angle-left When robots learn to collaborate
INNOVATION TUNGSTENO

When robots learn to collaborate

Collaboration between humans and machines will be necessary to achieve greater efficiency and precision especially in repetitive tasks. Credit: Universal Robots.

 

FRANCESCO RODELLA | Tungsteno

The barrage of robots increasingly able to perform complex tasks on their own is having a great impact on industry. Researchers and companies agree that their potential can increase exponentially if these machines are also able to create good chemistry with humans, as well as among themselves, and even more so if innovative technologies are integrate, say the experts. The effects of this interaction will be reflected in everything from factories and warehouses to space missions.

Companies from different sectors note that among the benefits of incorporating robots in assembly lines alongside humans will be greater efficiency and precision in the performance of monotonous and repetitive tasks, the reduction in the risk of work-related injuries and accidents, the optimization of waiting times and the limiting of overproduction. "A person brings skill, flexibility and the ability to solve problems, while a collaborative robot offers strength, resistance and precision in the realization of the task in question," says the Spanish Association of Robotics and Automation.

Jordi Pelegrí, in charge of Spain and Portugal for the company Universal Robots, affirms that "this cooperation can be more and more adaptive depending on the needs of the industry." In the COVAP food cooperative, for example, they needed to speed up the ham's vacuum packaging process to meet the established production standards. To achieve this, they deployed next to the cutting machine a robotic arm that is responsible for putting one after another, in the right place, the trays where the operators are going to place the cut ham. In this way, the company says, the objectives were achieved without affecting the quality of the product, food safety or the necessary environmental conditions.

Applications like this, believes Pelegrí, are increasingly common, and can go even further. Specifically, he believes that robots have the ability to learn to perform more "cognitive" repetitive tasks, for which it may be useful to apply techniques such as machine learning. "If I need to pick up objects that are similar, such as apples, and I have a pattern to recognize them, I can do that directly with a robot," he argues. By increasing the capabilities of machines, one can design models in which more than one robot interacts, for example one that carries an object and another that takes that piece and manipulates it, he adds.

The incorporation of a robotic arm together with the operators in the COVAP production chain optimizes the process, guaranteeing quality and safety. Credit: Universal Robots.

An amplified cooperation

The German multinational Mann+Hummel, producer of industrial filters, has launched some applications of collaborative robotics in its Zaragoza factory. In one of them, the cooperation is tripled: a robotic arm is responsible for moving a piece from a welding machine to a verification machine, waiting for it to finish its process, taking the piece and delivering it to an operator at the moment when it is needed. In another similar collaboration, the human worker has to execute the opposite task, that is, to load the robotic arm with a piece. The robot, in this case, is equipped with lights. When their color changes from blue to green, it means that it has been loaded correctly and the process can now be started.

As we can see, these machines need different technologies to be able to work. For example, the robotic arms used by Mann+Hummel, designed and manufactured by Universal Robots, are programmed through a tablet with installed software and can be configured to perform tasks ranging from packaging to assembly, and even incorporate a camera for performing quality control work.

Giving the robots "eyes" to orient themselves in the work environment is one of the possibilities that the sector explores when trying to increase its capabilities. Jordi Pelegrí maintains that in addition to this technology, called artificial vision, as well as machine learning, one can also apply other technologies such as the Internet of Things and Big Data. He explains that in many cases the key is to combine them. "If I collect information through artificial vision, then I have to see what to do with it and I can apply other techniques to process it."

The future involves the incorporation of collaborative robotics applications, such as this example of two robots working together to get through doors. Credit: Boston Dynamics.

Towards new horizons

Another innovation that is expected to provide more and more solutions in this regard is 5G, whose widespread implementation is expected starting from 2020-2021. This mobile network promises to guarantee more speed in the transfer of data and less delay in the response. In this way, it could allow better communication between machines and greater capacity in the remote control of autonomous robots.

Beyond the technology used in each case, according to Pelegrí it is essential that the collaboration between machines and humans be simple. In other words, robotic systems must also be easy to handle by unskilled workers. "It would not be viable to travel in a car if to drive it you needed the same concepts necessary to fly an airplane. The piloting of these machines has to be easy and manageable. That is our mission."

· — —
Tungsteno is a journalism laboratory to scan the essence of innovation. Devised by Materia Publicaciones Científicas for Sacyr’s blog.

También podría gustarte

FIND US ON FACEBOOK

INNOVATION VIDEOS

CONTACT SACYR

Paseo de la Castellana, 83-85.
28046 - Madrid (Spain)
Phone: +34 91 545 50 00

www.sacyr.com