Car-following (CF) algorithms are crucial components of traffic simulations and have been integrated into many production vehicles equipped with Advanced Driving Assistance Systems (ADAS). Insights from the model of car-following behavior help researchers to understand the causes of various macro phenomena that arise from interactions between pairs of vehicles. Car-following Models encompass multiple disciplines, including traffic engineering, physics, dynamic system control, cognitive science, machine learning, deep learning, and reinforcement learning. This paper presents an extensive survey that highlights the differences, complementarities, and overlaps among microscopic traffic flow and control models based on their underlying principles and design logic. It reviews representative algorithms, ranging from theory-based Kinematic Models, Psycho-Physical Models, and Adaptive Cruise Control Models to Learning-based algorithms like Reinforcement Learning (RL) and Imitation Learning (IL). To acknowledge the potential impact on CF models, Large GenAI Models are also included as Knowledge- Driven category. This manuscript discusses the strengths and limitations of these models and explores their applications in different contexts. This review synthesizes existing researches and available datasets across different domains to fill knowledge gaps and offer guidance for future research by identifying the latest trends in car following models and their applications.

In this study, we developed the solution of roadside LiDAR object detection using a combination of two unsupervised learning algorithms. The 3D point clouds are firstly converted into spherical coordinates and filled into the elevation-azimuth matrix using a hash function. After that, the raw LiDAR data were rearranged into the new data structure to store the information of range, azimuth, and intensity. Then, the dynamic mode decomposition method is applied to decompose the LiDAR data into low-rank backgrounds and sparse foregrounds based on intensity channel pattern recognition. The coarse-fine triangle algorithm (CFTA) automatically finds the dividing value to separate the moving targets from static background according to range information. After intensity and range background subtraction, the foreground moving objects will be detected using a density-based detector and encoded into the state-space model for tracking. The output of the proposed solution includes vehicle trajectories that can enable many mobility and safety applications. The method was validated at both path and point levels and outperformed the state of the art. In contrast to the previous methods that process directly on the scattered and discrete point clouds, the dynamic classification method can establish the less sophisticated linear relationship of the 3D measurement data, which captures the spatial-temporal structure that we often desire.

Reinforcement Learning from Human Feedback (RLHF) is popular in large language models (LLMs), whereas traditional Reinforcement Learning (RL) often falls short. Current autonomous driving methods typically utilize either human feedback in machine learning, including RL, or LLMs. Most feedback guides the car agent's learning process (e.g., controlling the car). RLHF is usually applied in the fine-tuning step, requiring direct human "preferences," which are not commonly used in optimizing autonomous driving models. In this research, we innovatively combine RLHF and LLMs to enhance autonomous driving safety. Training a model with human guidance from scratch is inefficient. Our framework starts with a pre-trained autonomous car agent model and implements multiple human-controlled agents, such as cars and pedestrians, to simulate real-life road environments. The autonomous car model is not directly controlled by humans. We integrate both physical and physiological feedback to fine-tune the model, optimizing this process using LLMs. This multi-agent interactive environment ensures safe, realistic interactions before real-world application. Finally, we will validate our model using data gathered from real-life testbeds located in New Jersey and New York City.

Vehicle merging is a tactical, dynamic optimizing process with a series of decision-making operations. Existing microscopic lane-changing simulation models have been criticized that they do not well capture the important sequential decision-making process during merging because of the lack of deeply investigating field data. This study first analyzes merging tactics based on noise-filtered next-generation simulation (NGSIM) data, and then a new combined sequential decision-making model was proposed for candidate gap generation, targeting gap selection, merge location choice, and acceleration decisions of the merging vehicles. This sequential decision-making model could dynamically simulate the merging vehicles’ choice of targeting gaps and merging tactics under the effects of changing traffic conditions. How the merging vehicles behave after rejecting their current adjacent gap and synchronize with the speed of the main-lane traffic during merging is embodied in this model. Such aspects were oversimplified and against reality in existing merging models. The proposed new merging model was calibrated and validated with a U.S. Highway 101 (U.S.101) data set. The superiority of this new framework is highlighted in comparison with other two merging models. The results demonstrate that the proposed model has an acceptable accuracy to reproduce the trajectories of merging vehicles. The simulated vehicles’ trajectories show that the gap-selection model and targeting gap approaching algorithm can well simulate the driving behavior of the merging vehicle, which carries gap-rejection experience. The empirical field data analysis and model calibration results in this study provide several useful thoughts on building merging models to capture more real-world freeway merging behaviors.