Modeling and Analysis of the Lane-Changing Execution in Longitudinal Direction

Lane changing is a common driving behavior in traffic, and the algorithm of lane-changing maneuver is an indispensable part of the design of autonomous vehicle control and adaptive cruise control. Although extensive research studies focused on modeling drivers' decision mechanism, lane-changing execution (LCE) happening after the lane-changing decision has not attracted much attention, which has a significant impact on driving safety and traffic simulation results. This paper attempts to replicate the real LCE behavior by proposing new LCE models. Depending on whether the lag vehicle on the target lane is considered in an LCE or not, two types of lane-changing execution are defined, namely, the cooperative LCE (CLCE) and the forced LCE (FLCE). The real vehicle trajectory data, i.e., the NGSIM data, are applied to train and test the proposed models. The results illustrate that the proposed models have good performance in replicating the CLCE and FLCE behavior and outperform the first LCE model proposed by Moridpour et al. Furthermore, when drivers decide to conduct a lane-changing execution, their considerations of the vehicles on the target lane have happened, and their considerations of the vehicles on the current lane decrease sharply during the LCE. In addition, the driver generally pays more attention to the preceding vehicle than to the lag vehicle on the target lane in an LCE.