Behavioral decision‐making model of the intelligent vehicle based on driving risk assessment

Abstract

AbstractIntelligent‐driving technologies play crucial roles in reducing road‐traffic accidents and ensuring more convenience while driving. One of the significant challenges in developing an intelligent vehicle is how to operate it safely without causing fear in other human drivers. This paper presents a new behavioral decision‐making model to achieve both safety and high efficiency and also to reduce the adverse effect of autonomous vehicles on the other road users while driving. Moreover, we attempt to adapt the model for human drivers so that users can understand, adapt, and utilize intelligent‐driving technologies. Furthermore, this paper proposes a combined spring model for assessing driving risk. Thus, we analyze some driving characteristics of drivers and choose “safety” and “high efficiency” as the two main factors that are pursued by drivers while driving. Based on the principle of least action, a multiobjective optimization cost function is established for the decision‐making model. Finally, we design six unprotected left‐turn scenarios at a T‐intersection and three unprotected left‐turn scenarios at a standard two‐lane intersection for applying simulation algorithm and provide a decision‐making map for developing intelligent‐driving technologies. Based on the principle of least action, this paper demonstrates that optimization theory can give insight into drivers’ behavior and can also contribute to the development of intelligent‐driving algorithms. The experimental results reveal that the behavioral decision‐making model can always avoid collision accidents on the premise of ensuring certain efficiency, and it can achieve 97.01%, 94.52%, 96.67%, 91.18%, 101.27%, 83.33%, 102.94%, 103.03%, and 105.77% of time to intersection's maximum pass rate in the considered nine scenarios.