Development of framework for in-vehicle collision warning system considering driver characteristics

Passing collisions are one of the most serious traffic safety problems on two-lane highways. The purpose of this dissertation is to develop a framework for a passing collision warning system (PCWS) for two-lane highways that can help drivers avoid passing collisions by reducing the chance of human error. Specifically, the objectives of this research are: (1) to conduct a comprehensive literature review of existing collision warning systems, (2) to design driving simulator and field experiments for data collection, (3) to develop deterministic and reliability-based models for passing sight distance (PSD) that consider driver characteristics, (4) to develop an in-vehicle passing collision warning system, and (5) to develop a Simulink model that implements the proposed PCWS in a real time environment. A driving simulator was used to collect data from participants (males and females of different ages). The field study data werecollected on real highways using an in-vehicle video camera and a GPS data logger. The driving simulator and field data were used to develop the passing parameters for the proposed PSD model. The estimated parameters included initial time, passing time, and average acceleration rate. The results of the model were compared with those from existing models and design guidelines. The results revealed that the existing PSD models were either too liberal or too conservative. The reliability-based PSD model was developed using the First-Order Second-Moments method and a Monte Carlo Simulation was used to validate the model. The proposed PCWS uses a radar sensor placed in the passing vehicle to detect opposing vehicles travelling in the left lane and calculate their relative distance and speed in order to estimate the time to collision. This time is then compared with the time required for the passing vehicle to clear the path. The “safe pass” signal can assist passing drivers in preparing for a safe passing maneuver during the overtaking process. A Simulink MATLAB model was developed and used to implement the methodology of the proposed warning system. The different factors that affect system accuracy were examined. The application of the system was illustrated using an example.