Ultra-wideband (UWB) based positioning with fewer anchors has attracted significant research interest in recent years, especially under energy-constrained conditions. However, most existing methods rely on discrete-time representations and smoothness priors to infer a robot's motion states, which often struggle with ensuring multi-sensor data synchronization. In this article, we present a continuous-time UWB-Inertial-Odometer localization system (CT-UIO), utilizing a non-uniform B-spline framework with fewer anchors. Unlike traditional uniform B-spline-based continuous-time methods, we introduce an adaptive knot-span adjustment strategy for non-uniform continuous-time trajectory representation. This is accomplished by adjusting control points dynamically based on movement speed. To enable efficient fusion of inertial measurement unit (IMU) and odometer data, we propose an improved extended kalman filter (EKF) with innovation-based adaptive estimation to provide short-term accurate motion prior. Furthermore, to address the challenge of achieving a fully observable UWB localization system under few-anchor conditions, the virtual anchor (VA) generation method based on multiple hypotheses is proposed. At the backend, we propose an adaptive sliding window strategy for global trajectory estimation. Comprehensive experiments are conducted on three self-collect datasets with different UWB anchor numbers and motion modes. The result shows that the proposed CT-UIO achieves$0.403m$,$0.150m$, and$0.189m$localization accuracy in corridor, exhibition hall, and office environments, yielding$17.2\%$,$26.1\%$, and$15.2\%$improvements compared with competing state-of-the-art UIO systems, respectively. The codebase and datasets of this work will be open-sourced athttps://github.com/JasonSun623/CT-UIO.

Effectively forecasting spatio-temporal data requires capturing intricate Effectively forecasting spatio-temporal data involves capturing complex spatial and temporal dependencies, which is challenging due to long-term dynamics, high-dimensional interactions, and diverse real-world data distributions. Traditional methods often fail to address these complexities, and while recent advances in Graph Neural Networks (GNNs) and attention mechanisms show promise, they face limitations such as high computational costs, dependence on large labeled datasets, and difficulty in balancing local and global dependencies. To overcome these challenges, we propose ESSTPNet, a novel framework that integrates advanced temporal encoding, dynamic spatial representation learning, and multi-scale hierarchical decoding. ESSTPNet introduces an enhanced Deep Echo State Network (DeepESN) for efficient multi-scale temporal feature extraction, a learnable graph displacement operator to dynamically capture both local and global spatial dependencies, and a multi-scale hierarchical decoder that adapts and refines spatio-temporal feature integration. Comprehensive evaluations on benchmark datasets demonstrate ESSTPNet’s state-of-the-art performance. Additionally, case studies on three real-world datasets—water flow, power generation, and electricity consumption—further validate its robustness and practicality, showcasing its ability to handle diverse and noisy data. These results highlight ESSTPNet’s significant contribution to advancing spatio-temporal forecasting.

Recently, the adversarial domain adaptation (ADA) methods have been widely investigated and applied in cross-domain hyperspectral image (HSI) classification. However, most ADA algorithms aim to align the cross-domain distribution without focusing on the class separability of the aligned target features and the information of samples within the domain. To address these issues, a new ADA framework based on calibrated prototype and dynamic instance convolution (CPDIC) is proposed in this paper for cross domain HSI classification. The CPDIC is composed of a generator, a calibrated discriminator and a classifier. The generator includes a static 3D convolutional network (SCN) and a dynamic instance convolutional network (DICN), where the SCN is used to extract coarse-grained features of HSI and the DICN can extract sample-specific fine-grained features using instance convolutions generated from dynamic instance convolution kernel generation (DCKG) module. As for the generator, the static and dynamic interactive feature extraction network extracts robust domain-invariant features with discriminability. The calibrated discriminator aligns the marginal distribution between domains and calibrate the predicted pseudo labels of target domain. For classification, a calibrated prototype loss (CPL) is introduced to align the class distribution across domains. The results of three cross-domain HSI classification tasks show that the proposed CPDIC outperforms existing unsupervised domain adaptation (UDA) algorithms.

In terms of transformer oil temperature prediction, it has been a challenge to effectively predict oil temperature by using time series data. In this paper, we focus on the problem of predicting transformer oil temperature to accurately get the running state of transformer. Firstly, we analyze the correlation between several data and the oil temperature, and then the data with the higher correlation were selected for the experiment. Secondly, we make predictions via the multi-layer perceptron and the LSTM networks. Finally, we can learn that LSTM networks has better performance than the multi-layer perceptron via the comparison experiment. Compared to the existing prediction strategies, the LSTM networks can also get better results and make more effective and efficient prediction.