Longitudinal vehicle platoon control has attracted much attention due to its potential for significantly mitigating traffic congestion, reducing the incidence of traffic accidents and improving fuel economy. To include platoon motion control in the two‐dimensional situations, this study proposes a lateral control system for vehicle platoon considering vehicle dynamic characteristics for enhancing the traditional longitudinal platoon control (one‐dimensional). The lateral control strategies are applied into two specific driving conditions: moving in parallel and moving in order which are representative in platoon lateral control applications and can be extended into a broader driving scenario. An integrated controller with motion control layer and actuator control layer is introduced to achieve the proposed lateral spacing control strategies. Three corresponding simulations are used to validate the proposed control system, and the results demonstrate that: during the two specific driving scenario, moving in order and moving in parallel, show different motion characteristics; platoon with proposed control system of platoon can realise a general curved driving task with a changing speed; the lateral following errors are influenced by the difference of dynamics characteristics between the four‐wheel independent driving vehicle and the front steering vehicle which should be taken into consideration in determination of expected lateral space.

Game theory has been extensively studied in recent years for its potential benefits on solving interactions between autonomous vehicles in platoon control. In this paper, we propose two behavioral decision-making approach based on non-cooperative game theory with both the complete information and incomplete information for cooperative vehicle platoon systems. The non-cooperative game payoff function takes the platooning performances of economy, comfort, safety, and further achieving self-driving functions into consideration. For the driving situations with incomplete information, a belief pool is constructed to represent the action probability for different behavioral types of vehicles, which will be updated by combining the driving intention identification with a Bayesian probability formula. With this, the stable strategies can be obtained for the two potentially conflicting parties, ensuring that neither of them has a motivation to change their driving behavior. Finally, the simulation results demonstrate that with the proposed behavioral decision-making approaches with complete information, the cooperative platoon performance can be markedly improved; with incomplete information, the platoon can determine the behavioral types of conflicting vehicles and complete collaborative decision-making approaches for solving not only the simple road-rights conflicting problems, but also for the extended vehicle platoon driving scenarios in a more complex transportation system.

The Controller Area Network with Flexible Data-Rate (CAN-FD) bus is the predominant in-vehicle network protocol, responsible for transmitting crucial application semantic signals. Due to the absence of security measures, CAN-FD is vulnerable to numerous cyber threats, particularly those altering its authentic physical values. This paper introduces Physical Semantics-Enhanced Anomaly Detection (PSEAD) for CAN-FD networks. Our framework effectively extracts and standardizes the genuine physical meaning features present in the message data fields. The implementation involves a Long Short-Term Memory (LSTM) network augmented with a self-attention mechanism, thereby enabling the unsupervised capture of temporal information within high-dimensional data. Consequently, this approach fully exploits contextual information within the physical meaning features. In contrast to the non-physical semantics-aware whole frame combination detection method, our approach is more adept at harnessing the physical significance inherent in each segment of the message. This enhancement results in improved accuracy and interpretability of anomaly detection. Experimental results demonstrate that our method achieves a mere 0.64% misclassification rate for challenging-to-detect replay attacks and zero misclassifications for DoS, fuzzing, and spoofing attacks. The accuracy has been enhanced by over 4% in comparison to existing methods that rely on byte-level data field characterization at the data link layer.

Recently, spacing policies of the vehicular platoon have been widely developed to enhance safety, traffic efficiency, and fuel consumption. However, the integrated spacing policies aim to maximum overall benefit, and the distributed spacing policies intense to get optimal monomer benefit. Ignoring the fairness of the benefit allocation of each vehicle will reduce the motivation to constitute the platoon. To fill this critical gap, this study proposes a spacing allocation method by treating spacing decisions as cooperative games. A flock’s model which is used to be the payoff function is introduced based on bionic motion principles. We present a characteristic function of the platoon for the cooperative game model considering the specific structure of the platoon. The τ value, Shapley value, and average lexicographic value are introduced and applied to allocate the spacing fairly. Proposed methods are compared with constant distance policy in some typical situations. The simulation results demonstrate that the spacing policy based on cooperative game theory improved the stable time for consistency control and the convergence of longitudinal following error.

Autonomous driving has demonstrated impressive driving capabilities, with behavior decision-making playing a crucial role as a bridge between perception and control. Imitation Learning (IL) and Reinforcement Learning (RL) have introduced innovative approaches to behavior decision-making in autonomous driving, but challenges remain. On one hand, RL’s policy networks often lack sufficient reasoning ability to make optimal decisions in highly complex and stochastic environments. On the other hand, the complexity of these environments leads to low sample efficiency in RL, making it difficult to efficiently learn driving policies. To address these challenges, we propose an innovative Knowledge Distillation-Enhanced Behavior Transformer (KD-BeT) framework. Building on the successful application of Transformers in large language models, we introduce the Behavior Transformer as the policy network in RL, using observation–action history as input for sequential decision-making, thereby leveraging the Transformer’s contextual reasoning capabilities. Using a teacher–student paradigm, we first train a small-capacity teacher model quickly and accurately through IL, then apply knowledge distillation to accelerate RL’s training efficiency and performance. Simulation results demonstrate that KD-BeT maintains fast convergence and high asymptotic performance during training. In the CARLA NoCrash benchmark tests, KD-BeT outperforms other state-of-the-art methods in terms of traffic efficiency and driving safety, offering a novel solution for addressing real-world autonomous driving tasks.

Previous research ignores the integration of the states of all vehicles in the platoon and static road information and uses only the lead vehicle or the average state for planning, which will lose the optimality of the platoon economy. The platoon's distributed optimization strategy is difficult to satisfy the optimality of the two-layer optimizer simultaneously. In this paper, we propose the centralized cloud-based platoon predictive cruise control (C-CPPCC) for highway scenarios, enhancing the platoon's efficiency and stability, while addressing the issue of increasing fuel consumption in existing platoon predictive cruise control (PPCC) systems. The C-CPPCC employs a centralized architecture, integrating the status information of platoon vehicles in the cloud to create a longitudinal planning model. It considers platoon stability and economical driving, introducing a centralized PPCC algorithm with rolling distance domains. The method's effectiveness was demonstrated by simulations using real vehicle and road data. Compared to the predecessor-leader following cruise control platoon (PLF-CC) and the distributed cloud-based PPCC (D-CPPCC), the C-CPPCC reduces fuel consumption by 7.52% and 1.57%, respectively. The fuel consumption increase for platoon-following vehicles is significantly reduced with the proportion from 9.25% and 1.59% to 0.07%. Simulation results verify the effectiveness of the proposed strategy.