Abstract
Waste management scene is in urgent need of robotic waste sorter. Nails and screws, as part of the construction waste scene, are hard to be found and can therefore, cause damage to the site's construction safety and increase the material loss. This paper presents a construction waste recycling robot. In order to complete the recycling tasks, robots are expected to inspect the entire working environment and identify the target objects. This research uses neural network technology to assist the robot patrol in an unknown work environment and to use faster R-CNN methods to find scattered nails and screws in real time, so that the robot can automatically recycle nails and screws. This study introduces computer vision technology and a full-coverage path-planning algorithm into the field of construction waste management and proposes a novel construction waste recycling approach. Based on this robot, we can continue our study of construction waste recycling robots that can automatically sort and recycle most construction waste in the future.
Original language | English |
---|---|
Pages (from-to) | 220-228 |
Number of pages | 9 |
Journal | Automation in Construction |
Volume | 97 |
DOIs | |
Publication status | Published - Jan 2019 |
Keywords
- Computer vision
- Construction waste management
- Faster R-CNN
- Mobile robot coverage
- Neural network
- Robotics in construction sites
ASJC Scopus subject areas
- Control and Systems Engineering
- Civil and Structural Engineering
- Building and Construction