Domestic waste detection and grasping points for robotic picking up

TL;DR

This paper develops a deep learning-based AI system using Mask-RCNN to detect and classify waste objects in indoor and outdoor environments, facilitating robotic grasping for recycling purposes.

Contribution

It introduces a new waste dataset and a complete pipeline combining 2D detection, 3D shape estimation, and grasp point computation for robotic waste collection.

Findings

Effective waste detection in diverse environments

Accurate 3D shape reconstruction for grasping

Enhanced recycling classification strategy

Abstract

This paper presents an AI system applied to location and robotic grasping. Experimental setup is based on a parameter study to train a deep-learning network based on Mask-RCNN to perform waste location in indoor and outdoor environment, using five different classes and generating a new waste dataset. Initially the AI system obtain the RGBD data of the environment, followed by the detection of objects using the neural network. Later, the 3D object shape is computed using the network result and the depth channel. Finally, the shape is used to compute grasping for a robot arm with a two-finger gripper. The objective is to classify the waste in groups to improve a recycling strategy.