Towards Robust Car Following Dynamics Modeling via Blackbox Models: Methodology, Analysis, and Recommendations

TL;DR

The paper investigates how the choice of target variable affects learning car-following dynamics across classical and data-driven models. It systematically compares classical CF models, Gaussian Processes, Kernel Ridge Regression, and LSTM using three target variables ($a$, $v$, $s$) on three diverse datasets, employing ANOVA to analyze interactions. The results show that optimal target variables depend on the model type and the space (function-space vs parameter-space) in which learning occurs, with classical CF models and LSTM favoring $s$, while GP/KRR favor $a$ (and $v$ in some cases). These findings highlight the nontrivial role of target selection in model calibration and provide guidance for selecting target variables and models in car-following and related traffic modeling applications across varied data regimes.

Abstract

The selection of the target variable is important while learning parameters of the classical car following models like GIPPS, IDM, etc. There is a vast body of literature on which target variable is optimal for classical car following models, but there is no study that empirically evaluates the selection of optimal target variables for black-box models, such as LSTM, etc. The black-box models, like LSTM and Gaussian Process (GP) are increasingly being used to model car following behavior without wise selection of target variables. The current work tests different target variables, like acceleration, velocity, and headway, for three black-box models, i.e., GP, LSTM, and Kernel Ridge Regression. These models have different objective functions and work in different vector spaces, e.g., GP works in function space, and LSTM works in parameter space. The experiments show that the optimal target variable recommendations for black-box models differ from classical car following models depending on the objective function and the vector space. It is worth mentioning that models and datasets used during evaluation are diverse in nature: the datasets contained both automated and human-driven vehicle trajectories; the black-box models belong to both parametric and non-parametric classes of models. This diversity is important during the analysis of variance, wherein we try to find the interaction between datasets, models, and target variables. It is shown that the models and target variables interact and recommended target variables don't depend on the dataset under consideration.

Figures (26)

• Figure 1: An LSTM cell
• Figure 2: The performance of four classical CF models with their respective learned parameters on (a) ASTAZERO ($s$) (b) JIANG ($s$). The black line on each plot indicates the 80/20% split between training and test data respectively.
• Figure 3:
• Figure 4:
• Figure 5: