- CoreSAEP (Vidi): Computational Reasoning for Socially Adaptive Electronic Partners
- About: The overall aim is to develop a reasoning framework that combines logic and quantitative techniques for Socially Adaptive Electronic Partners (SAEPs) that adapt their behavior to norms and values of people. This becomes more and more important as technology becomes an integral part of our daily lives. The computational reasoning techniques are aimed at determining when and to what extent norm-compliance can be guaranteed, and deciding what to do if in exceptional situations a norm cannot or should not be complied with.
- My role: project leader
- Funded by: NWO, Vidi personal grant
- Period: 2014 - 2019
- Size: 800K
- COMMIT - Project 2: Interaction for Universal acces - Work package 7: Socio-geographical support
- About: Social cohesion has been identified as positively influencing quality of life, and various kinds of technology can be used to increase cohesion and actively promote social life within a community. We will investigate how to develop a so-called electronic partner capable of providing support to its users by helping them navigate their social and geographical environment. Our target group is elementary school children and their social environment (e.g., parents, teachers, etc.). We propose to achieve our aim by creating an electronic partner that can adapt its behavior to the goals of the user as well as to norms of other important people and institutions in the user's social context (like the school or sports club). As the user of the electronic partner plays an important role, we will use a situated Cognitive Engineering methodology to establish a sound theoretical, empirically founded, and human-centered requirements baseline. This means that we will define the criteria and expectancies of the use of the electronic partner and iteratively develop and evaluate the system through prototyping and user feedback.
- My role: Work package leader and supervision of PhD student Abdullah Kayal
- Funded by: FES (Dutch Government)
- Period: 2012 - 2016
- SHINE: Data Science for Environmental Monitoring in Urban Environments (Sensing Heterogeneous Information Network Environment)
- About: The starting point of the SHINE research project is simple: better data leads to better understanding leads to better decisions. This holds for both individual citizens and government authorities. For this reason, SHINE develops ICT techniques to collect, process, and visualize data that concerns different aspects of urban life. Heavy rainfall is a good
example of this. More information is available from our flyer. SHINE is associated with the Delft Data Science initiative.
- My role: project leader and supervision of PhD student Thomas King, Thesis "Governing Governance: A formal framework for analysing institutional design and enactment governance"
- Funded by: DIRECT, Delft Institute for Research on ICT
- Period: 2012 - 2016
- Size: funds 5 PhD students and a postdoc
- Coactive Design
- About: Coactive Design is a new approach that we develop to address the increasingly sophisticated roles for both people and agents in mixed human-agent systems. The fundamental principle of Coactive Design is that the underlying interdependence of participants in joint activity is a critical factor in the design of human-agent systems. The idea is that in order to enable appropriate interaction, an understanding of the potential interdependencies among groups of humans and agents working together in a given situation should be used to shape the way agent architectures and individual agent capabilities for autonomy are designed.
- My role: supervision of PhD student Matthew Johnson; collaboration with Florida Institute for Human & Machine Cognition
- Period: 2009 - 2014
- Thesis cum laude: Coactive Design: Designing Support for Interdependence in Human-Robot Teamwork
- Slim Verbinden (Smart Connections)
- About: Slim Verbinden addresses the problem of how to get the right information to the right people at the right time in disaster situations. Improving this will lead to reduced risk for police officers, ambulance staff, etc., as well as for victims and bystanders and it should limit the impact of the disaster on the environment and the economy. We investigate the use of explainable AI techniques for improving sharedness of mental models among people involved in a disaster, as it is known from social science that increased sharedness improves team performance.
- My role: researcher and supervision of postdoc Maaike Harbers
- Funded by: Agentschap NL
- Period: 2011 - 2012