Reinforcement learning (RL)–based car-following (CF) control strategies have attracted significant attention in academia, emerging as a prominent research topic in recent years. Most of these control strategies focus solely on the motion status of the immediately preceding vehicle. However, with the development of vehicle-to-vehicle (V2V) communication technologies, intelligent vehicles such as connected autonomous vehicles (CAVs) can gather information about surrounding vehicles. Therefore, this study proposes an RL-based CF control strategy that takes multivehicle scenarios into account. First, the trajectories of two preceding vehicles and one following vehicle relative to the subject vehicle (SV) are extracted from a highD dataset to construct the environment. Then the twin-delayed deep deterministic policy gradient (TD3) algorithm is implemented as the control strategy for the agent. Furthermore, a sequence-to-sequence (seq2seq) module is developed to predict the uncertain motion statuses of surrounding vehicles. Once integrated into the RL framework, this module enables the agent to account for dynamic changes in the traffic environment, enhancing its robustness. Finally, the performance of the CF control strategy is validated both in the highD dataset and in two traffic perturbation scenarios. In the highD dataset, the TD3-based prediction CF control strategy outperforms standard RL algorithms in terms of convergence speed and rewards. Its performance also surpasses that of human drivers in safety, efficiency, comfort, and fuel consumption. In traffic perturbation scenarios, the performance of the proposed CF control strategy is compared with the model predictive controller (MPC). The results show that the TD3-based prediction CF control strategy effectively mitigates undesired traffic waves caused by the perturbations from the head vehicle. Simultaneously, it maintains the desired traffic state and consistently ensures a stable and efficient traffic flow.

Most of the existing research in the field of autonomous vehicles (AVs) addresses decision making, planning and control as separate factors which may affect AV performance in complex driving environments. A hierarchical framework is proposed in this paper to address the problem mentioned above in environments with multiple lanes and surrounding vehicles. Firstly, high-level decision making is implemented by a finite-state machine (FSM), according to the relative relationship between the ego vehicle (EV) and the surrounding vehicles. After the decision is made, a cluster of quintic polynomial equations is established to generate the path connecting the initial position to the candidate target positions, according to the traffic states of the EV and the target vehicle. The optimal path is selected from the cluster, based on the quadratic programming (QP) framework. Then, the speed profile is generated, based on the longitudinal displacement–time graph (S–T graph), considering the vehicle motion state constraints and collision avoidance. The smoothed speed profile is created through another QP formulation, in the space created by the dynamic-programming (DP) method. Finally, the planned path and speed profile are combined and sent to the lower level control module, which consists of the linear quadratic regulator (LQR) for lateral trajectory tracking control, and a double PID controller for longitudinal control. The performance of the proposed framework was validated in a co-simulation environment using PreScan, MATLAB/Simulink and CarSim, and the results demonstrate that the proposed framework is capable of addressing most ordinary scenarios on a structured road, with reasonable decisions and controlling abilities.

With the technical support of an intelligent networking environment, autonomous driving technology is facing a new stage of development, and the decision-making behavior of autonomous vehicles is changing fundamentally, so it is urgent to explore the lane-changing decision-making behavior mechanism of autonomous driving. Firstly, through the analysis of system similarity, the similarity between autonomous vehicles and moving molecules is sought, and the attraction and repulsion between molecules are applied to the lane-changing process of vehicles to effectively recognize the traffic scene of lane-changing vehicles. Secondly, the molecular interaction potential is introduced to unify the attraction and repulsion, and explore the dynamic influencing factors of lane-changing behavior for vehicles. Moreover, we systematically analyze the interaction relationship in the lane-changing process of Connected and Automated Vehicles, and establish the molecular interaction potential lane-changing model to explore the lane-changing decision-making behavior mechanism. Furthermore, we study the impact of micro lane-changing behavior on macro traffic flow. Finally, the SL2015 lane-changing model and the molecular interaction potential lane-changing model are compared and analyzed by using the SUMO platform. The results show that the speed fluctuation of Connected and Automated Vehicles based on the molecular interaction potential lane-changing model is reduced by 15.5%, and the number of passed vehicles is increased by 3.26% on average, which has better safety, stability, and efficiency. The molecular interaction potential modeling of lane-changing decision-making behavior for Connected and Automated Vehicles comprehensively considers the interaction relationship of dynamic factors in the traffic environment, and scientifically shows the lane-changing decision-making mechanism of Connected and Automated Vehicles.

To study the lane-change interaction characteristics of intelligent connected vehicles (ICVs) and reduce the risk of vehicle lane-changing decisions, a decision model based on the lane-changing game characteristics of the ICV is proposed in this paper. In the modeling process, the characteristics of vehicle lane-changing interaction behavior are analyzed based on evolutionary game theory and the vehicle game lane-changing payoff functions are quantified. The stability of the game equilibrium points is analyzed by using a dynamic evolution equation, and sensitivity analysis of the main factors affecting vehicle lane changes and the time to the collision of vehicles is conducted. The SUMO software is used to simulate and verify the vehicle game decision model, and the results show that the game decision system converges to different optimal strategy combinations under different traffic conditions, and this model can effectively reduce the decision-making conflict and the collision risk of vehicles.

To systematically study the influence of a propensity for a particular driving style on car-following risk, a safety potential field risk-quantification method that considers driving style is proposed. First, we classify driving behaviors and construct a field-based safety potential car-following model via analogy with intermolecular interactions; second, we establish a risk-quantification model by considering driving style, risk exposure, and risk severity and classify the consequent risk into four levels, high risk, medium risk, low risk, and safe, using the fuzzy C-means algorithm. Finally, we predict the car-following risk using the LightGBM algorithm in real time. The experimental results show that the LightGBM algorithm can recognize up to 86% of medium–high risk levels compared to the Decision Tree and Random Forest Algorithms. It can achieve effective prediction of car-following risk, which provides sufficient warning information to drivers and helps improve the overall safety of vehicle operation.

Along with the rapid development of autonomous driving technology, autonomous vehicles are showing a trend of practicality and popularity. Autonomous vehicles perceive environmental information through sensors to provide a basis for the decision making of vehicles. Based on this, this paper investigates the lane-changing decision-making behavior of autonomous vehicles. First, the similarity between autonomous vehicles and moving molecules is sought based on a system-similarity analysis. The microscopic lane-changing behavior of vehicles is analyzed by the molecular-dynamics theory. Based on the objective quantification of the lane-changing intention, the interaction potential is further introduced to establish the molecular-dynamics lane-changing model. Second, the relationship between the lane-changing initial time and lane-changing completed time, and the dynamic influencing factors of the lane changing, were systematically analyzed to explore the influence of the microscopic lane-changing behavior on the macroscopic traffic flow. Finally, the SL2015 lane-changing model was compared with the molecular-dynamics lane-changing model using the SUMO platform. SUMO is an open-source and multimodal traffic experimental platform that can realize and evaluate traffic research. The results show that the speed fluctuation of autonomous vehicles under the molecular-dynamics lane-changing model was reduced by 15.45%, and the number of passed vehicles was increased by 5.93%, on average, which means that it has better safety, stability, and efficiency. The molecular-dynamics lane-changing model of autonomous vehicles takes into account the dynamic factors in the traffic scene, and it reasonably shows the characteristics of the lane-changing behavior for autonomous vehicles.