Analyzing Behavioral Attributes of Drivers and Implementing Safe Driving Model

Abstract

Behavioral characteristics of a driver include - frequency of change of lane, speed of the car with respect to average speed on a specific road during a specific time, proximity of the vehicle with other on-road objects and the instantaneous speed. All these characteristics, when analyzed together, give us the necessary parameters that can judge the performance of a driver. In this project, we have built an embedded system with the help of various sensors in order to retrieve live data from the car. Initially, the data retrieved will be used to create a model for acceptable driving. We propose a driver behavior recognition method using Hidden Markov Models (HMMs) to characterize and detect driving maneuvers and place it in the framework of the cognitive model of human behavior. Later on, with the help of live data retrieved, the responses of the driver will be compared with the responses of the cognitive model and the driver will be judged based on his responses to the situation.