The accurate trajectory prediction of surrounding vehicles is crucial for the sustainability and safety of connected and autonomous vehicles under mixed traffic streams in the real world. The task of trajectory prediction is challenging because there are all kinds of factors affecting the motions of vehicles, such as the individual movements, the ambient driving environment especially road conditions, and the interactions with neighboring vehicles. To resolve the above issues, this work proposes a novel Heterogeneous Context-Aware Graph Convolutional Networks following the Encoder-Decoder architecture, which simultaneously extracts the hidden contexts from individual historical trajectories, varying driving scene, and inter-vehicle interactional behaviors. Specifically, the historical vehicle trajectories are fed into Temporal Convolutional Network to capture the individual context. Besides, a 2-Dimensional Convolutional Network with temporal attention is designed for transforming the scene image stream into compressing scene context. Then a Spatio-Temporal Dynamic Graph Convolutional Networks is devised to model the evolving interactional patterns, which incorporates the acquired individual and scene contexts as the representation of the node. Finally, the aforementioned three contexts are combined and fed into the decoder to produce future trajectories. The proposed model is validated on two real-world datasets which contain various driving scenarios. Results demonstrated that the proposed model outperforms state-of-the-art methods in prediction accuracy and achieves immense stability towards different vehicle states.

Traffic flow prediction plays an integral role in intelligent transport systems, helping to manage and control urban traffic and improving the operational efficiency of road networks. Although the current mainstream traffic flow prediction models have achieved good accuracy, they cannot effectively utilize the unique characteristics of the traffic network where the importance of a node in the traffic network is positively correlated with the traffic flow through the node. Actually, the historical traffic properties of nodes will have a great influence on the future. With this background, in this paper, we propose a node connection strength index by network representation learning to utilize the historical traffic attributes of nodes. Then, we design a graph convolution network based on the node connection strength matrix to predict the traffic flow of the node. A novel Dynamics Extractor is designed to learn the various characteristics of the traffic flow. Experimental results demonstrate that the proposed scheme has a better performance by comparison with baseline methods.

With the ability to acquire and process large-scale traffic big data, a cooperative intelligent transport system can be realized. Identifying important nodes in a traffic network contributes to better traffic control, which plays a more critical role in improving the traffic efficiency of a cooperative intelligent transport system. However, existing traffic node importance evaluation methods rely on manually designed metrics such as betweenness, degree, which may lead to biased results. Meanwhile, the traditional method of iteratively deleting nodes is unsuitable for a large-scale traffic network. In this paper, we propose a novel traffic node importance evaluation method based on clustering in represented transportation network. Specifically, the proposed method first construct a length-weighted network based on the geographic road network. Then, it learns the low-dimensional embeddings of nodes employing network representation learning. Finally, it clusters each nodes with a machine learning method and identifies the critical nodes through vehicle flow of different nodes. Experimental results on the real-word dataset show that our proposed method has excellent performance compared to baseline methods.

Networked systems widely exist in the modern society, and these systems are always operated in the presence of attacks and errors. The robustness of a network indicates its tolerance against potential damages, which is crucial for the network’s normal functionalities. Damage models including malicious attacks and cascading failures have been considered in previous studies. And the mitigation on these two destructive damages has attracted increasing attention, both single-objective and multiobjective optimization techniques are proposed to enhance the robustness of networks. The main weakness of these methods is that the obtained results can only handle one type of damages, either malicious attacks or cascading failures. In this way, multiple independent realizations are evitable to acquire robust candidates against different damages, which omits the inherent complementarities between the optimization of different objectives. Also, the design of networks with comprehensive robustness (CR) against both the two destructive damages is still pendent. For improving the efficiency in optimizing networks considering multiple damage models, in this article, evolutionary multitasking optimization is introduced into the optimization of network robustness. A multifactorial evolutionary algorithm (MFEA) for enhancing the CR of networks, called MFEA textsubscript CR, has been proposed. The algorithm is equipped with several dedicated genetic operators to find networks with robust structures against malicious attacks and cascading failures; meanwhile, the synergy between the optimization of the two objectives is exploited to improve the computational efficiency. Empirical studies demonstrate the competitive performance of ${\mathrm{ MFEA}}_{\mathrm{ CR}}$ over existing methods; the efficiency enhancement is remarkable via the simultaneous optimization of multiple objectives.

State-of-charge (SOC) estimation is essential for the safe and effective utilization of lithium-ion batteries. As the SOC cannot be directly measured by sensors, an accurate battery model and a corresponding estimation method is needed. Compared with electrochemical models, the equivalent circuit models are widely used due to their simplicity and feasibility. However, such integer order-based models are not sufficient to simulate the key behavior of the battery, and therefore, their accuracy is limited. In this paper, a new model with fractional order elements is presented. The fractional order values are adaptively updated over time. For battery SOC estimation, an unscented fractional Kalman filter (UFKF) is employed based on the proposed model. Furthermore, a dual estimation scheme is designed to estimate the variable orders simultaneously. The accuracy of the proposed model is verified under different dynamic profiles, and the experimental results indicate the stability and accuracy of the estimation method.