New advances in wearable and ubiquitous computing presented at UbiComp 2025
Author ADM+S Centre
Date 3 December2025
Researchers Kaixin Ji and Hiruni Kegalle from the ARC Centre of Excellence of Automated Decision-Making and Society at RMIT University recently presented their research at UbiComp 2025, hosted in Espoo, Finland.
The annual conference brings together leading international researchers, designers, and practitioners working across ubiquitous, pervasive and wearable computing.
Hiruni presented “Watch Out! E-scooter Coming Through!: Multimodal Sensing of Mixed Traffic Use and Conflicts Through Riders’ Ego-centric Views” examining rider behaviour across three types of transport infrastructure using eye-tracking, cameras and speed data collected in real-world environments.
Hiruni connected with members of the UbiComp and MobileHCI communities and explored ideas for future collaboration including potential joint studies on cross-cultural perceptions of micro-mobility safety, as well as opportunities to co-author comparative research and explore data-driven design frameworks using shared sensing datasets.
Hiruni said the conference reinforced the growing importance of considering design, development, and deployment aspects of ubiquitous and pervasive computing technologies, as well as understanding the human experiences and social impacts these technologies facilitate.
“ I was particularly inspired by the strong focus on translating research into real-world applications, especially those integrating large language models (LLMs) with sensor data to enable more context-aware and adaptive systems,” said Hiruni.
Kiaxin presented “SenseSeek Dataset: Multimodal Sensing to Study Information Seeking Behaviors,” which provides resources investigating complex cognitive mechanisms as people interact with information, using consumer-grade passive sensors to record their physiological and behavioral responses.
Kaixin also presented “GLOSS: Group of LLMs for Open-ended Sensemaking of Passive Sensing Data for Health and Wellbeing” an outcome of a collaboration with the UbiWell group at Northeastern University.
This research uses an agentic LLM network for open-ended sense-making on passive sensing data.
During the conference, Kaixin reconnected with collaborators from Northeastern University and Cornell Tech, strengthening ties between her work in wearable sensing and their research on mental wellbeing.
She also met researchers from Tsinghua University and the Hong Kong University of Science and Technology to discuss applications of the SenseSeek dataset, and spoke with Professor Michael Beigl from the Karlsruhe Institute of Technology about emerging ear-worn devices for monitoring neurological conditions.
“I was deeply moved by the enthusiasm and supportive spirit of the Ubicomp community,” said Kaixin.
“The researchers are genuinely committed to pursuing work that can make a real impact for users, and they maintain strong connections with industry. Some have even launched startups based on their research outcomes and received venture funding.”
Both Hiruni and Kaixin served as student volunteers throughout the conference. Hiruni co-organised the UbiComp4VRU workshop with colleagues from the University of Kassel and UNSW and Kaixin served on the program committee for the UbiSense workshop.
This research visit was funded by the ARC Centre of Excellence for Automated Decision Making and Society Research Training Program and the ADM+S RMIT node.
