Work

Falsification-driven reinforcement learning for maritime motion planning

Year: 2026

Type: article

Abstract: Compliance with maritime traffic rules is essential for the safe operation of autonomous vessels, yet training reinforcement learning (RL) agents to adhere to them is challenging. The behavior of RL agents is shaped by the training scenarios they encounter, but creating scenarios that capture the complexity of maritime navigation is non-trivial, and real-world data alone is insufficient. To address this, we propose a falsification-driven RL approach that generates adversarial training scenarios in which the vessel under test violates maritime traffic rules, which are expressed as signal temporal logic specifications. Our experiments on open-sea navigation with two vessels demonstrate that the proposed approach provides more relevant training scenarios and achieves more consistent rule compliance. • Efficient falsification-driven RL: counterexamples from temporal logic rules. • Maritime traffic rules extended with temporal logic robustness measures. • Falsification-driven RL improves traffic rule compliance.

Source: Ocean Engineering

DOI: 10.1016/j.oceaneng.2026.125579

Authors: Marlon Müller, Florian Finkeldei, Hanna Krasowski, Murat Arcak, Matthias Althoff

Institutions: Munich Center for Machine Learning, Technical University of Munich, University of California, Berkeley

Language: English

Topic: Maritime Navigation and Safety

Subfield: Ocean Engineering

Field: Engineering

Domain: Physical Sciences

Open Access status: hybrid

Estimated APC paid: $4,150