This paper presents an innovative event-triggered control approach for a class of nonlinear parabolic distributed parameter systems with time-varying delays. The novel event triggering mechanism enables control or measurement signals to be updated only when a predefined trigger condition exceeds a specified threshold. Multiple actuators and sensors, strategically distributed at specific points or partial regions of the spatial domain, are employed to perform pointwise/piecewise control and measurement. Two variations of event-triggered feedback (ETF) controllers are designed to address the collocated and non-collocated observation cases based on the distributions of actuators and sensors in space, respectively. The well-posedness of the open-loop and closed-loop systems is analyzed via the$C_{0}$-semigroup theory, respectively. Furthermore, the non-existence of zeno behavior is guaranteed by demonstrating that the inter-event time intervals are nontrivial. Finally, the proposed method is applied to address the temperature control problem in the catalytic reaction process. Numerical simulation results validate the effectiveness of the proposed ETF control method in practical applications. Note to Practitioners—This work is motivated by the temperature control challenges in catalytic reaction process, with an extended application to the production of hot-rolled steel strips. This paper proposes an innovative event-triggered control method to reduce the demands on communication and computational resources. Extensive comparative experimental results have thoroughly validated the effectiveness of the proposed method in practical applications.

This paper introduces the Adaptive Residual Decision-Control Synthesis (ARDCS) framework, a hierarchical control architecture that synergizes a model-based Linear Quadratic Regulator (LQR) with an adaptive Dynamic Proximal Policy Optimization (DPPO) agent for autonomous bicycle control. ARDCS is designed to master the bicycle’s complex nonlinear dynamics and adapt to environmental uncertainties by leveraging the stability of traditional control with the flexibility of reinforcement learning. A key innovation is a momentum-enhanced dynamic entropy adjustment mechanism within DPPO, which optimizes the exploration-exploitation trade-off for more stable and efficient learning. Comprehensive experiments on balancing and multi-target navigation tasks demonstrate that ARDCS consistently and significantly outperforms both conventional control methods and pure reinforcement learning strategies across varying levels of difficulty. The framework achieves enhanced adaptability and robust stability without relying on auxiliary mechanical stabilizers, offering a potent and generalizable solution for the control of under-actuated systems.

Intention tremor is a rhythmic and involuntary limb swing movement that causes significant inconvenience to the daily life of patients. Repetitive control is suitable for functional-electrical-stimulation-based intention tremor suppression because it can significantly attenuate the periodic signals. However, the performance of the repetitive controller may be weakened in tremor suppression due to uncertainties in the dynamics of the musculoskeletal model and the tremor frequency. In this paper, we propose an advanced methodology for tremor suppression by combining$H_{\infty }$control with high-order repetitive control. The proposed controller can not only guarantee the robust stability of the system subjected to model uncertainty, but also effectively suppress tremors with varying frequency. Comparative experiments on the unimpaired subjects and intention tremor patients were carried out to verify the effectiveness of the proposed method. The experimental statistical analysis results show that the proposed$H_{\infty }$high-order repetitive controller can suppress tremors by up to 84.97%, which is about 11% and 31% higher than the single memory loop repetitive controller and the traditional filter-based controller, respectively. Note to Practitioners—Patients afflicted with wrist intention tremors encounter significant challenges while executing routine activities such as eating, writing, and dressing. Functional electrical stimulation for tremor suppression operates by generating electrical pulses that oppose the tremor motion, thereby inducing muscle contractions and diminishing tremor magnitude. Given the periodic nature of tremor signals acting as system disturbances, repetitive control emerges as an effective method for tremor suppression. However, the conventional repetitive controller cannot significantly improve the tremor suppression performance in practice due to the uncertain property of the musculoskeletal dynamics and the variation of tremor frequency. In this paper, a robust controller combining$H_{\infty }$control with high-order repetitive control is proposed to address above issues. The high-order repetitive controller can effectively suppress the periodic tremor signals with varying frequency, and the$H_{\infty }$controller can provide robust stability and the desired tracking performance by properly choosing of the weighting functions. The comparative experimental results, conducted on our self-built wrist tremor suppression experimental platform, involving unimpaired subjects and patients with intention tremors, validate that the proposed control approach enhances tremor suppression efficacy by 11% and 31%, compared to the traditional repetitive controller and the filter-based controller, respectively. The feasibility and effectiveness of the proposed approach is only initially verified by small scale test, and more clinical verifications will be carried out in tremor patients in the future.

Pavement crack defect detection is an important task in road maintenance. Accurate detection of crack defects has far-reaching significance in maintaining the health condition of roads. Although many excellent crack defect detection algorithms have emerged, the detection effect on the edge details of the crack defects is still not ideal. In this paper, we propose a novel global-local fusion network based on edge enhancement and Transformer for pavement crack defect segmentation. Aiming at the structural characteristics of the pavement crack defects, combined with the edge detection algorithm (Sobel), an edge feature enhancement (EFE) module is presented to realize the accurate extraction of local detail information of the pavement crack defects. Meanwhile, a Transformer-based encoding path is also built to extract rich global information. Faced with the two different types of feature information, an adaptive fusion (AF) module is proposed to realize the efficient fusion of the two types of feature information. Furthermore, an attention-based local feature enhancement (ALFE) module and an edge refinement module (ER) are proposed to further suppress the interference in the local feature maps and refine the edge features of the pavement crack defects. Finally, a multi-scale feature enhancement (MFE) module is presented for multi-scale attention feature representation, by which we can provide high-quality input features for the decoding side. After extensive experimental validation, our proposed model has demonstrated a superior performance over existing mainstream models on multiple pavement crack defect segmentation datasets. The code of the model has been open to:https://github.com/MMYZZU/Crack-Segmentation.

Minimally invasive surgery, which relies on surgical robots and microscopes, demands precise image segmentation to ensure safe and efficient procedures. Nevertheless, achieving accurate segmentation of surgical instruments remains challenging due to the complexity of the surgical environment. To tackle this issue, this paper introduces a novel multiscale dual-encoding segmentation network, termed MSDE-Net, designed to automatically and precisely segment surgical instruments. The proposed MSDE-Net leverages a dual-branch encoder comprising a convolutional neural network (CNN) branch and a transformer branch to effectively extract both local and global features. Moreover, an attention fusion block (AFB) is introduced to ensure effective information complementarity between the dual-branch encoding paths. Additionally, a multilayer context fusion block (MCF) is proposed to enhance the network's capacity to simultaneously extract global and local features. Finally, to extend the scope of global feature information under larger receptive fields, a multi-receptive field fusion (MRF) block is incorporated. Through comprehensive experimental evaluations on two publicly available datasets for surgical instrument segmentation, the proposed MSDE-Net demonstrates superior performance compared to existing methods.

Currently, cracks are the most common defect in pavement diseases. Long-term non-maintenance can lead to crack lengthening and expansion, causing serious traffic accidents, as well as shortening the service life of pavement cracks. Therefore, it is of utmost importance to maintain cracks at an early stage. Due to the effect of some challenging factors, such as various shape information of the cracks, complex textured backgrounds, light shadows, similar texture objects, micro cracks and other factors, accurate crack detection still faces a certain challenges. To solve the above problems, a dual-encoding multi-scale fusion network based on the combination of convolutional neural network (CNN) and transformer network is proposed, named DMF-Net. To obtain stronger feature representations, a dual-encoding path is built to acquire global context features and local detail information simultaneously, where global context features are extracted based on the transformer branch, and the local detail features are extracted based on the CNN branch to detect tiny details of the cracks. Meanwhile, an interactive attention learning (IAL) module is introduced to effectively fuse the global features from the transformer branch and the local detail information from the CNN branch, achieving mutual communication and learning of different feature information. In addition, to enrich the feature representation ability, an attention-based feature enhancement (AFE) module is introduced to acquire more global contexts. Furthermore, faced with the crack detection task with class imbalance issue, a triple attention module (TAM) is built to emphasize the micro cracks. Finally, in the segmentation prediction stage, the deep supervision mechanism is also introduced to accelerate the convergence speed of the model, and serve effective multi-scale feature fusion. Compared with the current mainstream segmentation models, excellent performance has been obtained, which could provide a feasible scheme for the early maintenance of pavement cracks. The source code about proposed DMF-Net is available at https://github.com/Bsl1/DMFNet.git.

Pavement crack detection poses a formidable challenge due to the intricate texture structures of cracks and the complex environmental settings in which they are situated. In recent years, the advancement of deep learning techniques has prompted a surge in the utilization of Convolutional Neural Network (CNN)-based methods for pavement crack detection. While CNNs have exhibited remarkable results in crack detection tasks, they primarily excel at capturing local details with limited receptive fields, which can be insufficient for grasping global contextual information. Given the intricate nature of crack textures, it becomes imperative to leverage both global and local features for accurate detection. To address this issue, a transformer-based network with feature complementary fusion, refer to TFCF-Net, is introduced, which amalgamates Transformer and CNN architectures. Proposed TFCF-Net model prioritizes the Transformer branch for feature encoding, considering its strength in extracting global features, while the CNN branch is set as auxiliary encoding branch, which plays a complementary role for local feature extraction. Proposed TFCF-Net operates by utilizing global features as a foundation and iteratively refining them using local features, thus facilitating precise crack detection. This design enables proposed network to comprehensively capture both global and local information while judiciously fusing these two types of information based on the distinctive characteristics of cracks. To ensure effective fusion of global and local information, an Information Complementary Fusion (ICF) module is presented, which could efficiently merge the outputs of both encoding branches. To further optimize the fused information, a multi-dimensional attention (MA) module is proposed to embed into the, which enhances the model’s ability to capture long-range dependencies by optimizing information from multiple dimensions. Additionally, to improve the quality of input features on the decoding side, a multi-dimensional attention feature representation (MAFR) module is proposed, which expands the receptive field of the deepest semantic information, enabling the extraction of multi-scale feature representations. This paper rigorously evaluate proposed TFCF-Net against state-of-the-art (SOTA) models using three publicly available pavement crack datasets. Experimental results unequivocally demonstrate the superior performance of the proposed TFCF-Net.

Unmanned bicycle motion control has received increasing attention in recent years, and most studies focused on flat terrain. However, practical bicycle control scenarios require consideration of diverse road conditions and terrains. In this paper, we propose a reinforcement learning-based algorithm for unmanned bicycle path tracking, which maps the system states directly to control commands in an end-to-end manner. To overcome the challenge of poor algorithm convergence, we apply the concept of curriculum learning and devise multiple training curricula to progressively train the agent to control the bicycle. Simulation results demonstrate that our algorithm achieves successful unmanned bicycle path tracking control on complex terrains.

Automatic crack detection remains challenging due to factors such as irregular crack shapes and sizes, uneven illumination, complex backgrounds, and image noise. Deep learning has shown promise in computer vision for pixel-wise crack detection, but existing methods still suffer from limitations such as information loss, insufficient feature fusion, and semantic gap issues. To address these challenges, a novel pavement crack segmentation network, called PAF-Net, is proposed, which incorporates progressive and adaptive feature fusion. To mitigate information loss caused by feature downsampling, a progressive context fusion (PCF) block is introduced to capture context information from adjacent scales. To better capture strong features from local regions, a dual attention (DA) block is proposed that leverages both global and local context information, reducing the semantic gap issue. Furthermore, to achieve effective multi-scale feature fusion, a dynamic weight learning (DWL) block is proposed that enables efficient fusion of feature maps from different network layers. Additionally, a multi-scale input unit is incorporated to provide the proposed segmentation network with more contextual information. To evaluate the performance of PAF-Net, we conduct experiments using four common evaluation metrics and compare it with multiple mainstream segmentation models on three public datasets. The proposed PAF-Net demonstrates superior segmentation accuracy for pixel-level crack detection compared to other segmentation models, as evident from qualitative and quantitative experimental results.

Insulator fault diagnosis is a daily but key task for the power transmission system. Long-term exposure to complex natural environment will cause different insulator defects. As a common defects, missing-cap defects of insulators will not only affect the structural strength of power insulators, but also bring a certain effect to the stable power transmission. With the rapid development of machine learning, some machine learning-based defect recognition methods have been proposed for fast and high-precision power inspection. However, the handcrafted features could not effectively express the aerial images against complex inspection environment to affect detection performance of the shallow learning algorithms. And the detection precision of deep learning algorithms will be affected by the unbalanced small-scale defects. Therefore, the fast and high-precision power inspection still faces a certain challenge in the smart grid. To address the above issues, fusion with the deep convolutional neural network (DCNN) and transfer learning, a novel fault diagnosis algorithm of power insulators is proposed to provide a fast and accurate power inspection scheme. To remove complex backgrounds, a fast insulator location algorithm based on the lightweight YOLOV4 model is proposed which is served for the following defect recognition. On the basis, to imitate human vision, a defect recognition algorithm is proposed based on multi-feature fusion. Meanwhile, to ensure the feature expression ability of transfer learning on power insulators, a novel optimization strategy of transfer learning is proposed to improve the recognition precision. Experiments show that the proposed method could acquire a good recognition performance than other recognition models.

Dynamic gesture recognition is a typical human-computer interaction method owing to its great potential in practical applications. Currently, most of research work on gesture recognition has mainly focused on vision-based and surface electromyography (sEMG) methods. Compared to vision-based methods, the sequential sEMG signal can directly depict the muscle activity of different gestures which could lead to higher recognition efficiency. However, the effective feature design and selection of sEMG signal is still complicated since muscle fatigue and small electrode displacement will affect the recognition precision of sEMG signals. In this paper, a novel end-to-end dynamic gesture recognition method is developed. The raw sEMG signals are converted into an image form by using the time-frequency transformation method to obtain more comprehensive information for model training and test. And a recognition model based on Convolutional Neural Network (CNN) model is built for high-precision time-frequency image recognition. Experiments indicate that the proposed method could acquire distinguishing features from the pre-prossed images and the overall recognition accuracy on different gestures can reach up to 98.3%.

The soft fast pneumatic networks actuator (fPNA) featured as large-amplitude motion, and long life span provides a promising solution for varieties of innovative applications, such as the rehabilitation glove, the soft gripper, and the multi-gait robot. However, the infinite freedom in theory impedes its modeling for high-precision control. This paper proposes an analytical model of the fPNA based on the principle of the minimum potential energy. The tight integration of computationally efficiency into the analytical inverse solution of the proposed model enables the model-based control of the fPNA. The validation of the model is experimentally verified by four elaborate fPNAs. Furthermore, an inverse model-based iterative learning controller (ILC) is also constructed for position tracking control of the fPNA.

Rehabilitation is essential to recover the motor function of patients after stroke. In clinic cure, voluntary movements are encouraged to accelerate the recovery. However, for the rehabilitation system based on functional electrical stimulation (FES), voluntary movements are unpredictable and act as input disturbance, which would reduce the control precision. In addition, an accurate model of the human musculoskeletal dynamics is usually not available. In this paper, the upper-limb rehabilitation is described first and simplified to a linear nominal model. To deal with the aperiodic voluntary movements and model uncertainty, disturbance observer (DOB) is introduced as the inner-loop of the rehabilitation control system. The suppression of DOB for voluntary movements and model uncertainty is analysed in frequency domain. The stability of DOB is discussed and a criterion is given. To achieve high precision tracking control, iterative learning control (ILC) is employed. Combined with DOB, a variant gain gradient ILC method is designed based on the nominal model, which could enhance the performance and speed up the convergence. To validate the proposed methods, simulations are performed and compared in the end.

This article investigates a new collision-free routing problem of a multirobot system. The objective is to minimize the cycle time of operation tasks for each robot while avoiding collisions. The focus is set on the operation of the end-effector and its connected joint, and the operation is projected onto a circular area on the plane. We propose to employ a time-space network (TSN) model that maps the robot location constraints into the route planning framework, leading to a mixed integer programming (MIP) problem. A dedicated genetic algorithm is proposed for solving this MIP problem and a new encoding scheme is designed to fit the TSN formulation. Simulation experiments indicate that the proposed model can obtain the collision-free route of the considered multirobot system. Simulation results also show that the proposed genetic algorithm can provide fast and high-quality solutions, compared to two state-of-the-art commercial solvers and a practical approach.

The modeling of electrically stimulated muscles is of great importance for the tremor suppression via functional electrical stimulation (FES) approach. In this paper, with fully consideration of the characteristics of wrist muscles, a four input two output wrist muscle model with Hammerstein structure is proposed, by which the four-channel functional electrical stimulation signals can simultaneously stimulate the wrist muscles of flexor carpi radialis (FCR), extensor carpi radialis (ECR), flexor carpi ulnaris (FCU) and extensor carpi ulnaris (ECU) to realize 2 degrees of freedom (DOF) wrist movements. Then, we use the recursive least squares identification algorithm to identify the parameters of the pre-existing four input and one output system. Simulation results show that the least recursive squares identification algorithm of two-step method is advantageous in convergence and identification accuracy.

Gesture rotation and zooming have significant impact on the gesture recognition system and can greatly reduce the recognition rate. In this paper, we propose a novel recognition method for the multi-scale multi-angle gestures in skin-like noise backgrounds based on HOG-LBP feature extraction. The proposed gesture recognition system consists of pretreatment, feature extraction and classification. First, the single Gaussian model (SGM) and K-means algorithm was used to extract gesture images from a skin-like noise background region. Then, a HOG-LBP feature descriptor is proposed to represent multi-scale multi-angle gesture information. The HOG component provides the gesture edge gradient information and the LBP component provides the texture feature information, which can compensate for the lack of rotation invariance of a single feature and improve the recognition rate of gestures at multiple scales and multiple angles. Finally, the SVM classifier is utilized to realize the gesture classification. Experiment results on the home-made data sets show that the proposed method can achieve 99.01% recognition rate. Experiments on the NUS database and the MUGD database also demonstrate the performance of the proposed method.