Joint Module Research Articles

Robust visual question answering via polarity enhancement and contrast

The Visual Question Answering (VQA) task is an important research direction in the field of artificial intelligence, which requires a model that can simultaneously understand visual images and natural language questions, and answer questions related to images. Recent studies have shown that many Visual Question Answering models rely on statistically regular correlations between questions and answers, which in turn weakens the correlation between visual content and textual information. In this work, we propose an unbiased Visual Question Answering method to solve language priors from the perspective of strengthening the contrast between the correct answer and the positive and negative predictions. We design a new model consisting of two modules with different roles. We input the image and the question corresponding to it into the Answer Visual Attention Modules to generate positive prediction output, and then use a Dual Channels Joint Module to generate negative prediction output with great linguistic prior knowledge. Finally, we input the positive and negative predictions together with the correct answer to our newly designed loss function for training. Our method achieves high performance (61.24%) on the VQA-CP v2 dataset. In addition, most existing debiasing methods improve performance on VQA-CP v2 dataset at the cost of reducing performance on VQA v2 dataset, while our method not only does not reduce the accuracy on VQA v2 dataset. Instead, it improves performance on both datasets mentioned above.

The Precision Improvement of Robot Integrated Joint Module Based on a New ADRC Algorithm

This article analyzes the factors causing the precision error of the robot joint module, such as gear meshing disturbance, output elastic deformation, and load mutation. An improved active disturbance rejection control (ADRC) algorithm is proposed to overcome the uncertainty of nonlinear factors and variables of this research topic. The output precision loss of the joint module is introduced as a new input variable of ADRC, combined with the input variable of torque current of the joint module. The extended state observer is redesigned, and the online estimation of disturbance is realized. According to the disturbance estimation results, the current loop algorithm of permanent magnet synchronous motors is improved to compensate the torque disturbance of the robot joint module. The experimental results show that the improved ADRC algorithm can obviously suppress the disturbance of the joint module, weaken the meshing error and torque output deformation of the harmonic reducer gear, and improve the control accuracy of the joint module.

A Multi-Scale Mask Convolution-Based Blind-Spot Network for Hyperspectral Anomaly Detection

Existing methods of hyperspectral anomaly detection still face several challenges: (1) Due to the limitations of self-supervision, avoiding the identity mapping of anomalies remains difficult; (2) the ineffective interaction between spatial and spectral features leads to the insufficient utilization of spatial information; and (3) current methods are not adaptable to the detection of multi-scale anomaly targets. To address the aforementioned challenges, we proposed a blind-spot network based on multi-scale blind-spot convolution for HAD. The multi-scale mask convolution module is employed to adapt to diverse scales of anomaly targets, while the dynamic fusion module is introduced to integrate the advantages of mask convolutions at different scales. The proposed approach includes a spatial–spectral joint module and a background feature attention mechanism to enhance the interaction between spatial–spectral features, with a specific emphasis on highlighting the significance of background features within the network. Furthermore, we propose a preprocessing technique that combines pixel shuffle down-sampling (PD) with spatial spectral joint screening. This approach addresses anomalous identity mapping and enables finite-scale mask convolution for better detection of targets at various scales. The proposed approach was assessed on four real hyperspectral datasets comprising anomaly targets of different scales. The experimental results demonstrate the effectiveness and superior performance of the proposed methodology compared with nine state-of-the-art methods.

Design and Analysis of a Symmetric Joint Module for a Modular Wire-Actuated Robotic Arm with Symmetric Variable-Stiffness Units

Collaborative robots are used in scenarios requiring interaction with humans. In order to improve the safety and adaptability of collaborative robots during human–robot interaction, this paper proposes a modular wire-actuated robotic arm with symmetric variable-stiffness units. The variable-stiffness unit is employed to extend the stiffness-adjustment range of the robotic arm. The variable-stiffness unit is designed based on flexure, featuring a compact and simple structure. The stiffness–force relationship of the variable-stiffness unit can be fitted by a quadratic function with an R-squared value of 0.99981, indicating weak nonlinearity. Based on the kinematics and stiffness analysis of the symmetric joint module of the robotic arm, the orientation of the joint module can be adjusted by regulating the length of the wires and the stiffness of the joint module can be adjusted by regulating the tension of the wires. Because of the actuation redundancy, the orientation and stiffness of the joint module can be adjusted synchronously. Furthermore, a direct method is proposed for the stiffness-oriented wire-tension-distribution problem of the 1-DOF joint module. A simulation is carried out to verify the proposed method. The simulation result shows that the deviation between the calculated stiffness and the desired stiffness was less than 0.005%.

A Novel Cable-Driven Joint Module for Space Manipulators: Design, Modeling, and Characterization

A Novel Cable-Driven Joint Module for Space Manipulators: Design, Modeling, and Characterization

Shape Sensing and Kinematic Control of a Cable-Driven Continuum Robot Based on Stretchable Capacitive Sensors.

A Cable-Driven Continuum Robot (CDCR) that consists of a set of identical Cable-Driven Continuum Joint Modules (CDCJMs) is proposed in this paper. The CDCJMs merely produce 2-DOF bending motions by controlling driving cable lengths. In each CDCJM, a pattern-based flexible backbone is employed as a passive compliant joint to generate 2-DOF bending deflections, which can be characterized by two joint variables, i.e., the bending direction angle and the bending angle. However, as the bending deflection is determined by not only the lengths of the driving cables but also the gravity and payload, it will be inaccurate to compute the two joint variables with its kinematic model. In this work, two stretchable capacitive sensors are employed to measure the bending shape of the flexible backbone so as to accurately determine the two joint variables. Compared with FBG-based and vision-based shape-sensing methods, the proposed method with stretchable capacitive sensors has the advantages of high sensitivity to the bending deflection of the backbone, ease of implementation, and cost effectiveness. The initial location of a stretchable sensor is generally defined by its two endpoint positions on the surface of the backbone without bending. A generic shape-sensing model, i.e., the relationship between the sensor reading and the two joint variables, is formulated based on the 2-DOF bending deflection of the backbone. To further improve the accuracy of the shape-sensing model, a calibration method is proposed to compensate for the location errors of stretchable sensors. Based on the calibrated shape-sensing model, a sliding-mode-based closed-loop control method is implemented for the CDCR. In order to verify the effectiveness of the proposed closed-loop control method, the trajectory tracking accuracy experiments of the CDCR are conducted based on a circle trajectory, in which the radius of the circle is 55mm. The average tracking errors of the CDCR measured by the Qualisys motion capture system under the open-loop and the closed-loop control are 49.23 and 8.40mm, respectively, which is reduced by 82.94%.

[formula omitted]GCN: A hybrid hypergraph convolution network for skeleton-based action recognition

Recent GCN-based works have achieved remarkable results for skeleton-based human action recognition. Nevertheless, while existing approaches extensively investigate pairwise joint relationships, only a limited number of models explore the intricate, high-order relationships among multiple joints. In this paper, we propose a novel hypergraph convolution method that represents the relationships among multiple joints with hyperedges, and dynamically refines the height-order relationship between hyperedges in the spatial, temporal, and channel dimensions. Specifically, our method initiates with a temporal-channel refinement hypergraph convolutional network, dynamically learning temporal and channel topologies in a data-dependent manner, which facilitates the capture of non-physical structural information inherent in the human body. Furthermore, to model various inter-joint relationships across spatio-temporal dimensions, we propose a spatio-temporal hypergraph joint module, which aims to encapsulate the dynamic spatial–temporal characteristics of the human body. Through the integration of these modules, our proposed model achieves state-of-the-art performance on RGB+D 60 and NTU RGB+D 120 datasets.

A compound twisted and coiled actuators with payload-insensitive untwisting characteristics

Compound Twisted and Coiled Actuators (CTCAs) are promising candidates for artificial muscles due to their outstanding competences in producing large output forces and long strokes. However, as their thermal actuation performance depends on not only the driving temperature, but also the spandex deformation caused by the payload, it is rather difficult to obtain accurate actuation models for the conventional CTCAs and thus seriously restricts their control performance and application development. To tackle such difficulties, a CTCA with a novel strategic fabrication method is proposed in this work. Motivated by the kinetostatic performance of a stiffening spring with a high preload, the fiber draw ratio of the CTCA is significantly increased through pre-stretching the fiber during the fabrication process, which makes the thermal actuation performance of the CTCA insensitive to the external payload. The effectiveness of the proposed payload-insensitive characteristics of the CTCA is validated through experimental results, which shows that the CTCA has almost the same untwisting strain–temperature curves under different payloads. Consequently, a simplified yet accurate actuation model is obtained such that the displacement of the CTCA is mainly determined by the driving temperature within a meaningful range of payloads. Experiments are conducted to evaluate the accuracy of the actuation model and the model error is less than 7.6%. Such a simplified actuation model is implemented for the pose control of a 2-DOF CTCA-driven continuum robot joint module, in which the average steady-state error of the bending angle is less than 1.0∘.

Design and architecture of a slender and flexible underwater robot

This paper presents the design and analysis of a biomimetic underwater snake-like robot, addressing the main limitations of current underwater robotic systems in terms of maneuverability and adaptability in complex environments. The innovative design incorporates flexible joint modules that significantly enhance the robot’s ability to navigate through narrow and irregular terrains, which is a notable limitation in traditional rigidly connected underwater robots. These flexible joints provide increased degrees of freedom and enable the robot to absorb and release energy, ensuring stability even under external impacts, thus extending the operational lifespan of the robot. Finite element analysis demonstrates the flexible joints’ superior performance in various underwater conditions, offering a greater range of motion and workspace compared to rigid connections. The results indicate that the robot’s modular design, combined with the flexible joint module, leads to improved agility and maneuverability, allowing for precise and intentional operation. The control module, equipped with advanced sensors and a CPU, manages the complex dynamics introduced by the flexible joints, ensuring effective navigation and operation. The specific advantages of this design include the robot’s enhanced structural integrity, its ability to conform to irregular surfaces, and its adaptability to environmental variations. The paper concludes with a discussion on the implications of these findings for the future design and operation of underwater serpentine robots, emphasizing the need for a balance between the effects of elastic modulus and workspace to maximize the benefits of flexible joints.

Dynamic error modeling and analysis of articulated arm coordinate measuring machine with integrated joint module

The self-driven articulated arm coordinate measuring machine (AACMM) is a new non-orthogonal flexible coordinate measuring equipment providing automatic positioning and measurement with integrated joint modules introduced to its rotary joints. The dynamic measuring accuracy is the crucial indicator of the self-driven AACMM performance. However, the part manufacturing assembly error and structural dynamic deformation error will seriously reduce the dynamic measuring accuracy. Therefore, a modeling compensation approach based on error propagation and accumulation mechanism analysis of the above two aspects is necessarily required to preserve the automatic measuring accuracy of the self-driven AACMM. In this paper, the authors propose an innovative modeling method for evaluating and compensating for the dynamic measuring errors of the self-driven AACMM. The ideal measuring model is constructed assisted with on the classical Denavit–Hartenberg parameter methodology and the complex structure of the self-driven AACMM. The source and its propagation mechanism of the dynamic error from the self-driven AACMM were analyzed systematically to decouple the superposed error impact on measuring precision. The dynamic measurement error compensation mechanism model is presented considering the influence of instantaneous spatial attitude on the dynamic structural parameters of self-driven AACMM. Experimental results indicate that the single-point repeatability measurement error fluctuation of the self-driven AACMM is 0.046 mm. The radius measuring error of the gauge ball reduces from −0.26 mm to −0.19 mm after the error model compensation. Additionally, the measuring accuracy is demonstrated to be effectively improved by the dynamic error compensation mechanism model.

Advancing Hyperspectral Image Analysis with CTNet: An Approach with the Fusion of Spatial and Spectral Features.

Hyperspectral image classification remains challenging despite its potential due to the high dimensionality of the data and its limited spatial resolution. To address the limited data samples and less spatial resolution issues, this research paper presents a two-scale module-based CTNet (convolutional transformer network) for the enhancement of spatial and spectral features. In the first module, a virtual RGB image is created from the HSI dataset to improve the spatial features using a pre-trained ResNeXt model trained on natural images, whereas in the second module, PCA (principal component analysis) is applied to reduce the dimensions of the HSI data. After that, spectral features are improved using an EAVT (enhanced attention-based vision transformer). The EAVT contained a multiscale enhanced attention mechanism to capture the long-range correlation of the spectral features. Furthermore, a joint module with the fusion of spatial and spectral features is designed to generate an enhanced feature vector. Through comprehensive experiments, we demonstrate the performance and superiority of the proposed approach over state-of-the-art methods. We obtained AA (average accuracy) values of 97.87%, 97.46%, 98.25%, and 84.46% on the PU, PUC, SV, and Houston13 datasets, respectively.

Wind power forecasting: A transfer learning approach incorporating temporal convolution and adversarial training

Accurate forecasting of wind power is crucial for efficiently scheduling power grids that incorporate wind energy. However, distribution instability of time-varying data can undermine the generalization capability of forecasting models in future periods. This paper introduces a transfer learning approach incorporating adversarial training and temporal convolution to address wind power forecasting from the distribution-centric perspective. The proposed approach can be divided into two joint modules: the temporal domain split and domain-adversarial temporal convolutional network. The former segregates training data into temporal domains characterized by the most significant distribution differences, while the latter aims is designed to learn shared knowledge from the segmented temporal domains. The shared knowledge is independent of distribution shifts, which can be generalized to future testing well. Specifically, the domain-adversarial temporal convolutional network is composed of a base model incorporating temporal convolutional network and a domain classifier. Both components are jointly optimized by minimizing power prediction loss and maximizing temporal domain classification loss. The data from actual wind turbines are employed to validate the proposed approach. Experimental results show that the superior performance of the proposed approach compared to benchmark models, achieving a remarkable 24.56% improvement in accuracy by introducing adversarial training.

Research on a Biomimetic Flexible Ball Joint With Variable Stiffness for Robots

Abstract The human ball joint is a compact and flexible 3-degree-of-freedom (DOF) composite joint. The ball joints commonly used in robot design lack active driving capability. Ball joints with active drive generally consist of three single-degree-of-freedom joints connected in series, which is a noncompacted structure and easily leads to singular postures. In order to meet the demand for high-performance composite joint modules for service robots, this article designs a flexible biomimetic spherical robot joint with variable stiffness characteristics: the mechanism of muscle parallel antagonistic drive and ligament wrapping constraint is simulated; three parallel branch chains are used to drive three composite degrees-of-freedom; ropes, soft airbags, and series elastic drive gears are used to form a flexible transmission system; the contour of the rope winch has been optimized with the aim of transmission stability; and a pneumatic variable stiffness soft structure has been designed and fabricated. A compliance control algorithm for joints was developed based on the principle of impedance control. The research results indicate that the biomimetic ball-and-socket joint has a compact structure, a wide range of motion and good motion tracking performance, variable stiffness performance, and flexible interaction ability.

Joint module importance measures for multistate systems

Joint module importance measures for multistate systems

Joint-Module Health Status Recognition for an Unmanned Platform: A Time–Frequency Representation and Extraction Network-Based Approach

Due to the complex structure of the joint module and harsh working conditions of unmanned platforms, the fault information is often overwhelmed by noise. Moreover, traditional mechanical health state recognition methods usually require a large amount of labeled data in advance, which is difficult to obtain for specific fault data in engineering applications. This limited amount of fault data restricts the diagnostic performance. Additionally, the characteristics of convolutional neural networks (CNNs) limit their ability to capture the relative positional information of fault features. In order to obtain more comprehensive fault information, this paper proposes an intelligent health state recognition method for unmanned platform joint modules based on feature modal decomposition (FMD) and the enhanced capsule network. Firstly, the collected vibration signals are decomposed into a series of feature modal components using FMD. Then, time–frequency maps containing significant fault features are generated based on the continuous wavelet transform (CWT). Finally, a multi-scale feature enhancement (MLFE) module and an efficient channel attention (ECA) module are proposed to enhance the feature extraction capability of the capsule network, extracting more comprehensive global and local feature information from the time–frequency maps to achieve the intelligent state recognition of joint modules. This approach enhances fault features while reducing the impact of redundant features, significantly improving the feature extraction capability without increasing the model’s computational complexity. The effectiveness and superiority of the proposed method are validated through experiments on an unmanned platform joint-module testbed. An ablation experiment demonstrates the effectiveness of the MLFE and ECA modules, and a comparison with other advanced network models proves the superiority of the proposed method for health status recognition.

УДОСКОНАЛЕННЯ ІСНУЮЧИХ МЕТОДІВ ОЦІНЮВАННЯ ФУНКЦІОНУВАННЯ АЕРОПОРТІВ ПРИ МІЖНАРОДНІЙ ІНТЕГРАЦІЇ

The article is devoted to the study of methods for assessing the operation of airports under international integration and their improvement. The recent scientific studies on assessing the operation of international airports were analyzed and it was found that the most actively used methods are the initial data envelopment analysis (DEA) and various variants of quantile regression, Tobit regression, assurance region methods, SBI and DMU, Malmquist index, classification and regression tree, bootstrapping tool and other tools. It was confirmed that the existing methods for assessing the airports’ operation have a significant number of indicators for calculations based on the evaluation of economic, technical or overall efficiency. The parametric methods for assessing the operation of airports, their advantages and disadvantages were characterized. The methodology of stochastic frontier analysis (SFA), the use of a non-parametric approach to evaluate airport performance, and the evaluation of the overall efficiency of two-stage DMUs with shared resources were described. The methodical approaches based on the use of the spatial regression model, the AAG model and the stochastic frontier model were characterized. The methodology for determining the target function Effi to assess the efficiency of each airport DMUi was described. The scheme for assessing the airports’ operation by input and output indicators was proposed, which includes a three-module assessment system for the passenger, cargo and joint modules. The cargo module is the authors’ own development and its creation is justified by the fact that a significant number of air cargo flows are served by passenger aircraft. The group of input indicators of the cargo module includes the length of the loading and unloading work front, the number of cargo receiving points, the working area of rack and stack storage, the number of electric forklifts and lift trucks, as well as the number of cargo warehouses, and the number of general and special categories of cargo are defined as output indicators.

Dynamic Event-Triggered Strategy-Based Optimal Control of Modular Robot Manipulator: A Multiplayer Nonzero-Sum Game Perspective.

Due to the limited computing and processing ability of modular robot manipulator (MRM) components, such as sensors and controllers, event-triggered mechanisms are considered a crucial communication paradigm shift in resource constrained applications. Dynamic event-triggered mechanism is developing into a new technology by reason of its higher resource utilization efficiency and more flexible system design requirements than traditional event-triggered. Therefore, an optimal control scheme of multiplayer nonzero-sum game based on dynamic event-triggered is developed for MRM systems with uncertain disturbances. First, dynamic model of the MRM is established according to joint torque feedback technique and model uncertainty is estimated by data-driven-based neural network identifier. In the framework of differential game, the tracking control problem of MRM system is transformed into the optimal control problem for multiplayer nonzero-sum game with the control input of each joint module as the player. Then, the static event-triggered control problem of MRM system is studied based on adaptive dynamic programming algorithm. On this basis, the internal dynamic variable describing the previous state of the system is introduced, and the characteristics of dynamic trigger rule and its relationship with static rule are revealed theoretically. By designing an exponential attenuation signal, the minimum sampling interval of the system is always positive, so that Zeno behavior is excluded. Lyapunov theory proves that the system is asymptotically stable and the experimental results verify the validity of the proposed method.

STELO: A New Modular Robotic Gait Device for Acquired Brain Injury-Exploring Its Usability.

In recent years, the prevalence of acquired brain injury (ABI) has been on the rise, leading to impaired gait functionality in affected individuals. Traditional gait exoskeletons are typically rigid and bilateral and lack adaptability. To address this, the STELO, a pioneering modular gait-assistive device, was developed. This device can be externally configured with joint modules to cater to the diverse impairments of each patient, aiming to enhance adaptability and efficiency. This study aims to assess the safety and usability of the initial functional modular prototype, STELO, in a sample of 14 ABI-diagnosed participants. Adverse events, device adjustment assistance and time, and gait performance were evaluated during three sessions of device use. The results revealed that STELO was safe, with no serious adverse events reported. The need for assistance and time required for device adjustment decreased progressively over the sessions. Although there was no significant improvement in walking speed observed after three sessions of using STELO, participants and therapists reported satisfactory levels of comfort and usability in questionnaires. Overall, this study demonstrates that the STELO modular device offers a safe and adaptable solution for individuals with ABI, with positive user and therapist feedback.

Robust approximate constraint-following control design based on Udwadia–Kalaba approach and experimental validation for the joint module of cooperative robot

Robust approximate constraint-following control design based on Udwadia–Kalaba approach and experimental validation for the joint module of cooperative robot

A 3-DOF coupled tendon-driven humanoid waist

This study proposes a 3-degree-of-freedom (3-DOF) coupled tendon-driven humanoid waist featuring motion characteristics similar to those of the human waist. The tendon routing form of the waist was designed based on a 3-DOF coupled tendon-driven joint module considering the maximum balanced moment in the three directions. The ratios between the maximum joint torques for the pitch, roll, and yaw were designed to be 4:2:1 by the selection of a suitable tendon routing form and the adjustment of the pulley radius ratio. Two experiments were conducted to evaluate the maximum static torque and structural accuracy. The maximum static torques were 87.0, 53.0, and 22.2 Nm for the pitch, roll, and yaw, respectively. This was nearly consistent with the design considering the friction effect. An experiment for structural accuracy evaluation was conducted at different loads and different velocities to investigate the performance of the coupled tendon-driven mechanism. The experimental results indicate that, although the trajectory of the mechanism fluctuates greatly during the start and stop phases due to the large kinematic inertia of the mechanism and the stretching of the tendons, the trajectory also stabilizes when the velocity stabilizes.