Infer causality relationships in the motion of manned and unmanned vehicles using transfer entropy

Recently the interest in autonomous vehicles increases a lot, many studies are made and the researchers investigate the relationship between automatic vehicles and human being. Those studies focus on the relationship between autonomous vehicles and pedestrians or the drivers and the autonomous vehicle. In fact the presence of autonomous vehicle on the road is reality in some country, and it is fundamental to know how we can improve their efficiency. What we actually want to study is the relationship between the autonomous vehicles and the manned vehicles, since it is unlikely that since the beginning the autonomous vehicles will be the only ones on the streets. In this thesis we aim to understand the cause-and-effect relationship between the unmanned and manned vehicle. In order to verify this relationship we apply first a symbolic analysis and afterwards the transfer entropy analysis. We use and create many symbols and we test them studying how they performs with three different cases: when the two vehicles move in opposite direction facing each other, when the two vehicles proceed along parallel routes, and finally the case in which the two vehicles have to intersect their routes. On a secondary moment we make a parametric study using one of the symbol that better performs in the first study. In particular we analyze specific parameters that distinguish the number of total symbols of a type, and we study how the transfer entropy varies with the parameters.