Simulation platform tests AI fault diagnosis on underwater vehicles
A Simulation Platform for AUV Fault Recovery: Exploring LLM-Based Diagnostic Strategies
RoboticsArtificial Intelligence
Summary
Underwater robots need to fix themselves when things go wrong, but they can't always call for help. The authors created a system that uses language-based AI to figure out what the problem is and how to fix it. They tested this with many simulated failures to see how well different AI models worked. The best AI could suggest the right problem most of the time, while less advanced ones often guessed wrong too soon. This work helps improve how underwater robots can keep working on their own.
What this means in practice
- •For marine robotics developers: Test and improve fault diagnosis methods for underwater robots using a simulation system that integrates AI-based diagnostic reasoning.
- •For industrial automation engineers: Adapt an AI-supported fault recovery simulation approach to autonomous systems operating in communication-limited environments offshore.
Authors
Khalid Halba, Kylie Cooper, James G. Bellingham
Abstract
Autonomous underwater vehicles (AUVs) operating beyond reliable communications must recover from failures without human intervention. We investigate an architecture in which conventional deterministic layered control autonomy manages normal operations, while an invokable large language model (LLM) serves as a diagnostic and recovery planner when onboard anomaly detection identifies performance outside expected limits. Because language models are stochastic, rigorous evaluation requires ensemble testing rather than individual demonstrations. We present a closed-loop simulation architecture that couples real-time C vehicle software with a higher-level orchestration layer for physics-based fault injection, structured prompting, language-model interaction, mission file generation, validation, execution, and LLM-judge scoring. The framework, which we call SPAR (Simulation Platform for AUV Recovery), supports evaluation across fault realizations, prompt structures, reasoning models, and mission conditions. We vary these for a mass-shift fault over 480 SPAR trials, evaluating a frontier model and three off-the-shelf locally deployable LLMs. Model choice dominates diagnosis: the frontier model places the CG-shift mechanism in its top three hypotheses in 85-90% of trials, versus 60-78% for the best local model. Reasoning analysis indicates that local-model success is associated with following the complete diagnostic procedure, whereas weaker models often commit prematurely to elevator failure even though the actuator tracks its command. Diagnosis and operational decision performance do not appear to be coupled in this dataset. The contributions are an architecture extending unanticipated-fault recovery from detection to mitigation and an ensemble methodology for evaluating LLM-assisted mission management on low-power AUVs.