Development of an automatic sorting robot for construction and demolition waste

Abstract

An automatic sorting robot is designed in this report. The system makes use of height maps and near-infrared (NIR) hyperspectral images to locate the ROI of objects and to do online statistic pixel-based classification in contours. This approach has two advantages: (1) to generate training data for sorting without manual work; (2) to get more stable final result. Two kind of features in hyperspectral image were extracted, a scale-sensitive algorithm was used to identify amplitude feature and a scale-insensitive algorithm was used to identify trend feature. After location and classification, the robot grabs valuable targets based on their position and posture and places them into the corresponding recycling area based on their category. The prototype machine can automatically sort construction and demolition waste (C&DW) with a size range of 0.05–0.5 m. The sorting efficiency can reach 2028 picks/h, and the online recognition accuracy nearly reaches 100%.