The sensor‐enabled in‐vehicle communication and infrastructure‐centric vehicle‐to‐everything (V2X) communications have significantly contributed to the spark in the amount of data exchange in the connected and autonomous vehicles (CAV) environment. The growing vehicular communications pose a potential cyber security risk considering online vehicle hijacking. Therefore, there is a critical need to prioritize the cyber security issues in the CAV research theme. In this context, this paper presents a cyber security analysis of connected vehicle traffic environments (CyACV). Specifically, potential cyber security attacks in CAV are critically investigated and validated via experimental data sets. Trust in V2X communication for connected vehicles is explored in detail focusing on trust computation and trust management approaches and related challenges. A wide range of trust‐based cyber security solutions for CAV have been critically investigated considering their strengths and weaknesses. Open research directions have been highlighted as potential new research themes in CAV cyber security area.

The paper critically evaluates the CAV structure and component vulnerabilities, validates the identified attacks, and analyzes the trust‐based security solutions categorized according to the suggested trust taxonomy.image

The ability to predict the trajectory of an autonomous vehicle accurately is crucial for safe and efficient navigation. However, predicting diverse and multimodal futures can be challenging. Recent approaches such as attention and graph neural networks have achieved state‐of‐the‐art performance by considering agent interactions and map contexts. This study focused on multi‐agent prediction using an agent‐centric approach with transformers. This enables parallel computation and a comprehensive understanding of the environment. Two main features are introduced: an adaptive receptive field (ARF) that captures the relevant surroundings for each agent, and perception encoding, which serves as spatial context embeddings. The ARF adapts to the agent's velocity and rotation, focusing attention ahead at high speeds or to the sides when it is slower. Perception encoding divides agents or lanes into levels and encodes the information of each level. This approach enables the efficient encoding of complex spatial relationships. The proposed method combines these advances with transformer modelling for multi‐agent trajectory prediction while ensuring real‐time prediction capabilities. The approach is evaluated on the Argoverse benchmark and better performance than the state‐of‐the‐art baseline is achieved. By addressing challenges such as multimodal outputs and robustness, the study enhances the safety and efficiency of autonomous driving systems by more accurately predicting trajectories.

This study proposes an agent‐centric approach to accurately predict the trajectory of autonomous vehicles. The approach utilizes an adaptive receptive field that adjusts to the agent's velocity and rotation, focusing attention on relevant surroundings. Additionally, perception encoding divides agents or lanes into levels, allowing for the efficient encoding of complex spatial relationships. The proposed method combines these advancements with transformer modelling, achieving better performance than the state‐of‐the‐art baseline and enhancing the safety and efficiency of autonomous driving systems by accurately predicting diverse and multimodal trajectories.image

Connected and automated vehicles (CAVs) are being developed and designed to operate on existing roads. Their safe and efficient operation during roadworks, where traffic management measures are often introduced, is crucial. Two alternative measures are commonly applied during roadworks on motorways: (i) closing one or multiple lanes (ii) narrowing one or all lanes. The former can cause delays and increased emissions, while the latter can pose safety risks. This study uses a VISSIM‐based traffic microsimulation to compare the effectiveness of these two strategies on traffic efficiency and safety, considering various market penetration rates (MPR) of CAVs. The model was calibrated and validated with the data collected from M1 motorway in the United Kingdom. Results show that average delays per vehicle‐kilometre‐travelled decreased from 102.7 to 2.5 s (with lane closure) and 23.6 to 0.6 s (with narrow lanes) with 0% and 100% CAV MPR, respectively. Moreover, safety in narrow lanes improved by 4.8 times compared to 1.5 times improvement in lane closure with a 100% CAV MPR; indicating that narrow lanes would result in better safety performance. These findings could assist transport authorities in designing temporary traffic management measure that results in better CAV performance when navigating through roadworks.

A traffic microsimulation model was developed to evaluate two temporary traffic management measures under varying connected and autonomous vehicles market penetration rates. The model, validated with data from the M1 motorway in the UK, showed decreased delays and improved safety, particularly in narrow lanes, with increased connected and autonomous vehicles market penetration rates.image

In order to enhance the performance of safety and fuel economy of connected hybrid electric vehicles (CHEVs), a novel distributed hierarchical platoon control scheme of CHEVs is proposed. First, the non‐linear dynamic model of CHEVs platooning is established to accurately depict the multi‐process coupling characteristics of CHEVs. Then, a distributed hierarchical control framework for CHEVs platooning is proposed, which is consisted of a upper model predictive control (MPC) law and a lower energy management control law. The upper MPC control law is built to produce the desired accelerations of every vehicle in the platoon and the lower fuzzy‐based energy management control law is constructed to ensure the engine maintain at the rang of optimum working point and the motor work with the high efficiency of CHEVs platooning. Finally, the results manifest that the effectiveness of proposed platoon control scheme for CHEVs.

The non‐linear dynamic model of connected hybrid electric vehicles (CHEVs) platooning is established to accurately depict the multi‐process coupling characteristics of CHEVs. Then, a distributed hierarchical control framework for CHEVs platooning is proposed, which is consisted of an upper model predictive control (MPC) law and a lower energy management control law. The upper MPC control law is built to produce the desired accelerations of every vehicle in the platoon and the lower fuzzy‐based energy management control law is constructed to ensure the engine maintain at the rang of optimum working point and the motor work with the high efficiency of CHEVs platooning.image

Autonomous vehicles (AVs) present a paradigm shift in addressing conventional parking challenges. Unlike human‐driven vehicles, AVs can strategically park or cruise until summoned by users. Utilizing utility theory, the parking decision‐making processes of AVs users are explored, taking into account constraints related to both cost and time. An agent‐based simulation approach is adopted to construct an AV parking model, reflecting the complex dynamics of the parking decision process in the real world, where each user's choice has a ripple effect on traffic conditions, consequently affecting the feasible options for other users. The simulation experiments indicate that 11.50% of AVs gravitate towards parking lots near their destinations, while over 50% of AVs avoid public parking amenities altogether. This trend towards minimizing individual parking costs prompts AVs to undertake extended empty cruising, resulting in a significant increase of 48.18% in total vehicle mileage. Moreover, the pricing structure across various parking facilities and management dictates the parking preferences of AVs, establishing a nuanced trade‐off between parking expenses and proximity for these vehicles.

This study examines autonomous vehicle (AV) parking using utility theory to design user decision‐making, considering cost and time. An agent‐based model simulates real‐world AV parking dynamics, highlighting individual choices' impact on traffic and others' options.image