gearwheels of a car engine valve train

AccuPickCase Study

Automated Gear Loading Using AI


The customer is a large automotive component manufacturer based in Asia.


Automated Loading in Automotive Component Manufacturing

Automotive component manufacturers have traditionally relied on manual labor for loading tasks. This approach is labor-intensive, particularly due to the large quantity and weight of metal parts like gears. As a result, factors such as site conditions and worker capabilities can lead to fluctuations in loading efficiency. To address these challenges, auto component manufacturers are increasingly adopting automated loading systems to enhance production efficiency and reduce risks associated with manual labor methods.

gearwheels of a car engine valve train


Limitations of Ordinary Machine Vision Systems

Two major technologies currently dominate factory automation: (1) the use of industrial cameras placed within the operating area and (2) the use of industrial cameras mounted on robotic arms. These cameras precisely ascertain the relative positions of objects, facilitating the guidance of robotic arms in performing loading and unloading tasks. However, challenges arise in environments where illuminating objects from the side is hindered by environmental lighting interference. Additionally, issues like small gear size, reflective materials, oil contamination, and disordered stacking of workpieces in the bin present significant challenges for ordinary machine vision systems, resulting in inefficiencies in robot arm grasping.


Automated Gear Loading with AccuPick

Using industrial cameras and advanced AI, AccuPick learns and classifies gear stacking conditions and angles from a small number of image samples. AccuPick’s motion planning module swiftly calculates the optimal robot path, allowing the robotic arm to accurately pick gears from different angles, avoiding collisions with the bin.


Precise picking and collision avoidance ensured improved efficiency
Automation enabled continuous operation for greater productivity
Occupational health risks were reduced
More Case
  • cable and connector for USB, Type-C, Micro USB, Lightning, on a white background in isolation, collage

    AI Recognition and Classification of Connectivity Cables

    META-aivi transforms cable inspection with AI, ensuring precision in recognition, classification, and defect detection for efficient electronics manufacturing.
  • Inspecting Packaging Seals

    Based on deep learning technology, SolVision can determine whether a seal is intact by comparing a perfect packaging with potential variations
  • high quality Galvanized steel pipe or Aluminum and chrome stainless pipes in stack waiting for shipment in warehouse

    Parts Inventory Management Using AR + AI

    META-aivi uses artificial intelligence to accurately identify size variations of the pipes instantaneously. Using the fast counting feature, all the different parts are classified and counted within seconds.
  • a close up of a blue and yellow background

    Automated Visual Inspection for LEDs

    Streamlined LED visual inspection with AI SolVision features an Instance Segmentation tool designed to identify various mini LED defects through image processing.