PAL Robotics’ Highlights from HRI 2023
Our team had a fantastic time at the 18th ACM/IEEE International Conference on Human-Robot Interaction (HRI) held in Stockholm from March 13 to 16 2023. HRI is an annual conference for basic and applied HRI research where researchers from across the world present their best work to HRI to exchange ideas about the theory, technology, data, and science furthering the state-of-the-art in the field. PAL Robotics was a sponsor of the conference, and in addition, our HRI research team had the opportunity to take part in four workshops and two poster sessions. Moreover, our humanoid social robot ARI showcased speech and story-telling capabilities in a collaborative story-telling adventure set in an imaginary hospital. Read on for our highlights of the event.
The conference theme this year was “HRI for all” and aimed to promote diversity and inclusion in HRI through key theories, methods, designs, studies, and technical advances. The community was asked to consider ways to both make the field a more inclusive place as well as to encourage inclusion within research methods and practices.
In workshops – PAL Robotics’ Dr. Séverin Lemaignan joined a debate on life-long learning methods for HRI alongside Yale’s professor Brian Scassellati and Google’s researcher Karol Hausman; and as a presenter in a workshop on how fostering creativity and collaboration can help solve challenges in HRI, and Dr. Raquel Ros presented our recent work on the new PAL Robotics-led ROS4HRI standard. In addition, PhD candidate Lorenzo Ferrini presented his research on semantic gaze control, and Dr. Antonio Andriella participated in a session to discuss his papers on robot behaviour, and replicability of human studies in HRI.
PAL Robotics’ engagement with the HRI academic community also includes Dr. Séverin Lemaignan role as member of the HRI Steering Committee (that oversees many of the community’s academic activities) and as Publication Chair in charge of coordinating the publication of papers presented at this year’s HRI 2023 conference.
Workshops on topics including standardising Human-Robot Interaction and semantic scene understanding
“Lifelong Learning and Personalization in Long-Term Human-Robot Interaction” and “Pioneers”- Dr. Séverin Lemaignan
Dr. Séverin Lemaignan, Senior Scientist in Social Robotics at PAL Robotics participated in the “Lifelong Learning and Personalization in Long-Term Human-Robot Interaction” (LEAP-HRI) workshop, which aimed to promote interdisciplinary insights for lifelong robot learning and adaptation to users, context, environment, and activities in long-term interactions. He also participated in the “HRI Pioneers” workshop (the premiere forum for graduate students in HRI) as an invited speaker for the workshop’s Industrial panel. He discussed his experience as a research scientist in industry and the various pathways to industry for researchers.
“Advancing HRI Research and Benchmarking Through Open-Source Ecosystems” – Dr. Raquel Ros
Dr. Raquel Ros, Senior Scientist in Social Robotics at PAL Robotics presented her talk “ROS4HRI: Standardising an Interface for Human-Robot Interaction” in the “Advancing HRI Research and Benchmarking Through Open-Source Ecosystems” workshop. This workshop aimed to identify the preconditions and requirements to develop an open-source ecosystem that provides open-source assets for HRI benchmarking and comparison. Dr. Raquel Ros discussed how to standardise interfaces in HRI with the novel ROS4HRI standard, initially designed by PAL Robotics, and illustrated the work on the ARI robot within the project SHAPES to bring greater efficiency in healthcare for people in need.
“Semantic Scene Understanding for HRI” – Lorenzo Ferrini
Lorenzo Ferrini, PhD Student at PAL Robotics presented his paper “Unifying Bottom-Up and Top-Down Attention Models for Social Robots” focussed on semantic gaze control, in the “Semantic Scene Understanding for HRI” workshop. This workshop aimed to explore ways to learn and ground abstract semantics of the physical world towards assistive autonomy in areas such as unobtrusive learning from observations, preference learning from contextual observations, predicting and reasoning over physical effects of human actions, among others.
Poster sessions – Dr. Antonio Andriella
Dr. Antonio Andriella, Research Scientist at PAL Robotics participated in a poster session where he discussed his two papers on a computational approach for proactive robot behaviour and on improved replicability of human studies in HRI. Find out more here:
ARI demonstration of story-telling capabilities in a hospital scenario
ARI was joined by visitors for the creation of a collaborative story-telling adventure in an imaginary hospital showcasing speech and cognitive modelling abilities.
During the demo, visitors who came to our booth were presented with a map of an imaginary hospital with various tokens representing ARI, people, and objects.
ARI demonstrated story-telling capabilities by narrating the events unfolding and having ARI and people interact with each other. Visitors moved the tokens according to the story, and ARI was able to recognize the actions, such as when someone purposefully misplaced an item and reacted accordingly. Read the publication.
Our recent research published in HRI and social robotics
Here are some of the papers we have had published recently in HRI and social robotics:
Séverin Lemaignan, Sara Cooper, Raquel Ros, Raquel Ros, Lorenzo Ferrini, Antonio Andriella (2023): Open-source Natural Language Processing on the PAL Robotics ARI Social Robot
Ilenia Cucciniello, Antonio Andriella, Silvia Rossi (2023): “Towards a computational approach for proactive robot behaviour in assistive tasks”
Shelly Bagchi, Patrick Holthaus, Gloria Beraldo, Emmanuel Senft, Daniel Hernandez Garcia, Zhao Han, Suresh Kumaar Jayaraman, Alessandra Rossi, Connor Esterwood, Antonio Andriella, Paul Pridham (2023): “Towards Improved Replicability of Human Studies in Human-Robot Interaction: Recommendations for Formalised Reporting”
Lorenzo Ferrini, Séverin Lemaignan (2022): Kinematically-consistent Real-time 3D Human Body Estimation for Physical and Social HRI
Sara Cooper, Séverin Lemaignan (2022): Towards using Behaviour Trees for Long-term Social Robot Behaviour
Sara Cooper, Francesco Ferro (2022): Lessons Learnt from Deploying ARI in Residential Care
Sara Cooper, Óscar Villacañas, Luca Marchionni, Francesco Ferro (2021): Robot to support older people to live independently
Ioanna Dratsiou, Oscar Villacañas, Sara Cooper, Pavlos Isaris, Manex Serras, Luis Unzueta, Victor Fernandez-Carbajales (2021): Assistive Technologies for Supporting Wellbeing of Older Adults
ARI and TIAGo: social robots by PAL Robotics
ARI is a humanoid social robot developed by PAL Robotics in 2019 for HRI. The robot is a research platform in a number of EU-funded research projects including SHAPES, SPRING, and Pro-cared. The robot is easily programmed through the user interface and is equipped with voice and facial recognition systems.
TIAGo is a mobile manipulator as well as a humanoid social robot, that combines perception, navigation, manipulation, and Human-Robot Interaction skills. TIAGo is used by multiple universities and research institutions in daily research and has been part of EU research projects that PAL Robotics partners in, including SANDRo2, CANOPIES, and EVOLVED-5G.
We would like to thank everyone who participated as well as all event organisers and sponsors. We’re looking forward to applying some of the learnings to our ongoing research, and we can’t wait to see what the future of HRI holds. We look forward to opportunities to meet again in the future!
To find out more about PAL Robotics and our work, visit our blog and to ask any questions don’t hesitate to contact us. If you’re searching for project or event partners don’t hesitate to get in touch – we are always looking for new opportunities to collaborate.