Traffic forecasting is of great importance to transportation management and public safety, very challenging due the complicated spatial-temporal dependency essential uncertainty brought about by road network traffic conditions. Latest studies mainly focus on modeling spatial utilizing graph convolutional networks (GCNs) throughout a fixed weighted graph. However, edges, i.e., correlations between pair-wise nodes, are much more interact with each other. In this paper, we propose Multi-Range...

Among kinds of flexible tactile sensors, piezoelectric sensor has the advantage fast response for dynamic force detection. However, it suffers from low sensitivity at high-frequency stimuli. Here, inspired by finger structure-rigid skeleton embedded in muscle, we report a using rigid-soft hybrid force-transmission-layer combination with soft bottom substrate, which not only greatly enhances transmission, but also triggers significantly magnified effect d31 working mode sensory layer, instead...

Abstract Liquid metal is an ideal conductive material for soft electronics because of its high conductivity and fluidity at room temperature. However, the large surface tension mass density liquid make forming three-dimensional (3D) dangling structures a challenging task. Reported here suspension printing strategy direct deposition galinstan-based into 3D with shape fidelity spatial resolution (~150 μm). Acrylamide/nanoclay served as yield-stress fluid support bath, selected hydrogen...

Cross-Domain Few-Shot Learning (CD-FSL) is a recently emerging task that tackles few-shot learning across different domains. It aims at transferring prior knowledge learned on the source dataset to novel target datasets. The CD-FSL especially challenged by huge domain gap between Critically, such actually comes from changes of visual styles, and wave-SAN [10] empirically shows spanning style distribution data helps alleviate this issue. However, simply swaps styles two images. Such vanilla...

Radiation induced soft errors in combinational logic is expected to become as important directly on state elements. Consequently, it has develop techniques quickly and accurately predict error rates (SER) circuits. In this paper, we propose a new approach, which can be applied designs that use cell libraries characterized for analysis utilizes analytical equations model the propagation of voltage pulse input element. The average SER estimates using our approach compared obtained circuit...

Effective cooling solutions become increasingly crucial as micro-electrical devices advance in complexity and miniaturization. Microchannel heat sinks (MCHSs) are one of the solutions, researchers continuously exploring innovative designs optimization techniques to ensure that these meet evolving demands modern electronics. In current research, a novel disk-shaped MCHS with strategically positioned baffles was selected be investigated. Total thermal resistance (Rt) pressure drop (ΔP) were...

In the era of big data, using data to realize online estimation battery SOH has become possible. Traditional solutions based on theoretical models cannot take into account driving behavior and complicated environmental factors. this paper, an approximate degradation model real operating temperature electric vehicles (EVs) collected with a platform is proposed. Firstly, health indicators are extracted from historical equivalent capacity at 25 °C obtained capacity-temperature empirical formula...

Liking it or not, ready we are likely to enter a new phase of human history in which artificial intelligence (AI) will dominate economic production and social life—the AI revolution. Before the actual arrival revolution, is time for us speculate on how impact world. In this article, focus generative LLM-based AI, discussing societal factors that contribute its technological development potential roles enhancing both between-country within-country inequality. There good indications US China...

Abstract In response to the problems of insufficient accuracy and poor real-time performance in traditional paver positioning systems, this paper proposes a BDS(Beidou Navigation Satellite System) technology based on FPGA(Field Programmable Gate Array). Using FPGA as central processing system, analysis BDS messages is completed. Based Gaussian Luger projection plane four parameter conversion method, coordinate implemented convert latitude longitude information into CGCS2000(China Geodetic...

Soft errors that change configuration bits of an SRAM based FPGA modify the functionality design. The proliferation devices in various critical applications makes it important to increase their immunity soft errors. In this work, we propose use asymmetric (ASRAM) structure is optimized for error and leakage when storing a preferred value. key our approach observation bitstream composed 87% zeros across different designs. Consequently, ASRAM cell zero (ASRAM-0) reduces failure time by 25% as...

Abstract Better control of molecule‐electrode coupling ( Γ ) to minimize leakage current is an effective method optimize the functionality molecular diodes. Herein we embedded 5 isomers phenypyridyl derivatives, each with N atom placed at a different position, in two electrodes fine‐tune between self‐assembled monolayers (SAMs) and top electrode EGaIn (eutectic Ga−In terminating Ga 2 O 3 ). Combined electrical tunnelling results, characterizations electronic structures, single‐level model...

In the realm of real-time spaceborne synthetic aperture radar (SAR) imaging, accurate and swift phase factors calculation (PFC) holds significant importance. This paper introduces an innovative ARM-FPGA hybrid acceleration technique designed to expedite factor computation process SAR imaging. By combing strengths Advanced RISC Machines (ARM) Field-Programmable Gate Array (FPGA), our approach achieves optimal performance. The proposed methodology strategically allocates initial intermediate...

As technology scales, transient faults due to single event upsets have emerged as a key challenge for reliable embedded system design. This work proposes design methodology that incorporates reliability into hardware-software co-design paradigm systems. We introduce an allocation and scheduling algorithm efficiently handles conditional execution in multi-rate systems, selectively duplicates critical tasks detect soft errors, such the of is increased. The increased achieved by utilizing...

Previous few-shot learning (FSL) works mostly are limited to natural images of general concepts and categories. These assume very high visual similarity between the source target classes. In contrast, recently proposed cross-domain (CD-FSL) aims at transferring knowledge from nature many labeled examples novel domain-specific categories only a few examples. The key challenge CD-FSL lies in huge data shift domains, which is typically form totally different styles. This makes it nontrivial...

Due to technology scaling, smaller devices and lower operating voltages, next generation circuits are highly susceptible soft errors. Another important problem confronting silicon scaling is static power consumption. In this paper, we analyze the effect of increasing threshold voltage (widely used for reducing consumption) on error rate (SER). We find that improves SER transmission gate based flip-flops, but can adversely affect robustness combinational logic due higher voltages attenuation...

The representation of physical characteristics is the most essential feature mathematical models used for detection defects in automatic inspection systems. However, and formation defect image are not considered enough traditional algorithms. This paper presents a model inspection, denoted as localized (LDIM), different because it modeling features manual using local merit function to quantify cost that pixel defective. comprises two components: color deviation fluctuation. Parameters...

A deep learning with knowledge distillation scheme for lateral lane-level vehicle identification based on ultra-weak fiber Bragg grating (UWFBG) arrays is proposed. Firstly, the UWFBG are laid underground in each expressway lane to obtain vibration signals of vehicles. Then, three types (the signal a single vehicle, accompanying signal, and laterally adjacent vehicles) separately extracted by density-based spatial clustering applications noise (DBSCAN) produce sample library. Finally,...

Abstract A metasurface is a kind of ultrathin artificial composite composed subwavelength elements with unique abilities in manipulating electromagnetic waves. However, the static nature its conventional metallic/dielectric constituent material has limited fixed functionality narrow frequency ranges. The two‐dimensional carbon sheet, i.e., graphene, promising platform for effectively tuning as well operation band metasurface. Here, authors propose and demonstrate graphene hybrid dual‐band...