---
title: "Cleaning the Seabed: How AI and NASA Technologies Help Locate Sunken Munitions"
description: "Researchers from the University of Miami have created a system based on NASA technologies and AI to search for sunken munitions. 🌊🤖 The new method allows drones to see through waves and distinguish shells from debris with 100% accuracy, even if they are covered in silt. 🚁💣"
date: 2026-07-03T14:40:00.000Z
lang: en
url: https://xab.info/en/posts/cleaning-the-seabed-how-ai-and-nasa-technologies-help-locate-sunken-munitions
tags: []
publisher: "XAB.info"
---

# Cleaning the Seabed: How AI and NASA Technologies Help Locate Sunken Munitions

![Military personnel in camouflage uses a tablet displaying a seabed map to locate sunken munitions using AI and NASA technologies](https://xab.info/media/2026/07/03/ii-i-tekhnologii-nasa-pomogayut-nahodit-zatonuvshie-boepripasy/ii-i-tekhnologii-nasa-pomogayut-nahodit-zatonuvshie-boepripasy-1.webp)

The seabed often hides dangerous secrets — sunken munitions left over from military operations or tests. Their search and neutralization is a complex task requiring not only courage but also advanced technologies. Now, researchers from the University of Miami are offering a revolutionary solution that combines NASA's optical developments with artificial intelligence.

### The Distortion Problem: Why Drones Can't "See" the Bottom

Traditional methods of surveying the seabed from the air face a fundamental physical problem. The constant movement of waves on the water surface causes light rays to distort, making it impossible to obtain a clear image of underwater objects. For ordinary drones and aircraft, this is an insurmountable obstacle.

To bypass this limitation, a team led by Veda Chirayath from the Airborne Center for Earth Sciences (ACES) developed a comprehensive system that allows them to "see through water".

### Technological Breakthrough: The Three Pillars of the New System

The success of the project is based on the synergy of three advanced technologies:

    - **Fluid Lensing.** This technology works in real-time, calculating and compensating for optical distortions caused by waves. As a result, scientists obtain ultra-high-definition 3D images of the seabed, as if the water layer between the camera and the object simply did not exist.

    - **MiDAR System.** This tool provides active multispectral illumination of objects at various wavelengths. It records the level of light reflection, allowing it to distinguish metallic and artificial surfaces from the natural background.

    - **Machine Learning.** The resulting arrays of multispectral RGB images are processed by a neural network. The AI is trained to analyze geometric shapes and spectral signatures to accurately distinguish real munitions from natural debris, rocks, or coral.

### Testing in Real-World Conditions

The effectiveness of the new system was tested in practice in the Florida Keys at the Broad Key experimental station. The test conditions were as close to reality as possible: scientists submerged munition models and decoy objects in shallow water. Scanning was conducted by drones over several weeks.

During this time, the submerged objects became covered with a layer of sediment and grew marine vegetation — a process known as biofouling. This simulated a situation where munitions have been lying on the bottom for years.

### Results: High Accuracy and Safety

Artificial intelligence demonstrated outstanding results. The system successfully identified absolutely all real explosive targets, clearly highlighting them on the final maps with red markers indicating exact coordinates.

The active scanning method of MiDAR played a special role, showing high recognition accuracy. At the same time, the system confirmed a low level of false alarms, ignoring safe decoys and camouflaged underwater background objects.

### Future Plans

Despite the success of the first tests, the authors of the study note that the work is not finished. The system awaits additional rounds of testing. Scientists need to verify how well the AI adapts to different types of seabed — for example, muddy or heavily overgrown areas — as well as to various calibers of historical munitions.

The development of this technology could become a key factor in protecting people and the environment from the consequences of past military conflicts, making the process of searching and neutralizing mined areas safer and more efficient.