Object detection and graspability analysis for robotic bin-picking application in intralogistics

Primož Bencak (Author), Darko Hercog (Author), Tone Lerher (Author)

Abstract

Robotics has been gaining attention in intralogistics applications in recent years. Automation of intralogistics processes aims to cope with the rising trends of workforce deficiency, aging, and increasing demands that came with the rise of E-commerce. Many improvements aim at bin-picking applications since order-picking requires most contributions while adding little to the products' value. Robotic bin-pickers are showing promising results; however, they are still subject to many limitations. First, the vision system must correctly determine the object's location and orientation. Second, a correct robotic gripper must be chosen. Lastly, appropriate grasping points that lead to successful picking must be selected. In this paper, we explore the influencing parameters of object detection using a 3D vision system. Second, we analyze an actual bin-picking application to determine the most appropriate selection of the robotic gripper. Based on the experiments, we provide the guidelines for selecting the most appropriate robotic bin-picking configuration.

Keywords

intralogistics;robotic bin-picking;detection analysis;graspability analysis;

Data

Language:	English
Year of publishing:	2023
Typology:	1.08 - Published Scientific Conference Contribution
Organization:	UM FL - Faculty of Logistics
Publisher:	University of Zagreb, Faculty of Mechanical Engineering and Naval Architecture
UDC:	658:007.52
COBISS:	154849795
Views:	134
Downloads:	5
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	Slovenian
Secondary keywords:	intralogistika;robotsko komisioniranje;analiza detekcije;analiza oprijemljivosti;
Type (COBISS):	Other
Pages:	Str. 7-14
ID:	19638692