A BlueROV2-based platform for underwater mapping experiments

TL;DR

This paper introduces a cost-effective underwater mapping platform using a BlueROV2 ROV, combining sensor fusion and neural networks for pose estimation and object detection, validated through experiments in a controlled pool environment.

Contribution

It presents a novel low-cost platform integrating sensor fusion and deep learning for underwater mapping and object detection validation.

Findings

Successful pose estimation with extended Kalman filter

Effective object detection using deep neural networks

Validated mapping accuracy in controlled experiments

Abstract

We propose a low-cost laboratory platform for development and validation of underwater mapping techniques, using the BlueROV2 Remotely Operated Vehicle (ROV). Both the ROV and the objects to be mapped are placed in a pool that is imaged via an overhead camera. In our prototype mapping application, the ROV's pose is found using extended Kalman filtering on measurements from the overhead camera, inertial, and pressure sensors; while objects are detected with a deep neural network in the ROV camera stream. Validation experiments are performed for pose estimation, detection, and mapping. The litter detection dataset and code are made publicly available.