Highway Discretionary Lane Changing Behavior Recognition Using Continuous and Discrete Hidden Markov Model

Abstract

Lane changing behavior is one of the most essential and complex driving attributes. The lack of proper lane changing behavior can lead to collisions and traffic congestion. During highway driving, a human driver is able to estimate the intentions and behaviors of surrounding vehicles. In this paper, a driving behavior predictor model is proposed and evaluated. The proposed model is based on continuous Hidden Markov Model (CHMM) and discrete Hidden Markov Model (DHMM). The developed model has been evaluated using the real-world dataset of Interstate 80 and U. S. Highway 101 from Federal Highway Administration's Next Generation Simulation (NGSIM). The Time to Collision (TTC), velocity and acceleration of the target and surrounding vehicles are taken as input parameters in the model execution. The test results show that the proposed model exhibits 85.65% accuracy for left lane change, 86.06% accuracy for right lane change and 95.87% accuracy for lane keep behaviors. This model can be effectively used as a lane changing suggestion system in the advanced driver assistance systems (ADAS).