Case Study · Automotive parts

Automotive Arm Part Feeding That People Used to Do — 10–12 pcs/min on the Same Feeder Even as Variants Change

Manual feeding of automotive arm parts was automated with UNI FEEDER 5.0, RoboEye machine vision and an EPSON 6-axis robot, feeding and inserting at 10–12 pcs/min on a configuration built for a line whose variants keep changing.

Feed and insertion rate Automotive arm parts
10–12 pcs/min
Feeding and insertion process previously done by hand
Manual work replaced

Automotive Arm Part Auto-Feeding | EPSON 6-Axis Robot × UNI Flexible Feeder

Watch on YouTube →

What and how

Configuration of this case

Part
Automotive arm parts — many variants
Process
Part feeding → position and orientation recognition → pick-and-place · insertion
Robot
EPSON 6-axis robot
Feeder · Vision
UNI FEEDER 5.0 + RoboEye machine vision
Previous method
Manual feeding and insertion by hand
Industry
Automotive parts
Automotive Arm Part Feeding That People Used to Do — 10–12 pcs/min on the Same Feeder Even as Variants Change — part and detection view

Challenge

Parts with many variants, and more to come, were being picked and loaded one by one by hand

Feeding and inserting automotive arm parts was a process done by hand. There were many variants, and the production items were going to keep changing. What was needed was not equipment that handles only the part being made today, but a feeding method that could take on parts of different shapes on the same equipment.

Solution

RoboEye machine vision reads the position and orientation of the part, and the EPSON 6-axis robot picks at those coordinates

UNI FEEDER 5.0, RoboEye machine vision and an EPSON 6-axis robot were integrated into a single flow. The feeder spreads the automotive arm parts on the feed surface, vision recognizes the position and orientation of each part in real time, and the robot picks at those coordinates and inserts the part. Because there is no assumption that parts are loaded in a predetermined pose, parts of different shapes are handled with the same configuration.

  • Feeding — UNI FEEDER 5.0
  • Recognition — RoboEye machine vision (real-time position and orientation recognition)
  • Handling — EPSON 6-axis robot pick-and-place · insertion

Results

Manual feeding and insertion replaced at 10–12 pcs/min, and no equipment rebuild when variants change

  • Feed and insertion rate — automotive arm parts are fed and inserted at 10–12 pcs/min.
  • Labor replacement — the feeding and insertion process previously done by hand was replaced.
  • Variant handling — parts of different shapes are handled with the same feeder and the same vision configuration.
  • Future changes — the configuration was built on the premise that production items keep changing.

FAQ

Frequently asked about this case

What rate do you get when manual part feeding is automated?

This case feeds and inserts automotive arm parts at 10–12 pcs/min. It depends on the size and shape of the part and the gripping method, so verification against the target part is needed.

Can a flexible feeder handle parts with complex shapes, like automotive arm parts?

This case is exactly that. UNI FEEDER 5.0 spreads the parts on the feed surface, and RoboEye machine vision recognizes the position and orientation of each part and passes pick coordinates to the EPSON 6-axis robot. There is no assumption that parts are loaded in a predetermined pose.

Can it still be used when the production items keep changing?

This case started under exactly that condition. There were many variants and products would keep changing, so instead of equipment that handles only today's part, it was configured so the same feeder and the same robot take on parts whose shape differs.

Can it pick with the part's orientation taken into account?

Yes. RoboEye machine vision recognizes the position and orientation of the part together in real time and passes those values as robot coordinates. The robot performs pick-and-place at coordinates that reflect the orientation.

Can AIM vision and feeders be integrated with EPSON robots?

This case was built with an EPSON 6-axis robot. The engineering record of the integration through the EPSON RC+ plugin is documented in a separate research note.