ABSTRACT Robotic manipulators have become essential in demolition tasks involving hazardous, confined, or structurally unstable environments, where precise and responsive motion control is critical for safety and efficiency. Solving the inverse kinematics (IK) of demolition robot manipulators poses considerable challenges due to the inherent strong nonlinearity and coupling in their kinematic equations, along with potential singularities that can undermine real‐time computational efficiency and system robustness. To address these issues, this study introduces an enhanced Newton–Raphson (NR) approach, specifically optimized for inverse kinematic analysis of 6‐DOF manipulators equipped with a spherical wrist architecture frequently adopted in demolition robotic arms. The proposed method strategically partitions the manipulator into two three‐DOF segments and constructs NR iterative equations by utilizing the reconnecting constraints between these kinematic substructures. After solving a subset of joint variables iteratively, the remaining two joint angles are obtained analytically. The experimental results show that: Compared with the traditional Newton–Raphson method, the improved method has a faster convergence speed, higher accuracy, and better robustness, especially when dealing with singular points at the wrist. This makes the improved method applicable to the inverse kinematics control of dismantling robotic arms in complex environments.

ABSTRACT The NSGA‐II algorithm is widely applied in multiobjective mechanical structure optimization. In this study, the NSGA‐II algorithm was adopted to optimize the refractory lining structure of a ladle. First, a parametric model of the ladle was established using ANSYS Workbench, and the temperature and stress fields under typical operating conditions were calculated to generate the sample data required for training a BPNN prediction model. Second, to address the limitations of conventional BPNN, a genetic algorithm was employed to optimize the initial weights and thresholds. Taking the thicknesses of the working layer, permanent layer, and insulation layer as design variables, a GA‐BPNN single‐objective prediction model was developed, enabling high‐precision predictions of ladle mass, ladle volume, maximum ladle shell temperature, and maximum refractory lining stress. Finally, the NSGA‐II algorithm was utilized to solve the multiobjective optimization problem of the ladle refractory lining. In this optimization, ladle mass and capacity were imposed as constraints, while the maximum shell temperature and maximum lining stress were defined as objectives. In the simulation experiments, thermo‐mechanical coupling analysis was performed in ANSYS Workbench to generate 81 training samples and 10 test samples of temperature and stress data. The GA‐BPNN model optimized by the genetic algorithm achieved accurate predictions of ladle mass, volume, shell temperature rise, and lining stress. The results demonstrated that when the insulation, permanent, and working layers were 9.992, 83.998, and 137 mm thick, respectively, the ladle mass, volume, maximum shell temperature, and maximum lining stress reached 57,972.525 kg, 14.298 m 3 , 145.549°C, and 43.621 MPa. Under this parameter combination, the ladle exhibited optimal comprehensive performance in terms of insulation and service life. This method provides an effective approach to determining the optimal refractory lining structure of ladles, with significant implications for improving thermal performance, extending service life, and enhancing industrial economic efficiency.

ABSTRACT In recent years, intelligent control methods based on surface electromyographic signals (sEMG) have received extensive attention in the fields of bionic prosthetics and human‐computer interaction. Compared with high‐density sampling, sparse sampling sEMG has significant advantages due to its convenient collection, strong non‐intrusiveness and high equipment flexibility. However, it has deficiencies in spatial resolution and muscle activity information acquisition, which can easily lead to a decline in the recognition accuracy of complex gestures. In view of the limitations of sparse sEMG in feature extraction and channel information utilization, this paper proposes a hybrid neural network SERes‐L based on the attention mechanism. This model introduces the squeeze‐and‐excitation (SE) attention module in the residual unit to construct a residual attention structure to enhance the channel modeling of key information. At the same time, it combines the long short‐term memory network (LSTM) to capture the long‐term temporal dependencies in gesture actions, achieving a deep integration of spatial and temporal features. Experiments on the self‐built dataset show that SERes‐L achieves an average recognition accuracy rate of 93.90%. To verify its generalization ability, further tests were conducted on three public datasets, NinaPro DB1, DB4, and DB5. The average classification accuracy rates of the model reached 90.28%, 84.72%, and 91.87% respectively, significantly outperforming many existing mainstream methods. The above results indicate that the proposed SERes‐L architecture can effectively alleviate the challenges of insufficient utilization of channel information and inadequate feature extraction in sparse sEMG, demonstrating excellent generalization performance and practical application potential.

Autonomous driving has gradually become a research hotspot in recent years, but the robustness of loopback detection in complex environments such as dynamic and weak textures need to be improved. For the problem that traditional visual SLAM doesn’t make sufficient use of semantic information in the environment, a dynamic feature rejection method for visual SLAM is proposed based on geometric constraints in autonomous driving. This paper proposes to combine image segmentation and visual SLAM to build a semantic SLAM system and also proposes a data association method that combines semantic and geometric information to improve the traditional loopback detection method by using semantic information to increase the accuracy of loopback detection. The result of experiment on the TUM public dataset shows that the loopback detection accuracy of the improved loopback detection method is higher than that of the bag-of-words method in all four datasets; therefore, the method in this paper improves the accuracy of loopback detection of the SLAM system in general.

sEMG(surface electromyography) signals have been widely used in rehabilitation medicine in the past decades because of their non-invasive, convenient and informative features, especially in human action recognition, which has developed rapidly. However, the research on sparse EMG in multi-view fusion has made less progress compared to high-density EMG signals, and for the problem of how to enrich sparse EMG feature information, a method that can effectively reduce the information loss of feature signals in the channel dimension is needed. In this article, a novel IMSE (Inception-MaxPooling-Squeeze- Excitation) network module is proposed to reduce the loss of feature information during deep learning. Then, multiple feature encoders are constructed to enrich the information of sparse sEMG feature maps based on the multi-core parallel processing method in multi-view fusion networks, while SwT (Swin Transformer) is used as the classification backbone network. By comparing the feature fusion effects of different decision layers of the multi-view fusion network, it is experimentally obtained that the fusion of decision layers can better improve the classification performance of the network. In NinaPro DB1, the proposed network achieves 93.96% average accuracy in gesture action classification with the feature maps obtained in 300ms time window, and the maximum variation range of action recognition rate of individuals is less than 11.2%. The results show that the proposed framework of multi-view learning plays a good role in reducing individuality differences and augmenting channel feature information, which provides a certain reference for non-dense biosignal pattern recognition.

Autonomous driving has gradually become a research hotspot in recent years. Visual Simultaneous Localization and Mapping (SLAM) technology can help unmanned vehicles accurately explore the environment at a lower cost, and the readability of the map can be improved by integrating target detection algorithms. However, the location and 3D shape of the object in the map were not obtained. The method of RGBD-SLAM based on object detection with two-stream YOLOv4-MobileNetv3 convolutional neural network is proposed in this paper. RGBD SLAM algorithm and target detection algorithm are combined to build an algorithm model that can generate the global sparse map and build target dense map quickly. The two-stream network is integrated to obtain 2D information about the target, and further combined with the camera pose after the front-end key frame detection of the SLAM algorithm in this paper, and the dense 3D point cloud of the target and the center point position of the object is obtained. Then, the sparse point cloud of the SLAM system and the dense point cloud of the target can be obtained. The experimental results show that the number of point clouds decreases by about 50% and the time for mapping accounts for about 60% of the global dense mapping time. The method of this paper can efficiently decrease the computational space and improve the speed of semantic mapping, which verifies its feasibility and superiority. It can be used to achieve large-area mapping and the ability to update maps during autonomous driving.

Abstract With the rapid development of sensor technology and artificial intelligence, the video gesture recognition technology under the background of big data makes human‐computer interaction more natural and flexible, bringing richer interactive experience to teaching, on‐board control, electronic games, etc. In order to perform robust recognition under the conditions of illumination change, background clutter, rapid movement, partial occlusion, an algorithm based on multi‐level feature fusion of two‐stream convolutional neural network is proposed, which includes three main steps. Firstly, the Kinect sensor obtains RGB‐D images to establish a gesture database. At the same time, data enhancement is performed on training and test sets. Then, a model of multi‐level feature fusion of two‐stream convolutional neural network is established and trained. Experiments result show that the proposed network model can robustly track and recognize gestures, and compared with the single‐channel model, the average detection accuracy is improved by 1.08%, and mean average precision (mAP) is improved by 3.56%. The average recognition rate of gestures under occlusion and different light intensity was 93.98%. Finally, in the ASL dataset, LaRED dataset, and 1‐miohand dataset, recognition accuracy shows satisfactory performances compared to the other method.

Designing neural networks with bounded Lipschitz constant is a promising way to obtain certifiably robust classifiers against adversarial examples. However, the relevant progress for the important $\ell_\infty$ perturbation setting is rather limited, and a principled understanding of how to design expressive $\ell_\infty$ Lipschitz networks is still lacking. In this paper, we bridge the gap by studying certified $\ell_\infty$ robustness from a novel perspective of representing Boolean functions. We derive two fundamental impossibility results that hold for any standard Lipschitz network: one for robust classification on finite datasets, and the other for Lipschitz function approximation. These results identify that networks built upon norm-bounded affine layers and Lipschitz activations intrinsically lose expressive power even in the two-dimensional case, and shed light on how recently proposed Lipschitz networks (e.g., GroupSort and $\ell_\infty$-distance nets) bypass these impossibilities by leveraging order statistic functions. Finally, based on these insights, we develop a unified Lipschitz network that generalizes prior works, and design a practical version that can be efficiently trained (making certified robust training free). Extensive experiments show that our approach is scalable, efficient, and consistently yields better certified robustness across multiple datasets and perturbation radii than prior Lipschitz networks.

Abstract In sEMG (surface electromyography) pattern recognition, most of the research focuses on the static pattern recognition of different limbs, ignoring the importance of changing load intensity, and joint angle movement information. Traditional static qualitative pattern recognition cannot adjust the motion amplitude and load intensity, so it is of great significance to study the continuous prediction of wrist angle under different load intensities. Based on the correlation between the surface EMG signal and the joint angle signal, the article is based on the neural network to identify and predict the wrist angle under different loads continuously quantitatively. The sEMG signal in this article was collected with the approval and review of the Ethics Committee and the people's informed consent. Since qualitative pattern recognition cannot adjust the wrist movement range and the different load training intensity, the article establishes an angle prediction model based on a genetic algorithm to optimize the extreme learning machine (ELM). In addition, the article analyzes the influence of different loads on the continuous prediction accuracy of the wrist angle, realizes the continuous quantitative angle of the precise wrist prediction. Experimental analysis shows that the wrist joint angle predicted by the ELM optimized based on genetic algorithm is close to the actual angle, and the average error is about 5.96 degrees.

Summary For the problem of surface electromyography (sEMG) gesture recognition, considering the fact that the traditional machine learning model is susceptible to the sEMG feature extraction method, it is difficult to distinguish the subtle differences between similar gestures. The NinaPro DB1 dataset is used as the research object, and the sEMG feature image and the Convolutional Neural Network (CNN) are combined to recognize 52 gesture movements. The CNN model effectively solves the limitations of traditional machine learning in sEMG gesture recognition, and combines 1‐dim convolution kernel to extract deep abstract features to improve the recognition effect. Finally, the simulation experiment shows that compared with the accuracy of the raw‐sEMG images based on the CNN and the sEMG‐feature‐images based on the CNN and sEMG based on the traditional machine learning, the multi‐sEMG‐features image based on the CNN is the highest, which coming up to 82.54%.

Summary With the continuous development of sensor technology, the acquisition cost of RGB‐D images is getting lower and lower, and gesture recognition based on depth images and Red‐Green‐Blue (RGB) images has gradually become a research direction in the field of pattern recognition. However, most of the current processing methods for RGB‐D gesture images are relatively simple, ignoring the relationship and influence between its two modes, and unable to make full use of the correlation factors between different modes. In view of the above problems, this paper optimizes the effect of RGB‐D information processing by considering the independent features and related features of multi‐modal data to construct a weight adaptive algorithm to fuse different features. Simulation experiments show that the method proposed in this paper is better than the traditional RGB‐D gesture image processing method and the gesture recognition rate is higher. Comparing the current more advanced gesture recognition methods, the method proposed in this paper also achieves higher recognition accuracy, which verifies the feasibility and robustness of this method.

Abstract In the production and construction of industry, safety accidents caused by unsafe behaviors of staff often occur. In a complex construction site scene, due to improper operations by personnel, huge safety risks will be buried in the entire production process. The use of deep learning algorithms to replace manual monitoring of site safety regulations is a powerful guarantee for sticking to the line of safety in production. First, the improved YOLO v3 algorithm is used to output the predicted anchor box of the target object, and then pixel feature statistics are performed on the anchor box, and the weight coefficients are respectively multiplied to output the confidence of the standard wearing of the helmet in each predicted anchor box area, according to the empirical threshold determine whether workers meet the standards for wearing helmets. Experimental results show that the helmet wearing detection algorithm based on deep learning in this paper increases the feature map scale, optimizes the prior dimensional algorithm of specific helmet dataset, and improves the loss function, and then combines image processing pixel feature statistics to accurately detect whether the helmet is worn by the standard. The final result is that mAP reaches 93.1% and FPS reaches 55 f/s. In the helmet recognition task, compared to the original YOLO v3 algorithm, mAP is increased by 3.5% and FPS is increased by 3 f/s. It shows that the improved detection algorithm has a better effect on the detection speed and accuracy of the helmet detection task.

Summary Gesture recognition has always been a research hotspot in the field of human‐computer interaction. Its purpose is to realize the natural interaction with the machine by recognizing the semantics expressed by gesture. In the process of gesture recognition, the occlusion of gesture is an inevitable problem. In the process of gesture recognition, some or even all of the gesture features will be lost due to the occlusion of the gesture, resulting in the wrong recognition or even unrecognizability of the gesture. Therefore, it is of great significance to study gesture recognition under occlusion. The single shot multibox detector (SSD) algorithm is analyzed, and the front‐end network is compared. Mobilenets is selected as the front‐end network, and the Mobilenets‐SSD network is improved. In tensorflow environment, based on the improved network model, the self‐occlusion gesture and object occluding gesture are trained in color map, depth map, and color and depth fusion respectively. The recognition models of self‐occlusion gestures and object‐occlusion gestures in color map, depth map, and color and depth fusion are obtained. And compare and analyze the learning rate, loss function, and average accuracy of various models obtained for occlusion gesture recognition.

Abstract Robot grasping technology is a hot spot in robotics research. In relatively fixed industrialized scenarios, using robots to perform grabbing tasks is efficient and lasts a long time. However, in an unstructured environment, the items are diverse, the placement posture is random, and multiple objects are stacked and occluded each other, which makes it difficult for the robot to recognize the target when it is grasped and the grasp method is complicated. Therefore, we propose an accurate, real‐time robot grasp detection method based on convolutional neural networks. A cascaded two‐stage convolutional neural network model with course to fine position and attitude was established. The R‐FCN model was used as the extraction of the candidate frame of the picking position for screening and rough angle estimation, and aiming at the insufficient accuracy of the previous methods in pose detection, an Angle‐Net model is proposed to finely estimate the picking angle. Tests on the Cornell dataset and online robot experiment results show that the method can quickly calculate the optimal gripping point and posture for irregular objects with arbitrary poses and different shapes. The accuracy and real‐time performance of the detection have been improved compared to previous methods.

Summary For meeting the comprehensive requirements of “super insulation,” “lightweight,” and “longevity” of the contemporary ladle, this article designs a new kind of lightweight ladle with heat preservation and longevity performance, and based on steady‐state analysis method and numerical simulation technology, the comparison of temperature distribution between new lightweight and traditional ladle under typical operating modes is made and analyzed. The simulation results of temperature field prove that the performance of heat preservation in new kind of lightweight ladle has been improved obviously from the two aspects of ladle shell temperature and molten steel file rate. At the same time, the simulation results of stress field indicate that the stress of designed lightweight ladle reduced and distributed more evenly, which is conducive to prolonging the work time in‐service of the ladle. Finally, based on the field test, the simulation is proved to be effective, and the designed ladle structure achieves the expected purpose.

Vision-based gesture recognition accords with natural communication habits of human and can carry out long-distance and non-contact interactions. So it has become a hot direction in human-computer interaction research whose recognition effect largely depends on the performance of image preprocessing and recognition algorithms. In this paper, a gesture recognition method using color image and depth image combined is designed. For the influence of the angle on the same gesture, the skeleton algorithm is optimized based on the layer-by-layer stripping concept. The fast refinement algorithm improves the process of repeated scanning, extracts the key node information in the skeleton map of the hand, and establishes the spatial axis of the hand to determine the gesture direction. The gesture recognition experiment was performed based on convolutional neural network. The results showed the recognition accuracy rate was 96.01%, and the robustness and accuracy of the proposed recognition method were verified.