Detection and Collection of Waste Using a Partially Submerged Aquatic Robot
Author | Ayesh, Malek |
Author | Qidwai, Uvais |
Available date | 2024-05-07T05:39:55Z |
Publication Date | 2023 |
Publication Name | Lecture Notes in Networks and Systems |
Resource | Scopus |
Identifier | http://dx.doi.org/10.1007/978-3-031-16075-2_9 |
ISSN | 23673370 |
Abstract | With the amount of waste being dispersed into oceans on the rise, mitigating this issue has become a global concern. In the past few decades, governments, scientists, organizations, and individuals have been attempting to attenuate the effects of global warming, partially caused by improper waste disposal into oceans. This study presents a solar powered, partially submerged aquatic robot constructed from recycled, recyclable, upcycled, and sustainable materials. The robot aims to provide flexibility in the choice of construction materials by not being limited to what operating system, microcontroller, motors, and robot floaters are used. This robot detects and collects seven different categories of commonly littered waste namely cardboard (95.3%), wrappers (94.1%), metal cans (93.8%), surgical face masks (93.2%), plastic bags (96.2%), polystyrene (92.6%), and plastic bottles (93.8%). The custom detection system was evaluated based on whether it is capable of detecting waste and how well if little, medium, and high movement was introduced to the robot. Furthermore, the detection system's performance in low light situations along with the drivetrain's effectiveness was tested. Future improvements include forming larger dataset, enhancing the detection system's low light capabilities, and attaching a larger battery. |
Language | en |
Publisher | Springer Science and Business Media Deutschland GmbH |
Subject | Object detection Recycling Robotics Sustainability YOLOv3 |
Type | Conference |
Pagination | 133-149 |
Volume Number | 544 LNNS |
Files in this item
Files | Size | Format | View |
---|---|---|---|
There are no files associated with this item. |
This item appears in the following Collection(s)
-
Computer Science & Engineering [2402 items ]