Robotic Waste Sorter with Agile Manipulation and Quickly Trainable Detector

TL;DR

This paper presents a robotic waste-sorting system that combines agile manipulation, quick training of neural networks, and methods to adapt to appearance differences, enabling efficient and robust waste detection and handling in recycling environments.

Contribution

It introduces a combined manipulation approach, a rapid data collection method for neural network training, and techniques to mitigate appearance differences between dataset and real-world scenes.

Findings

Enhanced waste detection accuracy over baseline methods.

Rapid training of neural networks with automatically collected images.

Successful robotic manipulation of waste items in a simulated environment.

Abstract

Owing to human labor shortages, the automation of labor-intensive manual waste-sorting is needed. The goal of automating waste-sorting is to replace the human role of robust detection and agile manipulation of waste items with robots. To achieve this, we propose three methods. First, we provide a combined manipulation method using graspless push-and-drop and pick-and-release manipulation. Second, we provide a robotic system that can automatically collect object images to quickly train a deep neural-network model. Third, we provide a method to mitigate the differences in the appearance of target objects from two scenes: one for dataset collection and the other for waste sorting in a recycling factory. If differences exist, the performance of a trained waste detector may decrease. We address differences in illumination and background by applying object scaling, histogram matching with histogram equalization, and background synthesis to the source target-object images. Via experiments in an indoor experimental workplace for waste-sorting, we confirm that the proposed methods enable quick collection of the training image sets for three classes of waste items (i.e., aluminum can, glass bottle, and plastic bottle) and detection with higher performance than the methods that do not consider the differences. We also confirm that the proposed method enables the robot quickly manipulate the objects.