HorizonNet uses CNNs and DEM data for GPS-level USV navigation

Researchers have developed HorizonNet, a novel system for estimating the position of unmanned surface vessels (USVs) in coastal and archipelago environments. The system utilizes a Convolutional Neural Network (CNN) to extract horizon lines from panoramic images, compensating for camera orientation. This processed image is then used to correlate with digital elevation model (DEM) data via a MOSSE correlation filter, ultimately determining the USV's location with GPS-level accuracy. AI

IMPACT This research could improve autonomous navigation systems for marine vessels, potentially leading to more efficient and safer operations in complex environments.

RANK_REASON This is a research paper detailing a new method for navigation. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• Andreas Robinson
• arXiv
• CNN
• digital elevation model
• Global Positioning System
• HorizonNet

• paper
• product
• infra

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →