Accurate estimation of the state of health (SOH) for Lithium-ion batteries is crucial for ensuring safety and extending life. However, existing data-driven methods still face two challenges in practice, the lack of data and physical mechanism explanation. To address these issues, this study propose a novel hybrid frameork that integrates conditional generative adversarial network with physics-informed neural network (cGAN-PINN). The aging feature-guided cGAN generates synthetic voltage and temperature data using the SOH and normalized cycle number as conditional inputs, enhancing data diversity and preserving physical consistency. Subsequently, both synthetic and real data are fed into the PINN, which incorporates the Verhulst electrochemical model to embed prior physical knowledge of capacity fade dynamics, thereby enabling simultaneous SOH prediction and parameter estimation. Results demonstrate that the synthetic data generated by cGAN effectively alleviates overfitting and enhances generalization capability, reducing root mean square percentage error (RMSPE) by 42.0% on the test set. The framework achieves RMSPE and mean absolute error (MAE) of 0.269 and 1.158% on the test set, demonstrating robust performance for battery health management under data scarcity.

Against the backdrop of the rapid development of the energy internet, the role of energy storage systems in grid stability, energy balance, and renewable energy integration has become increasingly important. Among these systems, estimating the state of health (SOH) of battery storage systems, particularly lithium batteries, is crucial for ensuring system reliability and safety. While data-driven methods have poor interpretability and physics-based models are computationally expensive, physics-informed neural networks (PINNs) offer a compromise but struggle with high-dimensional inputs and dynamic variable coupling. This paper proposed a novel Kolmogorov–Arnold networks with physics-informed neural network (KAN-PINN) framework for lithium-ion battery SOH estimation. By leveraging KANs’ superior high-dimensional approximation capabilities and embedding the Verhulst model as a physical constraint, the framework enhances nonlinear representation while ensuring predictions adhere to degradation physics. Experimental results on a public dataset demonstrate the model’s superiority, achieving an RMSPE of 0.300 and MAE of 1.342%, along with strong interpretability and robustness across battery chemistries and operating conditions.

Target tracking in typical radar applications faces critical challenges in complex environments, including nonlinear dynamics, non-Gaussian noise, and sensor outliers. Current robustness-enhanced approaches remain constrained by empirical kernel tuning and computational trade-offs, failing to achieve balanced noise suppression and real-time efficiency. To address these limitations, this paper proposes the variational Bayesian-based maximum correntropy criterion cubature Kalman filter (VBMCC-CKF), which integrates variational Bayesian inference with CKF to establish a fully adaptive robust filtering framework for nonlinear systems. The core innovation lies in constructing a joint estimation framework of state and kernel size, where the kernel size is modeled as an inverse-gamma distributed random variable. Leveraging the conjugate properties of Gaussian-inverse gamma distributions, the method synchronously optimizes the state posterior distribution and kernel size parameters via variational Bayesian inference, eliminating reliance on manual empirical adjustments inherent to conventional correntropy-based filters. Simulation confirms the robust performance of the VBMCC-CKF framework in both single and multi-target tracking under non-Gaussian noise conditions. For the single-target case, it achieves a reduction in trajectory average root mean square error (Avg-RMSE) by at least 14.33% compared to benchmark methods while maintaining real-time computational efficiency. Integrated with multi-Bernoulli filtering, the method achieves a 40% lower optimal subpattern assignment (OSPA) distance even under 10-fold covariance mutations, accompanied by superior hit rates (HRs) and minimal trajectory position RMSEs in cluttered environments. These results substantiate its precision and adaptability for dynamic tracking scenarios.

The hyperspectral image classification community has long sought methods to handle scarce samples and unpredictable variations. Recent research aims to develop foundational models that overcome traditional limitations, such as inadequate representational capability, poor integration of prior knowledge, and incomplete feature extraction. However, the high cost of data collection and annotation, along with the billions of parameters involved, hinder real-world deployment. In this paper, we propose an insightful method for domain generalization using overall semantics, positional relationships, and morphological characteristics as domain constant knowledge carriers. This approach enables accurate cross-scene classification while significantly reducing deployment costs. Specifically, we design a lightweight vision-language hybrid domain generalization network (VLDG-Net) for few-shot cross-scene hyperspectral image classification. First, features from the source domain (SD) data are extracted through jointly optimized image and text encoders, with the resulting model transferable to the unseen target domain (TD) after pre-training. Second, we propose a residual hybrid backbone (RHB) as the image encoder, using residual 3D blocks (R3DB) to extract spatial-spectral features and residual 2D selective scanning blocks (RSS2DB) to model long-range dependencies. Additionally, we design an adapter for efficient prior knowledge embedding in the text encoder. Lastly, a two-stage contrastive loss (TSCL), combining vision-language alignment loss and instance discrimination loss, is proposed to create a more separable feature space. Extensive experiments on three public datasets show that our method outperforms several advanced approaches in few-shot cross-scene HSI classification, while also being efficient in training and testing.