The theory and practice of Human-Robot Interaction: reflections from a workshop on Collaborative Robots for Agile Production

On November 21st, 2023, NTNU co-organized with SINTEF the last of a series of workshops on robotics for agile production physically in Trondheim, Norway. The workshop’s title was “Cobots in Agile Production: Redefining the Future of Human-Robot Interaction”

On a cold but sunny Norwegian day, we invited people from the scientific community, industry and civil society to gather together in Trondheim and go on a journey that started with the psychology of humans interacting with robots and ended with a discussion on actionable points that the robotics community could take to improve the human experience of using robots. Human-robot interaction is charged with fears of job loss and security concerns but also excitement for innovation and welcoming solutions that can make people’s jobs easier. However, how can we ensure a balance between these two scenarios? How can we make human-centric robots for agile production?  

During the workshop, we explored these questions first with two keynote speakers: Federico Manzi, Researcher in Developmental and Educational Psychology at the Catholic University of Sacred Heart of Milan; and Fabio Sgarbossa, Professor of Industrial Logistics at NTNU.  

Dr. Manzi, an expert in Developmental Psychorobotics, presented his work on the Theory of Mind and human-robot interaction. Theory of Mind allows us to attribute mental states, that is emotions, desires, intentions, or thoughts, to oneself and others and to predict one’s and others’ behaviours. This concept, at the basis of developmental psychology, can help us better understand how humans perceive robots and reflect on robots’ mental states. The physical features of robots and their behaviours have a huge impact on how people interact with them. Humans tend to anthropomorphize robots and attribute to them mental states, this happens to a greater extent when robots look like humans. 

The application of the Theory of Mind to the area of collaborative robots is highly valuable for exploring trust and moral context. In Dr. Manzi’s research on trust, children have shown similar behaviours towards robots and humans but slightly higher trust towards humans. Regarding the study of moral transgression, Dr. Manzi presented a very interesting example based on the trolley dilemma which reflects upon robots’ mental states. In this case, robots were judged much harder for their moral decisions than humans. All of these findings provide useful insights for improving human-robot interactions in the agile production sector, where predicting robots’ intentionality and judging their choices are everyday activities for some workers. 

Zooming in on the specific case of agile production, Professor Sgarbossa presented his work on mobile robotics. Agile production is not only about smart systems but also about smart logistics and, while we are still far from full robotization, there are already a lot of robots in warehouses now. For example, Prof. Sgarbossa presented his work with mobile robots for warehouses moving boxes and materials around as well as the example of a project for civil society helping librarians move books between shelves. While robotics are useful and can make work easier, Prof. Sgarbossa concluded that humans are still the key players and decisions to automate should be the consequence of a thorough analysis of return on investment and increased efficiency, flexibility and reactivity. 

Following the two keynote speeches, the Responsible Robotics Compass was presented, setting the stage for concrete discussions on how these fascinating theoretical insights can be turned into recommendations for the robotics community. With a focus on low-code or no-code solutions, which are a new trend in the production of collaborative robots, two presentations preceded the group discussion. 

First, we heard from Sigmun Hennum Høeg (PhD Candidate in Mechanical and Industrial Engineering, NTNU) on the use of Large Language Models for robot manipulation. Through the use of large language models, we can program robots to decide which skill to use based on a selection of the most relevant ones connected to the prompted words. Language-based programming could make interaction with robots much easier.

Then, Andrej Cibicik (Research Scientist at SINTEF Manufacturing) presented the example of the Humantech project developing robots for the construction industry. SINTEF Manufacturing is doing a lot of work on collaborative robots, including with sensor technology and wearable technology, like exoskeletons. Andrej brought the example of his work on gesture-based human-robot communication for a robot handing bricks to construction workers. This robot would collaborate with the human worker by taking on what is a low-skilled and repetitive task, thus, helping the workers. 

Based on the keynote speeches and the examples presented, the participants discussed in small groups about human-robot interaction and actionable points to propose to the robotics community. Many groups had reflections on the role of fear of job loss in human-robot interaction. Robots presented as substituting humans do not follow a human-centric approach, according to the participants, and policy-makers should help workers widen the scope of their jobs to allow for reskilling and upskilling. Robot developers and employers are also responsible for ensuring workers’ physical and mental safety when interacting with robots and providing education on how the robots work. Overall, the participants saw collaborative robots as the future of agile production but in a harmonious way that can take into account diversity and inclusion and always have humans at the centre. 

The participants were very interested and engaged in the presentations and discussions and appreciated the interdisciplinary nature of the workshop which allowed them to reflect on news topics. The goal of the Robotics4EU project is to promote the more widespread adoption of AI-based robots in Europe. In this workshop, we collaborated with innovators, researchers, citizens, and decision-makers working in agile production, which helped raise awareness of non-technological aspects of robotics among stakeholders. This workshop has been exciting for Robotics4EU to gather feedback on the RoboCompass and understand sector-specific challenges.

Author: Silvia Ecclesia

cookies definitions

Robotics4EU Project may use cookies to memorise the data you use when logging to Robotics4EU website, gather statistics to optimise the functionality of the website and to carry out marketing campaigns based on your interests.

They allow you to browse the website and use its applications as well as to access secure areas of the website. Without these cookies, the services you have requested cannot be provided.
These cookies are necessary to allow the main functionality of the website and they are activated automatically when you enter this website. They store user preferences for site usage so that you do not need to reconfigure the site each time you visit it.
These cookies direct advertising according to the interests of each user so as to direct advertising campaigns, taking into account the tastes of users, and they also limit the number of times you see the ad, helping to measure the effectiveness of advertising and the success of the website organisation.