The HIANIC project will develop new technologies in term of autonomous navigation in dense and human populated traffic. It will explore the complex problem of navigating autonomously in shared-space environments, where pedestrians and cars share the same environment. Such a system will contribute both to urban safety and intelligent mobility in “shared spaces”.

Our team (HAwAI) is focused on developing a realistic simulation of pedestrians in interaction with autonomous vehicles in shared spaces. My post-doc is part of that project. My goal is to develop a realistic human behavior model of individual pedestrian perception and interaction with car. Such interactions include for exemple, stop and give way, change trajectory in function of perceived car trajectory or speed up to cross before the car is coming.

In this work, we are interested in the modelling of the human behaviour in natural hazard situations such as bushfires. We model the behaviour and social interactions between different actors involved in bushfires such as firefighter, police, emergency centre managers and civilians. We use the Belief, Desire and Intention (BDI) architecture to model realistic human behavior, and the FIPA-ACL standard to model the communications. We use geospatial data to represent the environment in a realistic way. The goal of this work is twofold: (1) provide a tool for emergency managers and stakeholders to allow them to explore different strategies and scenarios; (2) provide a serious game for human training based on that model.

In this project we were interested to understand why some children have difficulties when they read a text and how their lexical knowledge influences the reading ability. Indeed, learning to read requires the coordination of eye movements, attention, as well as lexical processing in order to achieve smooth reading. Contrary to what we perceive, the behaviour of our eyes is very complex and depends on several factors. We have analyzed the children's eye-movement while they read a text with statistical methods and with a computational model which can simulate the reading processes spatially and temporally in order to better understand why we observe differences in reading patterns.