AbstractsTransportation

Accidents at intersections are one of the most common causes of fatalities on roads. Statistics from the EU and the USA show that fatalities at intersections represent more than one fifth of all traffic fatalities. In particular, the left turn across the path of a vehicle coming from the opposite direction (LTAP/OD) is one of the riskiest situations at intersections. The goal of the thesis is to develop driver models describing human behaviour during left-turn-across path scenarios, for future use in active safety systems development for intersections. Video and vehicle dynamics data during daily driving were retrieved from the EuroFOT project database in order to find LTAP/OD scenarios suitable for the scope of the thesis. For the description of driver’s behaviour during the LTAP/OD scenario, two different driver modelling approaches have been considered: the Salvucci and Gray’s, which focuses on steering behaviour, and Nobukawa’s, which is based on vehicle acceleration. The work has been performed using MATLAB®, exploiting also genetic algorithms for model parameter optimization. The models describe the steering and speed behaviour reasonably well for the specific LTAP/OD events used in the model optimization and validation. However, generalization cannot be made across LTAP/OD as a scenario. Further efforts are required, e.g., to increase of number of intersections and number of drivers in the validation, to verify the reliability of the two models and develop it further. Once reliability has been established, the models may be used as part of the design of Intersection Driver Support Systems.