Principal Investigator: Praveen Edara
Co-Principal Investigator: Carlos Sun (firstname.lastname@example.org
About this Project
Brief Project Description & Background
Merging on freeways near lane drops is one of the most prominent safety risks that drivers consistently experience. A better understanding of the decision making behavior of drivers for merging will lead to the design of safer driver assistance systems and other safety interventions. Simulation models that incorporate realistic and accurate driving behavior models for merging will lead to accurate assessment of traffic operations. The next generation simulation (NGSIM) dataset provides detailed vehicle trajectories that can be used to investigate the merging behavior and to build models that replicate such behavior. In this project, several artificial intelligence methods will be explored for modeling merging behavior of drivers observed in the sites chosen from NGSIM data. The prediction accuracies of the developed models will be compared with those in the literature.
The PIs plan to apply different methods to model the merging behavior of drivers. Discrete choice methods such as Logit, artificial intelligence methods such as fuzzy logic and pattern classification will be explored.
Accurate driving behavior models for freeway merges will result from this project. The models will contribute to the development of better simulation models and safer driver assistance systems.
Changing lanes is an essential component of driving on freeways. Lane changes become mandatory if a lane drop occurs on roadways – natural lane drops and work zone or incident induced