Liberating robots with vision

Robots need ‘eyes’ in the form of machine vision and this has resulted in the two technologies developing symbiotically. Eyes can’t just be ‘bolted on’ to a robot. intelligence and communication need to be added to allow them to act as guides and it is these robot guidance systems that have the potential to open up countless more applications to many more new applications.

Until recently, most vision systems probably required expert support to design and install, and certainly a great deal of programming knowledge, and external computing power, to set up.  Today, however, they are becoming ‘plug and play’ – easy to install and commission.

Vision and robotics technologies have always developed symbiotically. The latest robot advances have brought lightweight and user-friendly cobots – affordable, easy-set-up articulated ‘arms’ that can be deployed and re-deployed on multiple applications without programming knowledge.

At the same time, imaging cameras have become self-contained ‘smart’ devices onto which vision functionality can be downloaded, just like a mobile phone ‘app,’ without the user having to do any development.  SICK developed its AppSpace software development platform to enable this.

SICK has been working with Universal Robots to develop an interface that makes configuring a vision task on a cobot very easy indeed. The SICK Inspector URCap software has been developed to ensure easy integration between a UR3, UR5 or UR10 robot and the SICK Inspector PIM60 2D vision sensor. It is a is simple yet powerful toolkit for creating vision-guided robot pick and place, quality inspection and measurement with minimum time and effort.

There are already numerous automated industrial applications where robots are the perfect candidates for the job of picking randomly-arranged parts or products. The need to pick up components that have been delivered to the factory in a container, bin or stillage and transfer them onto a conveyor belt for onward processing is a common task. However, until recently it would have taken a great deal of programming complexity and sophisticated, heavyweight robot hardware to achieve. 

Using the AppSpace platform,  SICK has developed both 2D and 3D vision-guided part localisation systems for smaller-scale robots and cobots. This sort of robotics is not designed to offer a high-speed substitute for manual picking but replaces a human’s repetitive and mundane task with a safer, high consistency alternative. It opens up new applications for picking specific small parts like bolts from a deep mixed parts bin and placing them on a conveyor or selecting part-completed items and placing on a press or machining centre.

The SICK PLOC2D provides a vision system for 2D localisation of parts, products or packages to be picked from a static workstation, moving belt, or feeder system. It uses a stereoscopic vision camera to enable 3D vision-guided bin picking applications of much smaller objects than was previously possible.

The PLOC2D and the PLB 520 have been developed to be directly compatible and simple to integrate with most industrial robot systems, including Universal Robots’ cobots. They can be connected directly to the robot control without programming skills or training and are ready to use almost immediately.  With installed software and an SD card, both systems have an easy-to-use interface which is compatible with webserver, Ethernet TCP/IP robot- and PLC interfaces, allowing site or remote configuration.  

The ‘democratisation’ of vision-guided automation is underway. Soon, almost no job will be too small for your robot to be given at the beginning of a shift – maybe even switching between jobs as needed all along the production, packaging and warehouse process.  With off-the-shelf hardware and ready to use ‘apps’, the versatility and flexibility promised for end users makes the future look very exciting indeed.