3D vision-guided robots benefit packaging applications

The new system, founded on RTS' modular PickCell software, is said to pave the way for high-speed pick-and-place robotics for new and challenging packaging applications, for example gripping complex shapes and profiles, picking variable heights, enhanced inspection and defect rejection, measuring volume and mass, and picking products individually from a random arrangement from a variety of in-feeds.

David Bradford, Managing Director of RTS Flexible Systems, comments: "This is an exciting development for end-of-line packaging processes. Advances in technology mean we can deliver applications that have been difficult for conventional 2D technology. Solutions will often depend on individual products and processes, but are likely to offer manufacturers opportunities to significantly improve throughput and raw materials yield."

The ability of 2D vision-guided robotic automation to generate the position and orientation of products is well established, and generally works on greyscale, colour or line-scanning vision systems as a means, for example, of enabling quality inspection and defect inspection. Because the 3D system can process the x, y and z co-ordinates of the product it can also gauge the height accurately, as well as determining the depth and surface contours of the product to be processed.

RTS says this technology has enabled its engineers to demonstrate picking of products with unpredictable or variable heights - such as natural food products or baked products - to tight tolerances without inflicting damaging.

Clear advantages

Unlike 2D technology that relies on contrast, 3D vision can work from the product's profile, so it enables inspection and picking of transparent or semi-transparent products such a vials or syringes, and presents the opportunity to pick from a random feed, pile or tote bin.

The fine surface contours of a complex product shape - which can be confusing or invisible to a 2D system - have been profiled in 3D so that the robot can identify the position and orientation, so that a label can be then be applied on an indented surface area.

Quality inspection is enhanced because a 3D system can detect the depth of a surface defect. So, for example, a 3D system could be programmed to reject baked products with air pockets or sinkage outside a certain tolerance, whereas a 2D system would normally only be able to select on the position and size of a defect.

A technology development team at RTS, which has been involved in robot application advances for nearly 20 years, has collaborated with Sick UK to use 3D vision technology using a laser-based triangulation technique to capture the profile and shape of the product as it enters the robotic cell. A single camera is set at an angle on the axis of an object moving along a conveyor. This camera is integrated with a laser vertically above the object. As the object moves through the laser line, the camera builds up a series of profile "slices' to create a digitised 3D image of the surface contours.

RTS is a well-established supplier of integrated automation systems. As a bespoke supplier it can combine the best available robots and vision systems with other machinery as required, such as conveyors, flow wrappers or checkweighers. The key to integration is RTS' modular approach based on its proprietary software, which enables product data to be processed and communicated to the picking robots as well to being integrated with production line in-feeds and out-feeds to ensure line balancing and process optimisation. One key strength is in robot co-operation, wherein robots collaborate to assemble a package of products. The robots "talk to each other' and work "as a team' to produce the desired result.