Static Modeling of Composite Continuum Robot Considering Coupling EffectSu, Jing; Gao, Tao; Lin, Zechen; Luo, Yulu; Song, Rui; Du, Fuxin
doi: 10.1088/1742-6596/3024/1/012030pmid: N/A
Continuum robots, characterized by their flexible and hyper-redundant structure, exhibit exceptional adaptability for navigating complex anatomical regions, demonstrating significant potential in minimally invasive surgical applications. This paper proposes a static model for the composite continuum robot incorporating the coupling effect. The tension transmission characteristics are derived by analyzing the driving cables’ frictional losses at nodal points. An accurate static model is established by considering both the end-load conditions of flexible beams and the coupling interactions between the concentric tube and notched continuum. Based on the coupling effect analysis and Euler-Bernoulli beam theory, deformation-driving relationships under various motion conditions are formulated for the composite continuum robot. Simulation and experimental verification demonstrate that the proposed static model achieves an average positioning error of 0.486 mm, exhibiting 1.846 times improvement in spatial accuracy compared with conventional constant curvature models. This static model provides theoretical foundations for precise control and clinical applications of composite continuum robots.
Kinematic Analysis and Printing Experiments of a Multi-DOF Robotic Arm-Based Additive ManufacturingLu, Kunteng; Han, Xingguo; Liu, Xuan; Liu, Nankang
doi: 10.1088/1742-6596/3024/1/012032pmid: N/A
This study developed a multi-degree-of-freedom (DOF) robotic arm additive manufacturing(AM) device to address the limitations of traditional three-axis and five-DOF additive manufacturing path planning, thereby enables unsupported printing for curved-surface models. The research employed a structural configuration combining a six-DOF robotic arm with a two-DOF Positioning machine. The kinematics model of the additive manufacturing system of the multi-degree-of-freedom robot arm is established by using the standard Denavit-Hartenberg (D-H) parameter method, and the forward and inverse kinematics equations are deduced systematically. Experimental results demonstrate that the proposed device can efficiently complete printing tasks for simple curved-surface components, providing both hardware solutions and methodological support for multi- DOF additive manufacturing technologies.
Turnout Switch Machine Fault Diagnosis Based on 1DCBAYuan, Zheyu; Qin, Baofu; He, Deqiang; Wang, Yanbo; Zhao, Jiayang
doi: 10.1088/1742-6596/3024/1/012001pmid: N/A
Turnout switch machine fault data is characterized by limited quantity and imbalanced class distribution. This limits the use of deep learning in fault diagnosis applications. Therefore, this paper presents a fault diagnosis method for turnout switch machines based on the 1DCBA model. Firstly, to mitigate the problem of insufficient fault data, the KMeans-SMOTE is employed to oversample the current data, effectively expanding the dataset. Secondly, a 1D-CNN is utilized to capture local features, and it is combined with a BiLSTM to exploit the temporal dependencies within the data. Finally, an attention mechanism is introduced to improve the model’s classification ability and fault diagnosis performance by allowing it to focus on the most critical features. The results of experiments indicate that the suggested approach attains a fault diagnosis accuracy of 100%.
CSEA: Confidence-guided Semantic-Enhanced Alignment for Online Test-Time AdaptationZhang, Tengfei; Wu, Ya; Huang, Yuan; Liu, Zhengfa; Zeng, Dequan
doi: 10.1088/1742-6596/3024/1/012034pmid: N/A
Online test-time adaptation (OTTA) is a more general method for accomplishing domain adaptation in real-world scenarios because it just needs a pre-trained source model to adapt the mini-batch target online. One popular approach is to adapt the model with empirical risk minimization using generated pseudo-labels. However, it can fail to learn optimal feature space due to the noisy pseudo-labels under domain shift. To address this restriction, we propose a novel online test-time adaptation technique called Confidence-guided Semantic-Enhanced Alignment (CSEA) that separates the mini-batch target into distinct subsets based on the pseudo-labels confidence and applies tailored constraints to each subset for reliable target structure mining. We first divide the mini-batch target data into confident and non-confident based on an online updated memory bank for each class. Then, we customize the alignment strategy to fit each subset best. Specifically, we propose a Re-weighted Local Clustering for the confident subset to learn the local semantic structure information of unlabeled target. Further, we propose a Semantic-enhanced Alignment that applied to the non-confident subset to guide the model to learn additional discriminative semantic knowledge. Extensive experiments on three OTTA benchmarks indicate effectiveness of the proposed CSEA.
Intelligent Test System Design for Aerospace Equipment HarnessesSun, Ping; Zhang, Yong; He, Nan; Zhou, Gongping; Li, Mei
doi: 10.1088/1742-6596/3024/1/012020pmid: N/A
Aiming at the current aerospace wiring harness, which is prone to breaks, short circuits, insulation failures, contract issues, and other quality problems due to the complex structure and numerous connections, an intelligent wiring harness test system was developed by combining the current development trend of intelligent manufacturing industry and Internet of Things technology. Experimental verification shows that the system greatly improves the detection efficiency of the wiring harness test, reduces the labor intensity of the wiring harness test, effectively reduces the test errors introduced by human factors in the control component products, and ensures the quality of the products and the production schedule.
Research on Obstacle Avoidance Path Planning for Robotic Arms Using Improved RRT AlgorithmPan, Fenglian; Zhao, Mingyu; Tan, Zhicheng; Li, Guo; Wang, Yongjun; Yao, Xing
doi: 10.1088/1742-6596/3024/1/012031pmid: N/A
In robot path planning, the Rapidly-exploring Random Tree (RRT) algorithm is extensively employed for its effective search performance. However, the conventional RRT algorithm exhibits several limitations, including non-smooth paths, low search efficiency, and insufficient adaptability to dynamic environments. To tackle these issues, this paper introduces an improved RRT algorithm that integrates a target-biased sampling strategy, bidirectional expansion, adaptive step size adjustment, multi-objective optimization, and path smoothing. The target-biased sampling strategy dynamically adjusts the sampling probability, enhancing global exploration capability and accelerating convergence speed. The bidirectional expansion mechanism enables the search tree to expand concurrently from both the start and goal points, thereby substantially decreasing the search time. Adaptive step size adjustment modifies the step size based on the distance to the goal, improving the accuracy of path planning. Additionally, multi-objective optimization, combined with genetic algorithms, optimizes path length, smoothness, and motion time. Path smoothing is achieved through the application of cubic B-splines, which optimize path continuity and minimize unnecessary sharp turns. The experimental outcomes indicate that the enhanced algorithm surpasses conventional RRT and RRT* algorithms in search efficiency, path quality, and computational performance. This makes it a more effective approach for robot path planning in complex and dynamic environments.
Research and Application of Multi-Source Sensor Fusion Localization Algorithm for Complex ScenariosWang, Qingshan; Zhang, Dengquan; Cai, Yongxiang; Cai, Cong; Wang, Jiaochen; Guo, Peng
doi: 10.1088/1742-6596/3024/1/012015pmid: N/A
Due to remote locations such as mines and ports, GNSS signals have not achieved full coverage. They are weak, resulting in significant positioning errors for autonomous vehicles and making them unable to meet task requirements. In response to the problem of substantial positioning errors in autonomous driving vehicles relying solely on GNSS in complex scenarios, this paper first adopts the multi-sensor fusion of GNSS, LiDAR, and IMU. It proposes a loop detection based on the NDT algorithm SLAM architecture to construct a lightweight scene map. It provides high-precision and low-data volume scene information for autonomous driving positioning. Secondly, based on constructing a lightweight scene map, a multi-feature matching method combining point cloud features and point cloud strength information is adopted to achieve precise positioning of autonomous driving. Finally, the volumetric Kalman filter is used to reduce the influence of noise points on the filter by weighting the observed and predicted values of sensors, thereby achieving high-precision positioning of autonomous driving in complex scenes through multi-sensor positioning fusion. The experimental results show that the multi-feature matching algorithm is less than 36ms, the backend optimization of lightweight scene maps is 80ms, the compression ratio of high-precision point cloud maps is 92%, and the fusion positioning accuracy is less than 5cm.
Optimization and Efficiency Analysis of Lunar Water Vapor Condensation Collection Structure DevicesWang, Yinchao; Yin, Zihao; Han, Junnan; Zhang, Weiwei; Zu, Lin; Tao, Guanghong; Yu, Suyang; Jiang, Chunying; Ye, Changlong
doi: 10.1088/1742-6596/3024/1/012009pmid: N/A
Aimed at the technical challenge of low water vapor collection efficiency in the in-situ thermal extraction of lunar polar water ice, this study, based on the heat and mass transfer characteristics of frost layer porous media, proposes a solution to enhance the gas-solid phase transition rate by optimizing the internal surface area of the water vapor collection unit. Through the establishment of a water vapor condensation dynamics model under vacuum and low-temperature conditions, the coupling effects of frost layer porosity and temperature field on the mass transfer process are revealed. A verification platform simulating the extreme lunar surface environment was constructed, and comparative experiments on different aluminum bead filling structures were conducted. The results indicate that the larger the internal surface area of the device, the higher the water vapor collection efficiency; the use of small-diameter aluminum beads significantly increases the internal surface area, but excessively small diameters may lead to pore blockage. The initial stage of water vapor collection is one of the key stages of mass transfer, where water vapor first comes into contact with the inner surface of the collection structure and undergoes a gas-solid phase change that affects the continued performance of the collection device. This research provides theoretical and experimental foundations for the design of water vapor collection units in in-situ lunar water ice extraction devices.
Design of Motor Speed Controller Based on FPGAWang, Yiding; Gao, Li; Gu, Mingxin; Wu, Guoguo; Lu, Peng
doi: 10.1088/1742-6596/3024/1/012022pmid: N/A
In the field of modern industrial automation, the accuracy and stability of motor speed control technology are essential to ensure production efficiency and product quality. The purpose of this research is to design and implement an FPGA motor speed controller based on a PID control algorithm to meet the needs of high-precision motor control. This design analyzes the principle of the PID control algorithm, including the function of proportion, integral, and differential three links and their adjustment methods. By optimizing the parameters of the PID controller, the high precision control of motor speed is realized. In the design process, this study elaborated on the FPGA implementation of PID controller-specific flow and key technology. The results show that the controller can quickly respond to the change in motor speed and output the corresponding PWM signal to adjust the motor speed.
Optimization Study of Precise Control of Commercial Vehicle Cooling System Based on MAP FeedforwardTang, Rongjiang; Zhao, Le; Liu, Quanhui; Wang, Wenwen; Jing, Xin
doi: 10.1088/1742-6596/3024/1/012024pmid: N/A
Aiming at the problems of overcooling, overheating, the insufficient control accuracy of water pump and fan, and wastage of power consumption of traditional thermal management systems of commercial vehicles under transient conditions, this paper constructs a thermal management system control strategy model based on GT-Suite and Simulink platform and carries out experimental validation of the model under the C-WTVC cyclic conditions, and the results show that the model has a high degree of accuracy. In this paper, the MAP feed-forward control strategy is adopted to query the MAP of heat dissipation demand through the engine speed and torque, combined with the ambient temperature correction algorithm to get the corrected heat dissipation demand and calculate the target speeds of the water pump and the fan based on the theory of minimum power allocation, and at the same time, introduce the fuzzy PID feedback control to realize the precise regulation. Simulation results show that the optimized thermal management system improves the coolant temperature control accuracy by 88.9%, reduces the total power consumption of the water pump and fan by 53.4% under the C-WTVC cycle condition, which effectively reduces the fuel consumption, and provides the theoretical basis and methodological support for the energy-saving optimization of thermal management system for commercial vehicles.