Abstract:Over the past 50–60 years, numerous driver behavior models have been proposed in the literature. However, the literature still lacks models describing bus drivers' behavior in traffic streams, even though buses comprise a non-negligible component of the traffic mix in many cities. Further, bus driver behavior might differ from other vehicles due to the differences in size, kinematic characteristics, maneuvering capabilities, and the number of occupants. Moreover, human factors such as multi-vehicle anticipation and stimuli perception contribute to this difference in driver behavior. Motivated by these reasons, this study presents a new modeling framework for mimicking bus driver behavior. The framework incorporates two important aspects of bus driver behavior: multi-vehicle anticipation and stimuli perception. Based on the proposed modeling framework, the study modifies the widely used Intelligent Driver Model (IDM). A variance-based sensitivity analysis is carried out to recognize the influence of model parameters (specifically, the new parameters) on the output of the IDM model. The modified IDM model is calibrated and validated using an empirical trajectory dataset of about 90 buses from a traffic stream in Chennai, India. In doing so, the study also contributes to modelling driver behavior in heterogeneous and disorderly traffic streams found in Indian cities and elsewhere. The parameter calibration results show that the average calibrated parameters of the modified IDM offer realistic interpretations, and the calibration and validation errors are small. Furthermore, it is evident from the results that the perceived space gaps by bus drivers can be longer or shorter than the actual space gaps. Overall, the modified IDM model outperformed the original IDM, highlighting the efficacy of the proposed multi-vehicle anticipation and stimuli perception features in the model. Finally, the study also evaluates the model performance by analysing its stability and macroscopic properties.

A Taxonomy and Review of Algorithms for Modeling and Predicting Human Driver Behavior

A Review on Driving Behavior Models

Driver Modeling through Deep Reinforcement Learning and Behavioral Game Theory

Pedestrian Models for Autonomous Driving Part II: High-Level Models of Human Behavior

Augmented Driver Behavior Models for High-Fidelity Simulation Study of Crash Detection Algorithms

A model of dyadic merging interactions explains human drivers' behavior from control inputs to decisions

Behavior Prediction of Traffic Actors for Intelligent Vehicle Using Artificial Intelligence Techniques: A Review

A review of game theory models of lane changing

Beyond RMSE: Do machine-learned models of road user interaction produce human-like behavior?

Driver Lane Changing Behavior Modeling And Simulation

A Review of HMM-Based Approaches of Driving Behaviors Recognition and Prediction

Modeling Driver, Driver Support, and Cooperative Systems with Dynamic Optimal Control

Toward Computational Simulations of Behavior During Automated Driving Takeovers: A Review of the Empirical and Modeling Literatures

A Review of Data Analytic Applications in Road Traffic Safety. Part 1: Descriptive and Predictive Modeling

A human factors-based modeling framework to mimic bus driver behavior

Modelling communication-enabled traffic interactions

A model of communication-enabled traffic interactions

Social Interactions for Autonomous Driving: A Review and Perspectives

A Systematic Solution of Human Driving Behavior Modeling and Simulation for Automated Vehicle Studies

Agents for Traffic Simulation

Social Interaction-Aware Dynamical Models and Decision Making for Autonomous Vehicles