Skip to main content Start main content

Research Laboratory for Advanced Social Robotics

 

2025 02 28 TRS Team members

Founded in 2025, the Research Laboratory for Advanced Social Robotics represents the confluence of three antecedent institutions: the Center for Advanced Media Research Amsterdam (CAMeRA, 2007-2012), Services of Electro-Mechanical Care Agencies (SELEMCA, 2011-2015), and the Social Robotics Pop-up Lab (ROBOpop Foundation, 2015-2020). Each of these entities contributed distinct strands to the laboratory’s intellectual mark-up: CAMeRA’s investigations centred upon emergent media technologies; SELEMCA pioneered the development of Caredroids, notably propelling the Alice R50 into public consciousness; and ROBOpop served as a lively conduit for scientific outreach and citizen science in social robotics.

The present laboratory, EARNEST, not only inherits these diverse legacies but also advances into the frontiers of foundational science, probing such phenomena as the interplay between affective and cognitive processes, the observer effect, and the application of quantum probability to decision making. Our methodological commitment is to instantiate theoretical constructs within robotic platforms, operating under the principle that computational realisability is a necessary – though not sufficient – touchstone for theoretical coherence and logical verification. Thereafter, these theories embodied in a robot are subjected to rigorous empirical tests, both in controlled laboratory settings and in situ field deployments, where the inevitable collision with practical realities serves as the ultimate arbiter of theoretical validity.

On the applied front, our research is conducted in close collaboration with end-users and other stakeholders (e.g., professional caretakers, industrial partners), exploring the dynamics of human-robot interaction across domains such as healthcare, education (particularly in tutoring contexts), and daily self-management tasks. Our EARNEST design aspires to deliver interactions that are not only personalised and culturally attuned but also safeguarded by robust data security (e.g., blockchain), thereby fostering the emergence of robots as universal interfaces in the IoT capable of serving as trustworthy confidants and advisors – free from the distortions of generative hallucinations.
 

Vision:

Envision a society wherein each individual possesses a profound comprehension of the capabilities and constraints inherent in robotics and artificial intelligence, particularly within the social sphere. Such understanding should serve to enhance the benevolent facets of these technologies while mitigating their potential misuse, thereby ensuring they contribute to a life of significance for all, rather than becoming instruments of exploitation or peril. Through research and design, we cultivate intellectual leadership that informs both present and prospective endeavours concerning robotics and AI. This guidance aims to inspire and galvanise individuals towards the development and implementation of social robots, which prove beneficial in areas where human abilities may be insufficient – be that physically or intellectually.

Use case scenario Alice in the IoT

 

 

Mission:

The principal aim of our endeavours at the Research Laboratory for Advanced Social Robotics is to cultivate independent thought and to demystify the nature of robots and artificial intelligence as technologies comprehensible from foundational principles. We seek to elucidate how human perceptions, those of engineers and scientists included, often ascribe attributes to these tools (i.e., consciousness, intelligence, sentience, understanding) that they do not intrinsically possess.

Our laboratory is devoted to fundamental research and theory development in areas such as probability theory, ‘true’ randomness, information entropy, and epistemics. This research underpins the design of robotic and AI systems capable of navigating complex scenarios and making informed decisions that surpass human capabilities, all while ensuring safety and security, with humans retaining ultimate decision-making authority. To this end, we develop sophisticated algorithms (e.g., quantum circuits), invent new statistics (e.g., for the study of outliers), and conduct empirical investigations into diverse topics, including social robots for problem solving and decision making, robots that support mental and physical health, those that serve educational purposes, and those that provide assistance within the home.

Our foremost objective is to research and design systems that simulate human-like (though not human-identical) social behaviours, thereby supporting facets of life where individuals face challenges and where human assistance is either unavailable or insufficient to meet the needs at hand.

 

Location:V611, 6/F, JCIT

 

 

 

Hosted by Research Laboratory for Advanced Social Robotics, the Research Grants Council funded project in the Theme-based Research Scheme (TRS) 2024/25 (Fourteenth Round) (T43-518/24-N) plays a prominent role (https://www.polyu.edu.hk/sd/news-and-events/news/2024/20240718-prof-johan-f-hoorn-awarded-funding-from-rgcs-theme-based-research-scheme/). The project Social Robots with Embedded Large Language Models Releasing Stress among the Hong Kong Population is set to build a Smart Inclusive Social Robot for a Dynamic Urban Environment, or sustAIner, for short.

Robotics-sustner

“Family wellbeing is the foundation of a harmonious society that promotes psychological health and individual flourishing” (Zeng et al., 2023). However, Hong Kong made it to The Lancet as an exemplary case of depression following broad social unrest. Around 12.8% of the population suffer from Post-Traumatic Stress Syndrome. COVID-19 stressors increased levels of anxiety and depression during the pandemic. Hong Kong is an overworked city with 61% of its population being stressed out, anxious, depressed, and in a bad mood. Occupational stress adds an annual economic cost to the city of 4.81-7.09 billion HK$. “Positive emotions buffer the negative impact of job stressors on absenteeism” (Siu et al., 2020) but the mental-care system is overburdened. We should “encourage and promote early health seeking treatment…” (Hung et al., 2022) but care avoidance is common. “Psychological support, such as brief, home-based psychological interventions, should be provided to citizens…” (Choi et al., 2020) but those are absent in HK.
We propose training Social Robots with Embedded Large Language Models (LLMs) on localised, cultural, and personal data to bring customised mental care to those who remain undetected by the official medical care system. Human-like social robots have shown to be natural interaction partners, assisting information search, improving health and mental wellbeing, and supporting educational tasks. We will connect robots and avatars to the 2021 TRS-awarded Hong Kong AI-hub, while providing a new software architecture for distributed computing, scalability, and privacy protection. We will develop training protocols, logic-symbolic AI, and design guidelines for novel methods and functionality, tested in-situ by local communities.
Tasks to be performed are 1) Building an architecture for dynamic access to the HK AI-hub with large-scale deployment of health-related information while handling Cloud delays, scalability issues, mobile distributed IoT, and data privacy; 2) Develop AI to continuously monitor topics citizens worry about and train Large Language Models (LLMs) on the most relevant topics, using HK AI-hub information; 3) Refine LLMs with logics, norms, and values (medical, cultural) to prevent improper conduct and hallucinating ‘facts’; 4) Develop localised and personalised conversation formats for topics citizens worry about; 5) Evaluate performance and efficacy through field studies into local communities.
Uniqueness and novelty lie in cutting-edge technology guided by design studies into urban dynamics. We target information relevant to citizens and take care that AI-enhanced information provision prioritises the role of emotions in technology use, avoiding market failure of apps and large-scale ICT infrastructure.

PC:
Johan F. Hoorn (D. Litt., D. Sc.), SD & COMP, PolyU

Co-PIs:
Jiannong Cao, COMP, PolyU
Kin Wai Michael Siu, SD, PolyU
Maggie Li, COMP, PolyU
Sek Ying Chair, Nethersole School of Nursing, CUHK
Xiaojuan Ma, CSE, HKUST

Co-I:
Chris Webster, Urban Planning and Design, HKU

Collaborators:
Ivy Yan Zhao, School of Nursing, PolyU
Bin Xiao, COMP, PolyU
Jiayan Pan, Social Work
Kee Lee Chou, Social Sciences and Policy Studies, EdUHK
Humayun Rashid, Xavor Robotics, CN

 

 

Director:
Johan HOORN (SD & COMP)

Deputy Directors:
LUXIMON Yan Tina (SD)
Fiona LIU (COMP)

Lab Manager:
Tesfa MERHATSIDK

 

 

Your browser is not the latest version. If you continue to browse our website, Some pages may not function properly.

You are recommended to upgrade to a newer version or switch to a different browser. A list of the web browsers that we support can be found here