| Title: | Enhancing robotic bin-picking through machine vision: investigating the impact of detection speed and bin fill levels : preučevanje vpliva zaznave objektov ob različno napolnjenih zabojih |
|---|
| Authors: | ID Mambayil Ebrahimkutty, Suhaib Ebrahim (Author)
ID Hercog, Darko (Mentor) More about this mentor... 
ID Bencak, Primož (Comentor) |
|---|
| Files: |  UN_Mambayil_Ebrahimkutty_Suhaib_Ebrahim_2025.pdf (1,97 MB)
MD5: 37AB735FDECA09F010218A48BE962FAA
|
|---|
| Language: | English |
|---|
| Work type: | Bachelor thesis/paper |
|---|
| Typology: | 2.11 - Undergraduate Thesis |
|---|
| Organization: | FL - Faculty of Logistic
|
|---|
| Abstract: | Robotic bin-picking systems have become important in modern intralogistics as warehouses automate to address labour shortages and rising e-commerce demands. These systems operate through the coordinated use of (a) a robotic manipulator, (b) a 3D machine vision system, and (c) robotic grippers. When integrated with Automated Storage and Retrieval Systems (AS/RS), these systems offer an efficient approach to improving the order-picking performance.
In this study, the performance of a selected 3D machine vision system was evaluated based on two key parameters: detection mode and bin fill levels. A series of structured laboratory experiments was conducted using a UR5e collaborative robot equipped with three types of robotic grippers and a Pickit M-HD2 3D vision system. Objects of varying shapes and orientations were tested under different detection modes and bin fill levels. Faster detection modes reduced the processing time but resulted in more detection failures, especially with complex shapes or densely populated bins. In contrast, slower modes improved accuracy but increased the cycle time. Normal mode offered the best balance. The detection reliability decreased at higher bin fill levels and with irregularly shaped objects, due to occlusion and limited visibility. By analysing the detection time and successful detection, insights were gained into how appropriate detection configurations can improve both reliability and throughput in robotic bin-picking systems integrated with AS/RS. |
|---|
| Keywords: | intralogistics, Robotic Bin-Picking, machine vision system, object detection |
|---|
| Place of publishing: | Maribor |
|---|
| Place of performance: | Celje |
|---|
| Publisher: | [S. E. Mambayil Ebrahimkutty] |
|---|
| Year of publishing: | 2025 |
|---|
| Number of pages: | 1 spletni vir (1 datoteka PDF (XIX, 69 str., 3 str. pril.)) |
|---|
| PID: | 20.500.12556/DKUM-93700 |
|---|
| UDC: | 007.52:621.86(043.2) |
|---|
| COBISS.SI-ID: | 244446723 |
|---|
| Publication date in DKUM: | 31.07.2025 |
|---|
| Views: | 206 |
|---|
| Downloads: | 41 |
|---|
| Metadata: |    |
|---|
| Categories: | FL
|
|---|
|
:
|
Copy citation |
|---|
| | | | Average score: | (0 votes) |
|---|
| Your score: | Voting is allowed only for logged in users. |
|---|
| Share: |  |
|---|
Hover the mouse pointer over a document title to show the abstract or click
on the title to get all document metadata. |
