Cooperative perception technology plays a crucial role in autonomous driving systems by improving safety and enabling real-time decision-making. However, existing LiDAR point cloud processing methods face significant challenges in both local geometric feature extraction and global feature fusion. To address these issues, this paper proposes DF-CoopNet, a cooperative perception framework comprising two core modules: Local Geometry Enhancement (LGE) and Sparse Key Feature Attention (SKFA). The LGE module enhances local geometric representations using a deformable k-nearest neighbor graph structure and adaptive fusion mechanism to effectively detect occluded targets. The SKFA module introduces a hierarchical sparse attention mechanism that balances performance and computational complexity through a Top-k strategy. Extensive experiments on the OPV2V and V2V4Real datasets demonstrate that DF-CoopNet significantly outperforms existing methods while maintaining robust detection performance even with substantially reduced point cloud data, validating its effectiveness for real-world cooperative perception applications.

Multiphase and multi-component fluid flows are highly intricate, exhibiting dynamic behaviors that challenge conventional simulation methods, especially across a wide range of operating conditions. Adding fluid-solid interactions further escalates this complexity, requiring simulations that are both physically accurate and computationally more tractable. Existing methods often fall short in aspects of either accuracy, efficiency, or stability, and sometimes may also consume excessive memory, rendering them unsuitable for many practical applications. In this paper, we introduce a novel, more practical solution combining a collision-enhanced lattice Boltzmann method with a finite-volume approach. This hybrid solver is designed to effectively simulate multiphase, multi-component flows via the phase-field modeling with a novel wetting boundary treatment in the finite-volume framework, while significantly reducing memory consumption. It enables simulations of multiple fluid phases under varied conditions, capturing rich interfacial behaviors. Additionally, it incorporates an immersed-boundary method to efficiently handle complex fluid-solid interactions. Validation tests confirm that our method produces physically consistent results and demonstrates accuracy across multiple evaluations. We provide comprehensive simulation results, along with comparisons of memory consumption and efficiency, highlighting the practical utility and superiority of our solver over conventional methods.

Simulating the coupled dynamics between rigid bodies and two-phase fluids, especially those with a large density ratio and a high Reynolds number, is computationally demanding but visually compelling with a broad range of applications. Traditional approaches that directly solve the Navier-Stokes equations often struggle to reproduce these flow phenomena due to stronger numerical diffusion, resulting in lower accuracy. While recent advancements in kinetic lattice Boltzmann methods for two-phase flows have notably enhanced efficiency and accuracy, challenges remain in correctly managing fluid-rigid boundaries, resulting in physically inconsistent results. In this article, we propose a novel kinetic framework for fluid-rigid interaction involving two fluid phases. Our approach leverages the idea of an overset grid, and proposes a novel formulation in the two-phase flow context with multiple improvements to handle complex scenarios and support moving multi-resolution domains with boundary layer control. These new contributions successfully resolve many issues inherent in previous methods and enable physically more consistent simulations of two-phase flow phenomena. We have conducted both quantitative and qualitative evaluations, compared our method to previous techniques, and validated its physical consistency through real-world experiments. Additionally, we demonstrate the versatility of our method across various scenarios.

BackgroundMixed-reality technologies, including virtual reality (VR) and augmented reality (AR) , are considered to be promising potential tools for science teaching and learning processes that could foster positive emotions, motivate autonomous learning, and improve learning outcomes. In this study, a technology-aided biological microscope learning system based on VR/AR is presented. The structure of the microscope is described in a detailed three-dimensional (3D) model, each component being represented with their topological interrelationships and associations among them being established. The interactive behavior of the model was specified, and a standard operating guide was compiled. The motion control of components was simulated based on collision detection. Combined with immersive VR equipment and AR technology, we developed a virtual microscope subsystem and a mobile virtual microscope guidance system. The system consisted of a VR subsystem and an AR subsystem. The focus of the VR subsystem was to simulate operating the microscope and associated interactive behaviors that allowed users to observe and operate the components of the 3D microscope model by means of natural interactions in an immersive scenario. The AR subsystem allowed participants to use a mobile terminal that took a picture of a microscope from a textbook and then displayed the structure and functions of the instrument, as well as the relevant operating guidance. This flexibly allowed students to use the system before or after class without time and space constraints. The system allowed users to switch between the VR and AR subsystems. The system is useful for helping learners (especially K-12 students) to recognize a microscope's structure and grasp the required operational skills by simulating operations using an interactive process. In the future, such technology-assisted education would be a successful learning platform in an open learning space.