The recent integration of educational technologies and emerging learning approaches into education systems has been driven largely by the pandemic. This paper conducts a systematic review and delves into the new wave of research on serious games designed for innovative learning using augmented reality (AR), virtual reality (VR), and mixed reality (MR). The review was referenced to the review protocol, PRISMA 2020. Using the Scopus Database with a time filter from 2007 to 2023 (27 July), we searched 329 articles and shortlisted 273 relevant studies. Notably, European countries contributed the most (62.9%) to this research area. Among the most frequent keywords, VR (90.9%) was commonly used in AR/VR/MR, while e-learning (95.3%) was among the popular innovative learning approaches. Further research studies are needed to employ AR and MR technologies, as well as other innovative learning approaches, to enable performance evaluation and comparison of various educational technologies and learning approaches. We conducted an in-depth analysis of the relevant studies and their basic characteristics. Additionally, we introduced 15 essential and recently published AR/VR/MR standards to ensure better reliability, quality, and safety of architectures, systems, products, services, and processes. To facilitate performance evaluation and analysis, we surveyed 15 recently published benchmark education datasets. This review suggested four future research directions, including multisensory experiences, generative artificial intelligence, personalization and customization, and real-time interaction.

Train operation control in urban railways is challenging due to its high dynamics, complex environment, and level of comfort and safety. To address these challenges, in this article, the authors propose a new deep reinforcement-based train operation (DRTO) method which includes: 1) A deterministic deep reinforcement learning algorithm, 2) a dynamic incentive system, which is used to ensure safe operation in a multitrain environment, and 3) an event-driven method, which is used to improve the DRTO performance based on an event-driven strategy. To evaluate the performance, we thoroughly compare the proposed method with other operation control solutions on both synthetic and real datasets. Our results demonstrate that DRTO is effective in: 1) Decreasing the energy consumption of train operation, 2) increasing passenger comfort, and 3) achieving a good tradeoff between efficiency and safety. In addition, the effectiveness of the event-driven strategy and the dynamic incentive system is demonstrated in the experiments.

Transportation-related consumer electronics technology has advanced rapidly, particularly for automated train operation on high-speed railways. To maximize transport capacity and meet growing demands, this manuscript proposes a new elastic tracking operation control method, that compresses the tracking interval while maintaining safety. The train operation process is formulated as a Monte Carlo process and the Twin Delayed Deep Deterministic policy gradient (TD3) algorithm is used to generate the basic operation strategy. A three-stage control principle and train tracking operation requirements are taken into account, and an elastic parameter-based train state transition rule is proposed. An improved cuckoo algorithm is then used to determine the elastic parameters for faster and more accurate solution convergence. Our results demonstrate that TD3-TOC is effective in i) improving the stability of the train operation process, ii) reducing the tracking interval, and iii) reducing delay in the case of emergency. In addition, the effectiveness of the elastic interval is demonstrated in experiments.

Kernel functions support a broad range of applications that require tasks like density estimation, classification, regression or outlier detection. For these tasks, a common online operation is to compute the weighted aggregation of kernel function values with respect to a set of points. However, scalable aggregation methods are still unknown for typical kernel functions (e.g., Gaussian kernel, polynomial kernel, sigmoid kernel and additive kernels) and weighting schemes. In this paper, we propose a novel and effective bounding technique, by leveraging index structures, to speed up the computation of kernel aggregation. In addition, we extend our technique to additive kernel functions, including X2, intersection, JS and Hellinger kernels, which are widely used in different communities, e.g., computer vision, medical science, Geoscience etc. To handle the additive kernel functions, we further develop the novel and effective bound functions to efficiently evaluate the kernel aggregation. Experimental studies on many real datasets reveal that our proposed solution KARL achieves at least one order of magnitude speedup over the state-of-the-art for different types of kernel functions.

Network kernel density visualization, or NKDV, has been extensively used to visualize spatial data points in various domains, including traffic accident hotspot detection, crime hotspot detection, ...