Algorithm for Determining the Position of a Ship Hull-Cleaning Robot

Algorithm for Determining the Position of a Ship Hull-Cleaning Robot

The paper explores the development and evaluation of algorithms for the positioning of ship hull cleaning robots, focusing on machine learning and sensor fusion techniques. The research employs Gradient Boosting, Kalman filters, and deep learning to enhance the accuracy of robot positioning. Gradien...

Full description

Saved in:
Bibliographic Details
Main Authors: Konrad Chwełatiuk, Anna Kubina, Marcin Michalak, Marek Sikora, Piotr Ściegienka, Łukasz Wróbel
Format: Article
Language:English
Published: MDPI AG 2025-05-01
Series:Applied Sciences
Subjects:
ship hull cleaning
positioning
Kalman filter
odometry data
IMU data
Online Access:https://www.mdpi.com/2076-3417/15/10/5705
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The paper explores the development and evaluation of algorithms for the positioning of ship hull cleaning robots, focusing on machine learning and sensor fusion techniques. The research employs Gradient Boosting, Kalman filters, and deep learning to enhance the accuracy of robot positioning. Gradient Boosting is used to predict displacement vectors and rotation angles, while the Kalman filter is applied to refine position estimates by integrating odometry and GPS data. Deep learning models are utilized to predict robot trajectories based on sensor inputs. Experiments conducted on the Rosario dataset and simulated environments demonstrate the effectiveness of these methods.
ISSN:2076-3417

Similar Items