Pick and place activities are prevalent in most robotic applications, both in industry (machine care, assembly or palletizing / depalletizing), as well as service sectors. These autonomous procedures are, therefore, more than consolidated in specific organized contexts and with certain types of components. When components with high variability or in less situations must be handled, this is not the case. Commercial container collection solutions are not suitable for this application, as it is impossible to configure each of the hundreds of references to control.
Tekniker, a member of the Basque Research and Technology Alliance, has developed new artificial vision algorithms for industrial robotics applications that allow the recognition of grip points of unknown objects organized inside boxes as part of the European PICK-PLACE project.
Ander Ansuategi, Tekniker researcher. He commented. “The method we built, which is based on Deep Learning techniques, allows us to directly predict the points of attachment of the elements without taking into account the identification of the objects.”
Tekniker has also been in charge of the technical part of the project and has built a monitoring system for the shared workspace between technicians and robots. “The aim is to improve robotic techniques related to artificial vision, flexible handling and grip, precision and reliability of operations to overcome the problems associated with handling complicated elements in the logistics sector”, says the researcher from Tekniker.
Computer code integration
To verify the effectiveness of the artificial vision system designed by Tekniker, 2 pilot cases have been distributed in the initiative.
The software developed has been integrated into an automatic order preparation and reception system from ULMA Handling Systems, project organizer; Furthermore, the system was evaluated in the creation of spare parts orders for the dealer network of TOFAS, a car manufacturer of the FIAT group.
Project outcomes include the commercialization of sophisticated information technology through the development of automation and artificial intelligence solutions for applications such as unit order generation and machine power.
For all information, feel free to send us a chat or to call us to receive our support specialist.