We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
SeaTrekker: a dual-mode USV for autonomous floating debris collection using YOLO and UWB localisation
AI summary Read the abstract
Researchers built a robotic boat called SeaTrekker that uses smart cameras to spot and scoop up floating trash like bottles and cans from the water. In tests, it successfully collected debris both by remote control and on its own, offering a safer, low-risk way to clean up plastic pollution before it breaks down into microplastics that can contaminate seafood and drinking water. While still a prototype tested in a pool rather than the open ocean, this technology points toward more efficient ways to protect marine ecosystems and, by extension, the food and water humans depend on.
Growing plastic bottle consumption and insufficient waste management are accelerating the accumulation of floating plastic debris in the UAE seas, which causes degradation of marine ecosystems and threatens sustainability, livelihoods, and essential coastal resources. Eliminating the floating debris from the water surface will clear the waterways, improve marine life, upgrade water quality, reduce the burden of manual cleaning and improve environmental, ecological and public health. The objective of this work is to design an Unmanned Surface Vehicle (USV) to collect the floating waste from the water surface to offer efficient, low-risk alternatives to manned operations in the challenging aquatic environments. SeaTrekker operates in both tele-operated and autonomous modes, with a curved catamaran-style dual-hull design inspired by Bernoulli’s principle, to optimises hydrodynamic flow while guiding debris into a passive net-based intake system. It is integrated with YOLO-based object detection, Intel RealSense depth sensing, and UWB localisation to identify and retrieve floating objects such as soda cans, water bottles, wood piece, plastic bags and rubber items. Its modular aluminium structure and 3D-printed ABS hulls support sustainability and ease of assembly. Initially challenges such as water leakage and localisation drift were addressed with targeted improvements. During controlled testing in a 20 m × 10 m swimming pool environment, SeaTrekker collected 22 floating cans per 15-minute trial in tele-operated mode (average 1.47 cans/min; peak 10 cans/min). In autonomous mode, it retrieved 9 floating cans per 15-minute trial (average 0.60 cans/min; peak 2 cans/min). These capabilities will help remove floating debris effectively. Future enhancements will focus on improving autonomy, durability, and adaptability through measures such as waterproofing, GPS integration, safer battery systems, and expanding its dataset. These improvements will position SeaTrekker for real-world marine clean-up missions.
More Papers Like This
Smart Floating Waste Collector
AI summary Read the abstract
Researchers built a solar-powered floating robot that uses YOLO-based deep learning to detect plastic bottles, polythene bags, and aluminum cans on water surfaces in real time and automatically collects them via a motorized conveyor, demonstrating feasibility as a low-maintenance aquatic waste management prototype.
The SeaClear system: An intelligent multi-robot solution for autonomous cleanup of marine debris on the seabed
AI summary Read the abstract
The SeaClear multi-robot system successfully demonstrated autonomous detection and collection of underwater marine debris using YOLO-based computer vision, validating AI-driven robotics as a scalable alternative to manual ocean cleanup. Automated debris collection technology is directly relevant to the microplastic problem because it could accelerate removal of macroplastic items before they fragment into the microplastic particles that are far harder to retrieve from the marine environment.
Smart Ocean Cleanup: An AI-Integrated Autonomous System for Marine Waste Management
AI summary Read the abstract
This paper presents an AI-powered autonomous boat system designed to detect and collect marine pollution — including plastics, oil spills, and microplastics — using deep learning image classification, IoT sensors, and robotic collection mechanisms. The system demonstrated over 94% accuracy for pollutant detection and classification across several AI models. While focused more broadly on ocean cleanup technology than on microplastic science specifically, it demonstrates how AI-integrated robotics could help address the practical challenge of removing plastic waste from ocean surfaces before it breaks down further.
“WAVECLEAN” – An Innovation in Autonomous Vessel Driving Using Object Tracking and Collection of Floating Debris
AI summary Read the abstract
Researchers designed an autonomous vessel called WAVECLEAN that uses object-tracking technology to identify and collect floating marine debris, including plastics. The system combines camera-based detection with machine learning to navigate waterways and gather waste without human operation. The study demonstrates a technology-driven approach to addressing plastic pollution in harbors, rivers, and coastal areas.
Real-time detection of macroplastic pollution in inland waters: development of a lightweight image recognition system
AI summary Read the abstract
This study developed a lightweight YOLOv8-based real-time detection system for macroplastic pollution in inland waters, optimized for deployment on fixed camera systems and UAVs. Pre-processing techniques including image tiling and blurring improved recall and mAP, and the system supports both decentralized on-site and centralized processing configurations.
Research digests by email
When a large batch of papers lands in the Atlas, we read through it and send a short write-up of what stood out.