415 Parts Seen: Machine Vision Guides Robot

The vibratory bowl conveyor provides an economical and reliable method of presenting small parts in a given orientation. However, a key weakness of the vibratory bowl feeder is its use of mechanical channels and guides to orient parts. This limits it to handling just one part or one family of similar parts.

By Edited by Mark T. Hoske, editor in chief, Control Engineering, mhoske@reedbusine May 6, 2009

Machine Vision, May ’09, Control Engineering

Articles in the May 2009 Control Engineering North American print edition supplement:
– Motion and Vision Combine to Detect Flaws
– Intelligent Vision Stops Bypass of Quality Control
– The New Look of Facial Recognition

The vibratory bowl conveyor provides an economical and reliable method of presenting small parts in a given orientation. However, a key weakness of the vibratory bowl feeder is its use of mechanical channels and guides to orient parts. This limits it to handling just one part or one family of similar parts. Loading aerospace fasteners into a press seemed to be ideally suited for automation with a vibratory bowl feeder and a pick-and-place robot, but sorting the 415 parts involved would have required dozens of feeders.

Recent improvements in the speed, accuracy, and reliability of vision systems made it possible to use a vision sensor to replace channels and guides to deliver a flexible vibratory bowl feeder, says Stephen Harris, president of Rixan Associates. The design created the ability to handle a “virtually unlimited number of

Feeder presents parts on plenum in lower left of picture. (Parts used in this application cannot be shown, so are replaced by candy bars of similar sizes.

parts within a certain size range,” he says.

The patent-pending Rixan RFS-1000 flexible vibratory bowl feeder system, which costs about $45,000, provides a 20% improvement in productivity, and frees up one operator for other tasks, providing less than a one year return on investment for most users. It uses the vibratory bowl feeder concept to randomly present the parts in the bowl in any orientation. When the vision system views the right part in the right orientation, it stops the feeder and sends the part’s location to a robot, which loads the part into the press.

Here’s how it came about. An aerospace manufacturer wanted to automate part picking and placing because of the potential danger involved in having an operator reach into a stamping press. Additionally, the repetitive nature of the task and potential for hand and wrist injury required an operator to take frequent breaks, limiting productivity.

Rixan engineers, led by chief technical officer Mark Battisti, developed a unique, flexible feeding concept that uses a track on a 24-in. vibratory bowl without obstructions and cutouts. Every part that reaches the top of the bowl is presented onto a raised semi-transparent plenum so the vision system mounted above can identify the

Complete vision-enabled robotic work cell shows the Mitsubishi Electric RV-6SL-S11 robot in use.

type of each part and its orientation. Correct parts are picked by the robot; incorrect parts return to the bowl.

Rixan selected a Cognex In-Sight 5401 vision system and a Mitsubishi Electric RV-6SL-S11 robot that have been integrated through a joint development effort between Cognex and Mitsubishi Electric. Mitsubishi Melfa-Vision software incorporates robotic programming software and Cognex In-Sight Explorer software to set up the robot and the vision system simultaneously.

Rixan engineers chose the Mitsubishi Electric robot because of its long reach (97+380+425=902 mm) and speed of up to 6 meters per second. (Specifications for the 6-axis robot claim a pick-and-place cycle time of 0.37 sec, faster than most SCARA robots). Rixan selected the Cognex In-Sight 5401 vision system because it provides up to seven times the processing power of other models, and acquires up to 60 full 8-bit images per second.

The Cognex In-Sight 5401 vision system (top center) acquires up to 60 8-bit images per second.

To develop the application, Rixan engineers did the following:

  1. Ran the Melfa-Vision wizard that calibrates the robot in relation to the feeder.

  2. Put a part in the robot gripper in the ideal orientation for pickup.

  3. Jogged the robot over the feeder and set the part down on the plenum. This process taught the robot to pick up the part in a specific orientation at a specific point on the plenum.

  4. Selected a Melfa-Vision wizard that runs the Cognex PatFind object location tool and drew a rectangle around the part on the screen. The PatFind tool recognized the part in the zero angle of orientation. This process (which takes only a few minutes to complete) programs the feeder to handle one part.

Top-down lighting is used for parts requiring pattern recognition. White lighting is used for some parts, and glare-free polarized lighting for others. The feeding system is slightly slower than an operator working at full speed, but because it operates without breaks, it is said to save 20% in productivity compared to a human operator. The vision system also eliminates errors that can occur in manual feeding, which can damage tooling and cause downtime, Harris says.

ONLINE extra
Additional details about this application:

  • Old vibratory bowls : Traditional vibratory bowl feeders use a helical shelf-like inclined ramp winding upwards from the bottom of a bowl to a discharge passageway. The passageway’s reciprocating motion pushes parts up the ramp. Obstructions, cutouts and other orienting devices are arranged along the passageway to topple off parts that are improperly oriented. Only properly oriented parts go through. Vibratory bowl feeders usually handle one family of very similar parts. Each part in a family requires changing or adjusting the obstructions and orienting devices, which takes considerable time. Therefore, despite high efficiency and relatively low cost, the vibratory bowl concept is usually not seriously considered for applications involving large numbers of different parts.

  • Orientation: Parts in the wrong orientation are recycled back into the bowl and continue going around until they become properly oriented. The vision system can distinguish among parts, so operators do not have to clean the bowl when changing from one part to another. The parts remaining from the previous run will pass through the feeder without being used until the corresponding program runs again.

  • Integrated productivity: The integrated approach makes it easy to create vision-driven robotic applications with capabilities such as locating and inspecting parts that go far beyond what robots alone can accomplish, Rixan says. Melfa-Vision includes standard job programs supporting most common robot applications, enabling novices to easily develop robotic vision applications with up to three robots and seven vision sensors.

  • After programming , the vision system automatically recognizes the specific part as it moves across the plenum. At recognition, the system sends a signal to stop the feeder. It then acquires another image to determine the position of the part and calculates the angle of presentation. The robot moves to the location of the part, twists its wrist to match the part’s orientation, picks up the part, and sends a signal to restart the feeder after removing all “pickable” parts.

  • Part details, symmetry : In this application, most parts are symmetrical top to bottom so they can be identified from their silhouette with the plenum backlit. Some other parts must have a certain side facing upwards before loading onto the press. The Cognex PatMax tool learns which side is up after placing a part on the plenum in the proper orientation and selecting the PatMax feature from within Melfa-Vision.

  • Additional advantages, praise : Ease-of-setup, accuracy, and reliability of the vision sensors helped considerably, Harris says; “The customer is extremely pleased with the vision system performance as well as the tight integration between the vision sensors and robots.” The vision software is “much faster and more accurate than other comparable industry tools that we have seen,” Harris says. “Finally, the easy integration between Cognex vision sensors and Mitsubishi robots drastically reduces the amount of time required to develop a vision-enabled robot applications and in most cases eliminates the need for writing code. Vision-enabled robots now make it possible for the first time to apply efficient and economical vibratory bowl feeding to automate applications with many parts.”

  • Future : The fastener manufacturer ordered one Rixan feeding system and programmed it for a few parts; two more were ordered to handle the full range of parts.