An intelligent systems framework for prototyping tactical driving decisions

This paper discusses a driving simulation framework, which focuses on the decision-making role performed at the tactical driving level. Our tactical driving technique using Genetic Algorithms (GAs), named GA-INTACT, accounts for the positions of a 'subject' vehicle and those of other vehicles and their speed parameters in the surrounding traffic condition, and selects favourable speed change and lane transition actions for the 'subject' vehicle, according to safety, speed and driving behaviour criteria. The use of GAs for obtaining near-optimum driving solutions eliminates the need for learning driving patterns, and handles the complex nature of modelling tactical driving. In particular, the proposed framework addresses the role of driving behaviour in influencing drivers' tactical decisions – a matter that is not treated systematically and sufficiently by other driving simulation methods