Cooperative lane change optimization with enhanced modeling of target-lane following vehicles

Merging lane-changing vehicle (LC) often cause sudden acceleration changes in the following vehicle in the target lane (TF), compromising driving comfort and increasing safety risks. Although the traditional Intelligent Driver Model (IDM) effectively captures car-following behavior between individual vehicles, it is less effective in scenarios involving multiple lead vehicles. Therefore, this study proposes an improved smooth transition IDM (ST-IDM) to model the response behavior of the TF during the transition from following its original leading vehicle to following a LC vehicle. Specifically, an acceleration transition function is introduced to quantify the car-following shift of the TF, and a transition weight is adjusted according to the time-to-collision (TTC). The ST-IDM is further incorporated into a cooperative lane-changing optimization framework, where a dual-objective optimization function is formulated based on lane-changing and interference costs. Empirical and Monte Carlo simulation data are used to analyze model adaptability under different scenarios and to investigate the results of a dual-objective optimization, as well as the impact of varying weightings. The results show that the proposed model effectively captures the dynamic response of TFs during lane changes, reducing RMSE by 23.1 % in the voluntary lane-change scenario and by 10.7 % in the forced lane-change scenario compared with baseline models. Moreover, the cooperative lane-changing model remains robust under stochastic perturbations, maintaining stable collective traffic states. These findings demonstrate that the proposed model enhances cooperative decision-making between LC and TF and can be applied to traffic flow simulation and optimization.