Artificial neural network modelling of driver handling behaviour in a driver-vehicle-environment system

Abstract

Modelling driver handling behaviour in a driver-vehicle-environment (DVE) system is essentially useful for the design of vehicle systems and transport systems in the light of the safety and efficiency of human mobility. Driver handling behaviour is reflected in two aspects: the mental workload and the performance. Further, this behaviour is exposed through the interactions between driver-vehicle and driver-environment. There is generally a lack of the first principle with which to develop a model for human behaviour. In this study, several more sophisticated artificial neural network architectures for developing models for human drivers in a DVE system were used. The vehicle dynamics are modelled by a 3-d.o.f. model derived from the first principle. The experiment was performed and compared with a DVE simulation system in which the developed human driver behaviour model was included, together with the vehicle dynamics model. The comparative study showed that the simulation result is in good agreement with the experimental result, which further justifies the effectiveness of the developed driver behaviour model.