Automatic Optical Inspection (AOI), as a core equipment in quality inspection process of printed circuit board (PCB) assembly lines, directly impacts overall production capacity through its inspection efficiency. However, existing research on AOI route scheduling problem exhibits limitations such as neglecting image acquisition center adjustment and lack of efficiency for large-scale PCBs. A hierarchical heuristic algorithm based on neighborhood search is proposed to address large-scale AOI route scheduling. The problem is decomposed into component clustering, path sequencing, and image acquisition center adjustment. The method features adaptive neighborhood construction through search area adjustment, enabling component clustering via neighborhood operations. Cluster centers are then sequenced using the Lin-Kernighan algorithm. A greedy heuristic algorithm for image acquisition center adjustment is further developed to optimize path length. Experimental results demonstrate that this algorithmic framework outperforms state-of-the-art methods, particularly showing significant improvements in solving large-scale problems.

This paper proposes an adaptive output feedback trajectory tracking control scheme for permanent magnet linear synchronous motors (PMLSMs). An integral sliding mode observer (ISMO) is developed to simultaneously estimate lumped disturbances and velocity state without requiring disturbance differentiability and velocity measurement. A command filter backstepping control (CFBC) strategy replaces derivative operations in conventional backstepping designs, with a compensation mechanism introduced for filter errors, simplifying the controller design and improving tracking performance. In addition, an adaptive law is designed to update model parameters online, enhancing robustness and adaptability. Experiments are conducted on an iron-core PMLSM platform and results show that the proposed method achieves superior tracking performance.

Significant differences in spatial structure characteristics make the coarse-grained few-shot scenarios easier to handle. However, fine-grained scenes remain often more challenging, and current related research falls far short of human-level recognition capabilities. To address this tough challenge, we get inspiration by the structures and functions of the human visual system, and then propose a human-centric visual cognition recognition framework, named VCRNet, including a shallow cognition network VCRNet-4 and the deep cognition neural network VCRNet-12. Specifically, in this framework, we cleverly design a visual perception module to simulate powerful visual pathway feature encoding of human visual system, an attentional regional sensing module and another pixel-level sensing module to simulate the structural and shape-detailed perception capabilities of the human occipital lobe, and a cognitive recognition module to emulate the integrative cognitive abilities of human frontal lobe. Ulteriorly, we have evaluated the performances of our proposed methods on three publicly available fine-grained benchmarks: CUB-200-2011, Aircraft-Fewshot, and Stanford-Cars. The compelling comparative experiments and thoroughly justified validation study demonstrate the structural excellence and superior performance of our VCRNet framework. To facilitate subsequent research, our codes have been available.

Meta-metric learning has demonstrated strong performance in coarse-grained few-shot situations. However, despite their simplicity and availability, these metametrics are limited in effectively handling fine-grained few-shot scenarios. Fine-Grained Few-Shot Classification (FGFSC) presents significant challenges to the network's ability to extract subtle features. Equipped with the symmetrical binocular perception system and complex neural networks in the brain, humans inherently possess exceptional and resilient meta-learning abilities, facilitating superior management of fine-grained few-shot scenarios. In this paper, inspired by the human binocular visual system, we pioneer the first human-like meta-metric paradigm: Binocular Singular Hellinger Metametric (BinoHeM). Functionally, BinoHeM incorporates advanced symmetric binocular feature encoding and recognition mechanisms. Structurally, it integrates two binocular sensing feature encoders, a singular Hellinger metametric, and two collaborative identification mechanisms. Building on this foundation, we introduce two innovative metametric variants: BinoHeM-KDL and BinoHeM-MTL. These are grounded in two advanced training mechanisms: knowledge distillation learning (KDL) and meta-transfer learning (MTL), respectively. Furthermore, we showcase the high accuracy and robust generalization capabilities of our approaches on four representative FGFSC benchmarks. Extensive comparative and ablation experiments have validated the efficiency and superiority of our paradigm over other state-of-the-art algorithms. Our code is publicly available at: https://github.com/ChaofeiQI/BinoHeM.

This paper proposed a neural networks-based adaptive control scheme for linear motors considering cogging force compensation. The cogging force is modeled and compensated for improving the tracking performance. An indirect parameter adaptive strategy is proposed to address the problem of parametric uncertainties. Compared with the direct adaptive strategy, this strategy can promote the parameter estimations to converge to the true values. In addition, radial basis neural networks are designed to estimate remaining system uncertainties, including unmodeled dynamics, model errors, and external disturbances. Comparative experiments are conducted on an iron-core permanent magnet linear synchronous motor platform. The experimental results show that the proposed control scheme can achieve excellent control performance.

Pneumatic actuators gain much popularity in many industries where there is great demand for a safety working environment and dynamic performance of a system. But the nonlinear characteristics such as friction and air compressibility add to difficulty of controlling so that constrain its wider application. In this paper, in order to overcome the disadvantage like the inaccuracy of parameters, uncertainty of the model and disturbance, a sliding mode observer with RBF neural network is proposed. The RBF neural network is designed to appropriate the nonlinear parts of the model, and the robustness of sliding mode control can guarantee the stability of control system under perturbation and model uncertainty. The stability of this algorithm is proved by Lyapunov theory. Finally, simulations done with Simulink is designed to examine the effectiveness of our algorithm. The result shows this algorithm has good performance.

A sliding mode control method based on nonlinear disturbance observer and nominal model is proposed to track the trajectory of manipulator with uncertain interference. A dynamic model of the manipulator is established, a sliding mode control law is designed, and the stability of the system is verified by the Lyapunov stability theory. A nonlinear disturbance observer is introduced to improve the performance of the control system. The results show that the control method reduces the unmodelled dynamic errors and the influence of uncertain external interference. Furthermore, simulation results based on Matlab prove that compared with the traditional PD position control method, proposed method has higher accuracy and better robustness.