Projects

Our Research Domains

Domain 1: Human-Building Interaction

Domain 2: Information visualization

Domain 3: Augmented and Multimodal Interaction

Domain 4: Behavior change

Domain 5: Mediated and collaborative technology

Domain 6: Human-centered Data Privacy

Domain 7: Cognitive computing

 

Domain 1: Human-Building Interaction

smartlivinglab.ch

2015-now

Human-IST has recently developed a new research field: Human-Building Interaction (HBI) with the ambition to apply HCI knowledge to the habitat and have a real impact on the way people live and work in buildings. Human-IST research works on HBI are developed within the Smart Living Lab project, co-funded by the Canton of Fribourg and EPFL in the Bluefactory Innovation Quarter. The goal is to develop innovative interactive technologies, for instance to understand human interactions in the habitats, to better design buildings, to favour behaviour change, to increase inhabitants’ comfort and reduce energy consumption.

Human-IST collaborators: Denis Lalanne, Julien Nembrini, Hamed Alavi, Himanshu Verma, Raphaël Tuor, Michael Papinutto

Partners: UniFr, EPFL, HEFR

 

Comfort: a coordinate of user experience in built environments

We study building occupants comfort across its four dimensions (thermal, visual, acoustic, air quality), with a novel perspective that highlights the importance of human’s perception and behavior. We propose a reformulation of the question of comfort that extends its scope from a “quality of built environment” to an “interactive experience of the human occupants”. With this perspective we answer questions that embody the complexity of human experience within the built environment:  what are the elements that influence our perception of comfort? Can awareness have an impact on how we pursue comfort in interaction with built environments? Can the interaction style with the building elements have impact on our comfort? How would artificial intelligence in buildings (building automation systems) change our expectations from and perception of built environments?

Human-IST Collaborators: Hamed Alavi, Michael Papinutto, Denis Lalanne

Start date: January 2016

 

Spatiality and User Experience of Hot-desking Offices

Through a longitudinal study in smart living lab, we have studied the space-use behavior of office workers in two refurbished offices in hot-desking situation. The 3D modeling and evaluation of space design using isovist and space syntax methods, we found correlations between the visual attributes of a workplace and its occupancy rate.

Human-IST Collaborators: Hamed Alavi, Himanshu Verma, Denis Lalanne

External Collaborators: EPFL, Atelier OI

Start date: January  2016

 

ComfortBox: Learning personal comfort in interaction with users

Human-IST has developed an interactive device that measures the environmental conditions, communicates the objective comfort measurements  with the users, receives feedback from them, and learning from the user's feedback tunes its prospective behavior. Comfortbox is equipped with ten indoor environmental sensors that together can create an insight into the indoor thermal, visual, acoustic comfort, and evaluate the quality of air. ComfortBox is intended to make a channel of dialogue between human and building, but also can evaluate indoor condition in relation to the productivity and health of the occupants especially in a long-term.

Human-IST Collaborators: Hamed Alavi, Denis Lalanne

External Collaborators: Logitech

Start date: January 2016

 

Smart Living Design Studio

In collaboration with the architects educators in the school of engineering and architecture in Fribourg (HEIA-FR), we plan to study the learning-by-doing practices in school design studios. Using such context where working, learning, living, resting, socializing often happen at the same place, our attempt is to study a possible future where in general the borders between different activities will be fading. We will first make a thorough understanding of  how different activities are carried out in relation to spatial configurations of the studio. Then, we empower the studio users to analyze their own space-use behavior, find the inefficiencies, and redesign the reevaluate their studio with a rapid pace.

Human-IST Collaborators: Hamed Alavi, Himanshu Verma, Denis Lalanne

External Collaborators: School of Engineering and Architecture, Fribourg

Start date: January 2018

 

Visual analytics of building data

Building Management Systems (BMS) are more and more standard building equipment. These senso-motoric systems generate great amounts of data. The aim of this project is to enhance traditional data presentation format such as graphs and tables to support expert users in early open-ended exploration of building data. Visualization prototypes involving established machine learning and interactive data visualization techniques allow expert users  to derive meaningful conclusions in limited time.

Human-IST Collaborators: Julien Nembrini, Florian Evéquoz, Denis Lalanne

External Collaborators:

Start date: April 2015

 

Insolight

The goal of the research is to establish a methodology and to develop a test and demonstration platform to measure and assess the benefits of using innovative luminaires from both technical and users’ perspective. The startup Insolight has developed a smart LED panel with the ability to dynamically change its light beam output (direction and intensity). The panel takes the form of a smart luminaire for multi-functional spaces adapting the light beam with the users’ task. The aim is to install, benchmark and test the 4th-generation prototype with real users of the “Halle Bleue”. Ultimately, this project will demonstrate and possibly validate the interest of such a luminaire in an office environment.

Human-IST Collaborators: Julien Nembrini, Denis Lalanne

External Collaborators: Prof. Jérôme Kaempf, School of Engineering and Architecture, Fribourg, Insolight

Start date: September 2017

 

MUBI Mobile user-building interface

The goal of the project is to propose a mobile interface to interact with a building database, allowing to browse in real-time environmental and energy data gathered by the building sensors, relative to the space occupied by the user. Taking advantage of the mobile platform, it will also augment sensor data by gathering user comfort perception data, in order to analyze correlations between both data streams. In this project, focus is put on the development of a mobile tool with large diffusion potential, in order to allow for its adaptation to different building contexts.

Human-IST Collaborators: Julien Nembrini, Himanshu Verma, Raphaël Tuor, Denis Lalanne

External Collaborators: School of Engineering and Architecture, Fribourg

Start date: June 2017

 

ELSA Exploration Tool for Sustainable Architecture

Website: http://elsa.epfl.ch

The goal of this project is to build a decision-making tool for architects and engineers. Its main objective is to help and guide the user at the early design stages of the building. This assistance is provided by offering different design alternatives and their relative environmental impact. These impacts are then compared to the 2000-Watt society target values in order to quantify the performance of each design alternative.

Human-IST Collaborators: Denis Lalanne, Raphaël Tuor

External Collaborators: EPFL, EPFL+ECAL Lab

Start date: January 2016

Back to top

 

Domain 2: Information visualization

Visualizing the influence of language in 15 years of political elections in a bilingual canton

This interdisciplinary project held in cooperation with HES-SO Valais Wallis and the Institute of European Studies of the University of Fribourg aims at understanding the influence of language in political elections in the bilingual canton of Valais. Based upon the data from the cantonal government elections from 2001 to 2017 in Valais, the researchers combine data visualization and machine-learning approaches with contemporary history methods to interpret the role of language in the choice of political representatives.

Human-IST Collaborators: Florian Evéquoz, Denis Lalanne

External Collaborators: Bernhard Altermatt (Unifr), HES-SO Valais

Start date: May 2017

 

CLISEL project

Website: http://clisel.eu

CLISEL is a “Coordination and support action” led by the European Union to study and gather data about law instruments, migration and climate-security with a focus on Sardinia. Human-IST is working with the World Trade Institute (WTI) to build various OD-Data (Origin-Destination) visualizations. These will allow researchers not only to communicate about their findings, but also to explore datasets that are not explicitly related together. We expect these visualizations to shape a better understanding of the relation between climate change and migration, as well as the impact of policies in response to climatic disasters on migration.  

Human-IST Collaborators: Pierre Vanhulst, Florian Evéquoz, Denis Lalanne

External Collaborators: Elisa Fornale (WTI)

Start date: August 2017

 

Visual analytics of building data

Building Management Systems (BMS) are more and more standard building equipment. These senso-motoric systems generate great amounts of data. The aim of this project is to enhance traditional data presentation format such as graphs and tables to support expert users in early open-ended exploration of building data. Visualization prototypes involving established machine learning and interactive data visualization techniques allow expert users  to derive meaningful conclusions in limited time.

Human-IST Collaborators: Julien Nembrini, Florian Evéquoz, Denis Lalanne

External Collaborators:

Start date: April 2015

Back to top

 

Domain 3: Augmented and Multimodal Interaction

iKnowU – Exploring the Potential of Multimodal AR Smart Glasses for the Decoding and Rehabilitation of Face Processing in Clinical Populations

Our ability to identify faces and decode facial expressions of emotion is essential for everyday life social interactions. Very recent advances in wearable and augmented reality technologies provide a unique opportunity to create perceptual and cognitive prostheses. This innovative project aims at developing smart glasses to help visually or cognitively impaired individuals with face and emotion recognition. Using collected information about the user’s relatives, friends and colleagues, the system will automatically recognize their presence in the visual field of the camera, their identity and emotion. This information will then be reported to the user via audio, tactile and/or visual feedback.

Human-IST Collaborators: Denis Lalanne

External Collaborators: Roberto Caldara, Simon Ruffieux, Nicolas Ruffieux

Start date: July 2016

 

Autonomous Vehicles and Us : Designing Pedestrian-AV interaction

Autonomous vehicles currently lack dedicated interfaces that allows them to communicate their intentions clearly to other road users. In this project that was started in collaboration with CarPostal, we aim to develop novel interfaces to mediate the interaction between autonomous vehicles and pedestrians. The design of the tool is realized using a participatory approach with test users. CarPostal also provides an access to their “SmartShuttles” to help us understant the various interactions that take place in real settings, and users acceptance of new interfaces.

Human-IST Collaborators: Florian Evéquoz, Himanshu Verma, Denis Lalanne

External Collaborators:  Grace Eden (HES-SO), CarPostal

Start date: September 2017

Back to top

 

Domain 4: Behavior change

CO2 Correlation

Shifting peak electricity load towards times of lesser demand has been the subject of many research efforts. The aim of the project is here to compute the CO2 content of the electricity supply to shift the load towards times of more climate-friendly production, e.g. when solar panels are producing or when hydroelectric production is at its maximum. Through the analysis of plug load smart metering data and dedicated user experiments, the potential for CO2-based load shifting in office environment is evaluated.

Human-IST Collaborators: Julien Nembrini, Agnes Lisowska, Denis Lalanne

External Collaborators: Didier Vuarnot, Building 2050 group EPFL, SLL

Start date: February 2016

 

BES-Online

Website: http://bes-online.unifr.ch

BES-Online is an therapy created by our partners from the “Klinische Psychologie und Psychotherapie” unit. The aim is to create an online version of a treatment they designed for Binge-eating disorders, which they validated in real-life context. Human-IST took charge of the technical development - web platform, database, data visualization. On the top of this, our team seeks to dynamically adapt the interface, in ordrer to minimize the number of patients that abandon the therapy.

  • Human-IST Collaborators: Marin Tomic, Fabio Sodani, Pierre Vanhulst, Denis Lalanne
  • External Collaborators: Simone Munsch, Andrea Wyssen
  • Start date: January 2017  

Back to top

 

Domain 5: Mediated and collaborative technology

Situated organization of video-mediated interaction

Video-based communication has become a common way of interacting with remote interlocutors, whether through complex videoconferencing systems or webcams integrated into consumer technologies. Ethnomethodology and conversation analysis (EM/CA) are sociological approaches that have been influential in Human-Computer Interaction (HCI) for nearly three decades due to their focus on the situated organization of practical activities. This three-month project resulted in an article presenting a state-of-the-art literature review of empirical research on video-mediated social interaction studied from the perspective of EM/CA. It puts forward an original organization of the findings on the interplay of talk, bodily behavior and spatial and material resources, and underscores the ways in which technology enables and constrains interaction, shaping familiar and novel social activities.

Human-IST Collaborators: Denis Lalanne, Jakub Mlynář

External Collaborators: Esther González-Martínez (UNIFR - Department of Social Sciences)

Start/end dates: January 2017 - March 2017

 

Luigi

Improving the overall experience of people who participate in meetings from a remote location has been a focus of research for some time now. The technological solutions proposed range from variants of Skype-type systems, to creation of specialized teleconference rooms, to the use of robots as embodiments of remote participants. However, no solution has thus far been proposed that successfully addresses more than a couple of the issues involved. We would like to explore whether the use of tangible and embodied interaction would be an effective medium for improving the user experience of participants in hybrid meetings (meetings where one or more participants are remote, which other are grouped together in a single room).

Human-IST Collaborators: Agnes Lisowska Masson, Himanshu Verma, Denis Lalanne

External Collaborators: Jakub Mlynář (UNIFR - Department of Social Sciences)

Start date: March 2017 -

 

cBoard

To improve collaborative work on large interactive display, users have to be provided with the possibility of creating personalized input. So we built a virtual whiteboard able to track users and assign each interaction with it to the correct person. The data collected to create this association can be used to further improve user experience. We explore several ways to use this information to create a proactive user experience, where the system guides user interactions in particular situations instead of only reacting to their interactions with the display, as we are generally used in classic interactive systems.

Human-IST Collaborators: Moreno Colombo, Himanshu Verma, Denis Lalanne

External Collaborators: -

Start date: March 2017

Back to top

 

Domain 6: Human-centered Data Privacy

Usable Privacy

Many users of technology have only a limited understanding of how their data is used by technology, and how using particular technologies can affect their privacy, both in the real and virtual worlds. This project aims to explore how users can be made more aware of what happens to their data and how to understand and assess privacy risks and benefits in various contexts by taking a holisitc approach to privacy. That is, by understanding privacy from the perspectives of technology (current practices and constraints), human behaviour and attitudes and legal, social and economic norms and practices, and using that understanding to design and develop prototypes of methods that help improve user awareness and understanding of their privacy.

Human-IST Collaborators: Agnes Lisowska Masson, Denis Lalanne

External Collaborators: Grace Eden (HES-SO Valais-Wallis)

Start date: April 2017 -

Back to top

 

Domain 7: Cognitive computing

Internet of Postal Things (IoPT)

In this project, we are experimenting with innovative postal services in the field of Internet of Things (IoT) sensors and incidental business models. The basic notion is to make better use of the physical Web Swiss Posts owns and to provide analyzed data of the IoT-sensors to customers or citizens. The project includes technical challenges (i.e. LoRa WAN, new kind of sensors, etc.) as well as business challenges (i.e. new business models as Freemium services, etc.), which should be explored experimentally, as they go hand-in-hand and therefore need to be addressed holistically (i.e. applying action-design-research). It is a design-science research project, where sociotechnical issues as pioneering citizen-centered Smart City services are addressed.

Human-IST Collaborators: Edy Portmann

External Collaborators: Swiss Post

Start date: September 2017

 

Cognitive Computing: An AI Hierarchy of Needs for novel Swiss Post Services

This project looks at postal data (e.g. scans, logins, sensor data, user generated content, etc.) to feed smart systems, which could provide users (e.g. citizens) with personalized and contextual information (e.g. within their city). The data come from different parts within Swiss Post (e.g. PostBus, PostFinance, PostGroup, etc.) as well as in different forms and granularity (e.g. ambiguous sensor data, natural language sources, etc.). The quest of this design-science research is, to design and engineer AI-systems, which offer value to the users (e.g. through providing users/citizens personalized information in the right context). The basic challenges thereby range from collecting data via exploring them through to self-learning systems, which may communicate with users/citizens (e.g. in natural language and with visualization).

Human-IST Collaborators: Edy Portmann

External Collaborators: Swiss Post

Start date: September 2017

 

Computing with Words and Perceptions

Computing with Words and Perceptions is a relatively new method of computing for computer systems. As words occur in the “natural language” and not in the numerical format, they are imprecise and cannot be processed by traditional systems. Because of this, a lot of relevant information gets lost. In order to change that, the team of the Human-ist examines linguistic computing and tries to create intelligent and smart computer systems. With the help of Computing with Words and Perceptions, new ways of computing should be developed which make the transformation of imprecise information into computer-readable data possible so that they can subsequently be applied in computations. It should be possible to include all necessary information in decision-making processes. It is thus the aim to make today’s processing of information more efficient such that the interaction between humans and computer systems can be optimized.

Human-IST Collaborators: Edy Portmann, Sara D’Onofrio

External Collaborators: Swiss Post

Start date: September 2017

 

Back to top