mlsanityData to 14 Oct 2026

Physics-informed offline reinforcement learning eliminates catastrophic fuel waste in maritime routing

arXiv 2603.17319 ↗First posted 18 Mar 2026cs.AI cs.LG cs.ROcited by 1

On the map: Offline RL and imitation learning from data → Offline RL algorithms

Abstract (arXiv)

International shipping produces approximately 3% of global greenhouse gas emissions, yet voyage routing remains dominated by heuristic methods. We present PIER (Physics-Informed, Energy-efficient, Risk-aware routing), an offline reinforcement learning framework that learns fuel-efficient, safety-aware routing policies from physics-calibrated environments grounded in historical vessel tracking data and ocean reanalysis products, requiring no online simulator. Validated on one full year (2023) of AIS data across seven Gulf of Mexico routes (840 episodes per method), PIER reduces mean CO2 emissions by 10% relative to great-circle routing. However, PIER's primary contribution is eliminating catastrophic fuel waste: great-circle routing incurs extreme fuel consumption (>1.5x median) in 4.8% of voyages; PIER reduces this to 0.5%, a 9-fold reduction. Per-voyage fuel variance is 3.5x lower (p<0.001), with bootstrap 95% CI for mean savings [2.9%, 15.7%]. Partial validation against observed AIS vessel behavior confirms consistency with the fastest real transits while exhibiting 23.1x lower variance. Crucially, PIER is forecast-independent: unlike A* path optimization whose wave protection degrades 4.5x under realistic forecast uncertainty, PIER maintains constant performance using only local observations. The framework combines physics-informed state construction, demonstration-augmented offline data, and a decoupled post-hoc safety shield, an architecture that transfers to wildfire evacuation, aircraft trajectory optimization, and autonomous navigation in unmapped terrain.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Offline RL algorithms, one of 13 directions in Offline RL and imitation learning from data. Builds on 7 of its 40 references that are in the map. First posted on arXiv 18 Mar 2026.

Builds on

7 of 40 references.

Closest ancestors · 2

Same direction, newest first

  1. 2021Offline Reinforcement Learning with Implicit Q-Learning12 Oct
  2. 2020Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problems4 May

Other references · 5

  1. 2021Physics-Informed Neural Operator for Learning Partial Differential Equationscited by 527
  2. 2021Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learningcited by 371
  3. 2019Tackling Climate Change with Machine Learningcited by 265
  4. 2024From PINNs to PIKANs: Recent Advances in Physics-Informed Machine Learningcited by 73
  5. 2023Maximum diffusion reinforcement learningcited by 9

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026On the Limits of Univariate Deep Learning for Significant Wave Height Forecastingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Realistic Curriculum Reinforcement Learning for Autonomous and Sustainable Marine Vessel Navigation
  2. 2025Goal-Conditioned Reinforcement Learning for Data-Driven Maritime Navigation
  3. 2026Does Latent Context Help? A Controlled Evaluation of Inverse Reinforcement Learning in Arctic Shipping
  4. 2026EcoFair-CH-MARL: Scalable Constrained Hierarchical Multi-Agent RL with Real-Time Emission Budgets and Fairness Guarantees
  5. 2026Critic-Free Deep Reinforcement Learning for Maritime Coverage Path Planning on Irregular Hexagonal Grids
  6. 2025Khalasi: Energy-Efficient Navigation for Surface Vehicles in Vortical Flow Fields
  7. 2024Evaluating Robustness of Reinforcement Learning Algorithms for Autonomous Shipping
  8. 2025CH-MARL: Constrained Hierarchical Multiagent Reinforcement Learning for Sustainable Maritime Logistics