Daily Archives

02/04/2020

SADAKO visits INDUMETAL’s facilities

By | Blog | No Comments

SADAKO visits INDUMETAL’s facilities for a second round of recordings.

Last year, on July and September, a first round of recording sessions was carried out at Bianatt’s facilities and conducted by CERTH.

Indumetal was selected for a second round and, this time, Sadako was the partner in charge of the coordination and the recordings. On 17th and 18th December, Sadako arrived at Indumetal’s facilities, loaded with all their cameras equipment and ready for two hard working days. On the other hand, Indumetal arranged a specific area for the classification and disassembly capturing and, following Bianatt’s experience and indications, a metal structure for camera installation and a correct illumination were provided as well. The use cases selected for this round were microwaves and PC towers.

Additionally, human motion analysis and prediction were recorded in order to detect human actions at cell-level and hand gestures for human robot interaction and to identify task on the human worker. For that, different communication gestures and some of the described actions on D3.1 were performed by Indumetal’s workers.

Do we recycle together? Human-robot collaboration for recycling

By | Blog | No Comments

Do we recycle together? Human-robot collaboration for recycling

With the Information Society and the steadily growing consumption of electrical and electronic equipment (EEE) worldwide, society is facing the challenge of dealing with increasing amounts of waste electrical and electronic equipment (WEEE) in a sustainable and responsible way.

Nowadays manufacturing companies are going through an increasing public and government pressure to reduce the environmental impact of their operations. But when dealing with WEEE, some difficulties arise in classifying and dismantling electronic devices.

How to improve the current process that is mainly manual and time consuming? Can technology help in the improvement of collaboration between humans and robots? A novel solution is to promote cooperation between an operator and robots in this type of processes.

More specifically, the overall goal of HR-Recycler project is to create a hybrid collaborative environment, where humans and robots will harmoniously share and undertake, at the same time, different processing and manipulation tasks, targeting the industrial application case of WEEE recycling.

This goal will be achieved through Human Robot Collaboration (HRC) systems. In the words of Sara Sillaurren, TECNALIA´s project coordinator: “In this new environment in which human and robot have to work collaboratively, it is essential that process design take into account the state of the human for a better understanding between the two of them.”

Within the scope of this project, TECNALIA will bring Human Factors and User experience Lab, which combines physiological sensors and technologies for monitoring human behavior.

Through the usage of wearable devices and platform in this laboratory, and with the collection of psico-physiological data from users, human behavior can be better understood when collaborating with robots and thus the system can be adapted so that processes are more fluid and reliable.

According to this, all types of signals captured give us very interesting information about the user and referred to the activity they are performing. For example, they can give us cognitive information, that is, if the user is paying attention and if the user is able to retain the information we are showing. Another type of information that can be measured is the emotional one. For example, the affective valence or degree of attraction towards the activity they are completing, or the activation and emotional impact, that measure the level of calm or excitement of the user. Finally, there are other types of metrics such as visual attention (that is, what first calls the attention of the user) or the implicit association (between a concept and an attribute).

Furthermore, considering the multiple factors that affect Human Robot Collaboration, TECNALIA is going to define the experimental studies targeting end- user with a purpose of understanding trust in collaborative disassembly and the validation of trust factor model to improve it.

For doing so, a variation of the inspection game has been designed in order to expose participants to different trust stimuli towards machines. Inspection game is a mathematical model of a non-cooperative situation where an inspector verifies that another party, adheres to legal rules instead of shrinking work duties. This experiment will allow us to detect how (un) predictability and prior experiences affect trust.