Online Navigation Planning for Long-term Autonomous Operation of Underwater Gliders

2026-02-22

AI Summary

This study addresses the challenges of oceanic environmental uncertainty and large-scale fleet management faced by underwater gliders during long-term autonomous operations. The authors propose an online navigation planning method based on a stochastic shortest-path Markov decision process, which uniquely integrates a physics-informed simulator with Monte Carlo tree search. By leveraging ocean current forecasts and historical data, the approach generates environmental samples that account for control uncertainties and enables closed-loop replanning at each surfacing event. The method was validated through two field experiments in the North Sea, collectively spanning approximately three months and 1,000 kilometers. Results demonstrate a significant improvement in navigation efficiency over straight-line strategies, confirming the effectiveness and practicality of this sample-driven online planning framework for long-duration autonomous missions in dynamic marine environments.

Technology Category

• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Search and Optimization: Sampling/Simulation-based Search
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• Economics, Online Markets and Human Computation: Research challenges in human and human-AI computation
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Underwater glider robots have become an indispensable tool for ocean sampling. Although stakeholders are calling for tools to manage increasingly large fleets of gliders, successful autonomous long-term deployments have thus far been scarce, which hints at a lack of suitable methodologies and systems. In this work, we formulate glider navigation planning as a stochastic shortest-path Markov Decision Process and propose a sample-based online planner based on Monte Carlo Tree Search. Samples are generated by a physics-informed simulator that captures uncertain execution of controls and ocean current forecasts while remaining computationally tractable. The simulator parameters are fitted using historical glider data. We integrate these methods into an autonomous command-and-control system for Slocum gliders that enables closed-loop replanning at each surfacing. The resulting system was validated in two field deployments in the North Sea totalling approximately 3 months and 1000 km of autonomous operation. Results demonstrate improved efficiency compared to straight-to-goal navigation and show the practicality of sample-based planning for long-term marine autonomy.

Problem

• underwater gliders
• long-term autonomy
• online navigation planning
• ocean current uncertainty
• autonomous marine operations

Innovation

• Monte Carlo Tree Search
• stochastic shortest-path MDP
• physics-informed simulation
• online replanning
• autonomous underwater glider