Joseph +1 more
2/6/2024
cs.LGcs.CVcs.RO

No official code found. Best implementation: fzi-forschungszentrum-informatik/informed_rl (5 stars, MIT, last commit 2024-06-12).

Abstract

Reinforcement Learning is a highly active research field with promising advancements. In the field of autonomous driving, however, often very simple scenarios are being examined. Common approaches use non-interpretable control commands as the action space and unstructured reward designs which lack structure. In this work, we introduce Informed Reinforcement Learning, where a structured rulebook is integrated as a knowledge source. We learn trajectories and asses them with a situation-aware reward design, leading to a dynamic reward which allows the agent to learn situations which require controlled traffic rule exceptions. Our method is applicable to arbitrary RL models. We successfully demonstrate high completion rates of complex scenarios with recent model-based agents.

View on arXivView PDF

Code Implementations(1)

fzi-forschungszentrum-informatik/informed_rlCommunity100%
MIT

Informed Reinforcement Learning for Situation-Aware Traffic Rule Exceptions

52PythonJun 5, 20242 years agoMIT

Similar Papers

Scenario-Based Hierarchical Reinforcement Learning for Automated Driving Decision Making

86% match2025

Provable Traffic Rule Compliance in Safe Reinforcement Learning on the Open Sea

86% match2024

Diverse and Adaptive Behavior Curriculum for Autonomous Driving: A Student-Teacher Framework with Multi-Agent RL

86% match2025

Bootstrapping Reinforcement Learning with Sub-optimal Policies for Autonomous Driving

85% match2025

Safety-aware Causal Representation for Trustworthy Offline Reinforcement Learning in Autonomous Driving

85% match2023

Efficient Reinforcement Learning for Autonomous Driving with Parameterized Skills and Priors

85% match1 repo1082023

Cite this paper

@article{bogdoll2024informed,
  title  = {Informed Reinforcement Learning for Situation-Aware Traffic Rule Exceptions},
  author = {Daniel Bogdoll and Jing Qin and Moritz Nekolla and Ahmed Abouelazm and Tim Joseph and J. Marius Zöllner},
  year   = {2024},
  eprint = {2402.04168},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2402.04168v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2402.04168.svg)](https://codeofpaper.com/paper/2402.04168)

Discussion