AccuPickCase Study

Kitting and Assembly Line Automation

Customer

The customer is a renowned European automotive parts manufacturer. Their production line is dedicated exclusively to manufacturing air conditioners for a leading German carmaker.

Challenge

Fully Automated Kitting and Assembly Line with Limited Space and Complex Assembly Parts

The customer wanted a solution for setting up a fully automated kitting and assembly line. However, they faced challenges in arranging numerous robot arms in a limited workspace due to the bulky vibrating mechanisms and 2D vision systems, which typically require ample space.

Additionally, the customer had to handle a large number of intricate assembly parts, necessitating extensive custom mechanical design work and fine-tuning of the 2D vision system. The project demanded precise coordination among multiple robots on separate conveyors to execute ready-to-pick tasks simultaneously and swiftly.

Solution

AccuPick Streamlines Assembly Line Process and Optimizes Space Utilization

We seamlessly integrated three vision scanners with a single robot controller capable of parallel processing, optimizing space utilization while efficiently handling multiple component types on separate conveyors. Our solution transformed the assembly line process by eliminating the need for additional equipment, occupying just 1 sqm of space.

Equipped with AccuPick, Solomon’s AI-based 3D bin picking software, and SolScan, a structured light scanner, our solution excelled at accurately recognizing intricate objects. The robot could now effectively identify the correct pick-point among complex overlapping objects, meeting stringent tolerance requirements of less than 1 mm. This precise performance ensured flawless placement of the components.

Combining a high-speed structured light scanner with advanced AI algorithms, our smart bin picking system significantly reduced the cycle time to under 5 seconds for each component. The quick recognition capabilities, taking less than 2 seconds per part, contributed to this impressive efficiency.

Additionally, our multi-functional gripper design further streamlined operations by handling 2-3 different components simultaneously and accurately placing them in their designated assembly spot, enabling the robot to easily execute pick-and-place tasks, resulting in a further reduction of cycle time.

AccuPick 3D interface on a desktop monitor and a snapshot of the kitting and assembly lines in an automotive factory with a row of SolScan industrial 3D cameras and an ABB robot picking arm

Outcome

Optimized space utilization, occupying just 1 sqm of space for a fully automated assembly line
The vision system accurately recognized intricate objects and achieved precise component placement, meeting tolerance requirements of less than 1 mm
Reduced cycle time to under 5 seconds per component through quick recognition and multi-functional gripper design, improving overall efficiency and productivity
相關文章
  • AI影像辨識 – OCR電子元件字元

    電子元件製造過程追蹤為半導體之產出基石,辨識元件編號被視為生產重要環節之一。但較差環境下讓AOI辨識更加困難,對於提升產線效率以及降低字元的誤判度有很大改善空間。利用SolVision技術執行光學字元辨識,有別於傳統AOI,不受底色、環境光線及字元種類多等限制,可精準識別個別編碼。
  • 汽車鋼板烤漆AI瑕疵檢測

    透過AI視覺技術檢測汽車鋼板烤漆上的瑕疵,加速整體效率,提升產線生產力。
  • SMT製程的回焊短路檢測解決方案

    SMT製程回焊過程中,過多錫膏量或是印刷偏移可能導致錫球間短路,過去以人工方式檢測,效率不彰。SMT多餘錫膏在高溫下的流動型態無法預測,難以傳統AOI檢出。運用SolVision AI技術,將SMT製程影像樣本中的回焊短路瑕疵定位並標註,訓練AI模型。可輕易檢出錫球間短路情形。
  • close-up of PCB assembly

    使用 AI 進行 PCB 組裝驗證

    META-aivi 為領先的IPC 製造商優化 PCB 組裝和偵測,透過 AR + AI 驗證提高精度和效率。