# World-Model-Grounded LLM Planning for AUV and ASV Navigation Near Offshore Wind Farms

arXiv: https://arxiv.org/abs/2608.19661 · HTML: https://mlsanity.com/paper/2608.19661/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in World models for navigation, one of 11 directions in World models.
- Builds on 7 of its 27 references that are in the map.
- First posted on arXiv 20 Aug 2026.

## Placed in

- World models / [World models for navigation](https://mlsanity.com/wm/world-models-for-navigation/)

## Builds on (most cited first, 7 of 7)

- [World Models](https://mlsanity.com/paper/1803.10122/) (2018)
- [Stonefish: Supporting Machine Learning Research in Marine Robotics](https://mlsanity.com/paper/2502.11887/) (2025)
- [Distributed AI Agents for Cognitive Underwater Robot Autonomy](https://mlsanity.com/paper/2507.23735/) (2025)
- [Tethered Multi-Robot Systems in Marine Environments](https://mlsanity.com/paper/2508.02264/) (2025)
- [A Collaborative Reasoning Framework for Anomaly Diagnostics in Underwater Robotics](https://mlsanity.com/paper/2511.03075/) (2025)
- [OVerSeeC: Open-Vocabulary Costmap Generation from Satellite Images and Natural Language](https://mlsanity.com/paper/2602.18606/) (2026)
- [Mil-SCORE: Benchmarking Long-Context Geospatial Reasoning and Planning in Large Language Models](https://mlsanity.com/paper/2601.21826/) (2026)

## Abstract (arXiv)

Large language models can turn a natural-language mission into a sequence of robot actions, but they do not have a sense of physics: they cannot judge how long a command should run, or whether it will make the robot drift into an obstacle. We proposed the use of a world model to expand the capabilities of Large Language model-based planners. Our method has three components: a physics-grounded neural world model, a three-phase gradient-based trajectory optimizer, and a Model Predictive Controller (MPC)-style closed-loop replanner with a trust-region guard. The language model decides what to do, and the world model decides how long, whether that means driving eight thrusters through 6 DOF or two differential thrusters through 3 DOF. We evaluate two marine vehicle classes operating near offshore wind infrastructure: a 6-DOF Autonomous Underwater Vehicle (AUV) and a 3-DOF differential-drive Autonomous Surface Vehicle (ASV). In five benchmark missions per platform, both vehicles reach every goal with zero predicted collisions, and both transfer to GazeboSim under ocean current, waves, and thruster dynamics, remaining collision-free and cutting GazeboSim goal-distance error versus the ungrounded baseline by 70-82% (ASV) and roughly 93% (AUV), after a residual fine-tuning pass that separately reduces surrogate rollout Root Mean Square Error (RMSE) by 60% (AUV) and 69% (ASV). For the ASV we further demonstrate a Vision language model (VLM)-assisted semantic-mapping pipeline that extracts obstacles and environmental context from satellite imagery, nautical charts, and forecast Application Programming Interface (API) instead of onboard sensors, reaching 96% navigability accuracy as a drop-in replacement for hand-specified obstacle geometry.