High repeatability is a prerequisite in any highly functional manufacturing operation. In machine vision applications, being able to select parts off a conveyor belt in the same way with the same precision each time is vital. To achieve repeatability, many manufacturers have put robots to the task. But accommodating the robot so that the job is done properly can be a cumbersome process.For example, if the job requires the robot to pick up and place a product in a given location, the manufacturer has to construct a fixture that would hold the item in a repeatable position, or, in some cases, the product may have to be moved to the fixture so that the robot can perform the given task.But a robot that can "see" would eliminate the need for this time-consuming, costly preparation. It would be able to make decisions based on sensory feedback from its environment. Just such a robot was what one supplier of building products was looking for. The company has 42 products, such as concrete paving slabs, in all sizes, colors and shapes, and moving them from one place to another was problematic. To move a square concrete paving slab onto a pallet, four paddles had to be used to center it, but if the slab was a circle, a triangle or a rectangle, the job was not so easy.Once the control panel or teach pad for the vision-enabled robot is programmed, it can be used by an untrained operator.The company now uses a "vision-enabled" robot based on technology developed by Kuka Automation + Robotics of West Midlands, UK, and DVT Corp.'s UK facility in Milton Keynes. The vision system connects directly to a Kuka robot controller via fast Ethernet. Machine vision is accessed directly from the robot's control panel, which is a handheld programming device with a color screen. Machine vision is no longer managed separately from the robot, nor does it require its own PC monitor.Rather than using fixtures to locate the product, the vision system uses two slab-position checks on the conveyor before it picks up the slab and puts it on the pallet. The robot determines the center and the rotation of each slab so it can lift it correctly and place it accurately on the pallet prior to banding and shrink-wrapping. No fixture is required, and the slabs can be in any position on the conveyor.Cost savingsVery fast data transfer speeds helped the building supplier realize a cost benefit. As the concrete slab moves into position, the robot takes three images, compares them and calculates the mean location. Previously, a separate camera and robot were used, and the transfer of this information could have a delay of about a second while the systems analyzed the data. Now the total time to pick up and place the product is just 4.5 seconds, which is considerably faster than when the operation was performed manually."This reduces cost and simplifies operation of the robot's vision," said Ian Tatton, DVT's director of business development for northern Europe. "It also means the operator's perception is of a robot with integral vision, so there is going to be an increased expectation that the robot will perform tasks that incorporate a vision function."Besides this application, the robot can check for any corners that might be missing from the pieces it's moving. This determination can be made before placing the piece on the pallet, saving time and headaches down the line.Brett Green, general sales manager at Kuka, sees the overall benefits as cost and time savings. "Quality control improves dramatically, and there are also benefits for distribution. The manufacturer's distribution centers and retail garden centers receive uniform packaged product, and once palletized, no one sees or has to touch the product again."The heart of the vision-enabled robotic technology is its twin graphical user interface, also jointly de-veloped by Kuka and DVT. It controls how the robot's vision is managed and operated. DVT's Framework Machine Vision software remains accessible through the robot's control panel, so a trained operator can set up any number of work routines, inspections, tests, locating coordinates, scans, optical character recognition activities or other tasks, and any operator can perform them without training. As complex routines become programmed into the robot, its vision can be taught to recognize different products that trigger certain routines. In the case illustrated here, the different slabs are inspected and stacked in particular ways, just as, in another scenario, a robot might use its vision to learn paths to locate a variety of tools to perform different tasks.With cost and time savings, operational improvements and the ability to carry out new, more efficient functions, machine vision robots will become even better at adapting and responding to their environment. In the future, vision-enabled robots will continue to lend a helping hand and a watchful eye to a variety of applications.