Mandatory Lane Change Strategy And Algorithm For VANET Using Stackelberg Model

Mandatory lane change occurs when the vehicle has no option and required to shift lane,mostly because of the geographical condition,such as,exits,on and off-ramps.It happens in expressway,mostly because of several lane closures and ramps.High-speed characteristics of driving in expressways makes it more dangerous for vehicles to do the mandatory lane maneuver.This dissertation studies Stackelberg equilibrium to give drivers a better strategy of decision making for a more efficient maneuver.As part of the game theory,Stackelberg naturally has payoffs the vehicles have to measure before executing lane change.We add drivers' aggressiveness to the other payoffs;speed payoff and comfort payoff.Therefore,all of these payoffs will be calculated within twenty steps ahead every time the vehicle moves a step forward.To predict the motions of surrounding vehicles a few time steps in the future,Model predictive controller(MPC)is utilized.Furthermore,the dissertation is divided into two different following models,the classic two-seconds following model and SUMO's default,Krauss following model,to see which one results in better timing and safety between the two.Thus,Stackelberg method and MPC will be combined with these two following models.MPC and the game to output the best time for subject vehicle to shift.To test if the dissertation's methods offer better efficiency,first,the simulation will be held in three different traffic densities with different numbers of vehicles for each density.Then,we simulate the algorithms.Finally,to determine if the proposed algorithms are more or less efficient,the dissertation extracted the simulation's output to compare its both algorithms to the other existing algorithms and breaking the results down to various indicators.