Parallel Mechanism Research Articles

Parallel Processing of Image Databases for Accelerated Morphological Analysis

Nuclear forensics relies on different signatures to identify the source of nuclear material. Such signatures include crystalline structure, chemical composition, and particle morphology. One way to quantify morphology in electron microscope imagery is through image segmentation, where pixels are assigned to several partitions (or groups) that correspond to particles, grains, and other objects of interest within the image. Once pixels are assigned to segments, it is relatively straightforward to quantify other quantities of interest, such as grain size, circularity, etc. However, the range and diversity of microscope images make it difficult to obtain an accurate segmentation automatically. The accuracy of segmentation can be improved through supervised learning, but this requires many images to be manually segmented. Another way to improve the accuracy is to use interactive segmentation. Interactive segmentation requires a human to provide image-specific user input to improve performance. However, the amount of user input (effort) is generally far less than is required for supervised learning. In this paper, we investigate several parallelization strategies to automatically explore the user input parameter space of interactive segmentation algorithms across a large number of images. Specifically, we investigate four different parallelization mechanisms in a high-performance computing (HPC) environment and use the Amdahl fraction to evaluate efficiency on multiple processor cores across multiple nodes. Ultimately, the parallelization strategy that was most efficient utilized the message passing interface integrated with the segmentation and quantification code. This strategy had an Amdahl fraction of 0.985 and a performance of about 0.251 s/image. These results indicate that the parameter space of interactive segmentation algorithms can be efficiently explored using HPC. This opens the door to future work where user input is reduced and where interactive image segmentation algorithms are automatically applied to large image sets.

Design, simulation, control of a hybrid pouring robot: enhancing automation level in the foundry industry

AbstractCurrently, workers in sand casting face harsh environments and the operation safety is poor. Existing pouring robots have insufficient stability and load-bearing capacity and cannot perform intelligent pouring according to the demand of pouring process. In this paper, a hybrid pouring robot is proposed to solve these limitations, and a vision-based hardware-in-the-loop (HIL) control technology is designed to achieve the real-time control problems of simulated pouring and pouring process. Firstly, based on the pouring mechanism and the motion demand of ladle, a hybrid pouring robot with a 2UPR-2RPU parallel mechanism as the main body is designed. And the equivalent hybrid kinematic model was established by using Eulerian method and differential motion. Subsequently, a motion control strategy based on HIL simulation technique was designed and presented. The working space of the robot was obtained through simulation experiments to meet the usage requirements. And the stability of the robot was tested through the key motion parameters of the robot joints. Based on the analysis of pouring quality and trajectory, optimal dynamic parameters for the experimental prototype are obtained through water simulation experiments, the pouring liquid height area is 35–40 cm, the average flow rate of pouring liquid is 112 cm3/s, and the ladle tilting speed is 0.0182 rad/s. Experimental results validate the reasonableness of the designed pouring robot structure. Its control system realizes the coordinated movement of each branch chain to complete the pouring tasks with different variable parameters. Consequently, the designed pouring robot will significantly enhance the automation level of the casting industry.

Configuration design and kinematic performance analysis of a parallel wheelset suspension for a multi-wheeled pressurized lunar rover

When a multi-wheeled pressurized lunar rover (MWPLR) travels over rough terrain, the ability of its suspension to maintain wheel–ground contact is critical for achieving heavy load distribution and increasing its payload capacity. If the MWPLR uses an independent suspension, increasing the number of wheels worsens the wheel-off-ground problem. If the MWPLR utilizes a rocker–bogie suspension with a planar open-loop kinematic chain, it will have difficulty withstanding lateral impacts. In this study, to solve the above problems, the configurations of the 4-SS parallel mechanism were enumerated based on the Grassmann line geometry and screw theory to design a two-degree-of-freedom parallel wheelset suspension. The suspension integrated two axially parallel wheels on the same moving platform. Kinematic analysis showed that no singularities occurred in the workspace. The moving platform enabled large vertical movement and rotation of ±30°. Simulations showed that the suspension dramatically improved the ride comfort and wheel–ground contact capability of the MWPLR.

On the extension of the mechanism for parallel import of goods

The material highlights the positive and negative aspects of the implementation of the parallel import mechanism. The main changes in its work in 2024 are presented. The use of a simplified certification scheme by market participants is noted to be highly effective.

Research on Ground Microgravity Simulation System Based on Parallel Mechanism

Research on Ground Microgravity Simulation System Based on Parallel Mechanism

Enhancing Vegetable Sales Forecasting with A CNN-LSTM-Transformer Hybrid Model

Due to the short shelf life of vegetable products, a significant portion of the inventory cannot be resold the following day. To facilitate more informed procurement decisions in superstores and minimize vegetable wastage, this study proposes a hybrid prediction model based on CNN-LSTM-Transformer for enhancing the accuracy of forecasting vegetable sales volumes. Firstly, an LSTM model is incorporated to account for the recurring and seasonal variations in vegetable sales. Secondly, a CNN model is introduced to address the limitations of LSTM in capturing spatial data components. The convolutional and pooling layers of CNN help establish spatial relationships among different feature values in the dataset. Finally, a Transformer model is integrated to tackle the issue of long-term dependencies, which LSTM alone struggles to resolve. The Transformer model employs a parallel attention mechanism, eliminating temporal dependencies and effectively addressing LSTM's long-term dependency challenge. This integration also accelerates model training. The evaluation metric employed in this paper is RMSE (Root Mean Square Error). The three-year sales volume of 11 vegetable types is predicted using seasonal ARIMA, XGBoost, LSTM, CNN-LSTM, and CNN-LSTM-Transformer models. The results indicate that the CNN-LSTM-Transformer model achieves the lowest RMSE at 0.0758, followed by ARIMA at 0.0792, CNN-LSTM at 0.0830, XGBoost at 0.0858, and LSTM at 0.0913. These findings demonstrate that the CNN-LSTM-Transformer model exhibits superior accuracy in forecasting vegetable sales volumes, yielding more precise predictions. This accurate forecasting not only aids superstores in reducing waste and enhancing profitability but also assists government authorities in rationalizing vegetable subsidy policies and optimizing the vegetable production and marketing system.

High-Speed Cell Assembly with Piezo-Driven Two-Finger Microhand

In the past few decades, researchers have conducted extensive studies on cell micromanipulation methods. However, there has consistently been a lack of a micromanipulation system that excels in both precision and speed. Additionally, many of these methods rely on manual control, thus significantly reducing efficiency. In this paper, a robotized micromanipulation system employing a two-finger microhand is proposed. The microhand has a 3-DoF parallel mechanism driven by three piezoelectric actuators, enabling high-precision micromanipulation. Replacing the needle-tip end-effector with a hemispherical end-effector makes cell grasping easier and more stable. In addition, a vibration-based release method combined with gel coating is proposed to reduce the release difficulty caused by adhesion forces. Through multiple sets of experiments, we have determined the optimal grasping and releasing conditions while balancing precision, stability, and damage degree to cells. An automated cell assembly strategy based on microscopic visual feedback and pick-and-place path planning is proposed to achieve the robotized high-speed cell array. Hela cells were chosen as the operation objects, achieving a 95% success rate in grasping and a 97% success rate in releasing. A “T” letter array formed by cells was successfully assembled with an average grasp and release time of less than 0.8 s and an assembly accuracy of 4.5 μm for a single cell. This study holds significant implications for the fields of biology and medicine, presenting potential applications in tissue engineering.

Active Compliance Smart Control Strategy of Hybrid Mechanism for Bonnet Polishing.

Compliance control strategies have been utilised for the ultraprecision polishing process for many years. Most researchers execute active compliance control strategies by employing impedance control law on a robot development platform. However, these methods are limited by the load capacity, positioning accuracy, and repeatability of polishing mechanisms. Moreover, a sophisticated actuator mounted at the end of the end-effector of robots is difficult to maintain in the polishing scenario. In contrast, a hybrid mechanism for polishing that possesses the advantages of serial and parallel mechanisms can mitigate the above problems, especially when an active compliance control strategy is employed. In this research, a high-frequency-impedance robust force control strategy is proposed. It outputs a position adjustment value directly according to a contact pressure adjustment value. An open architecture control system with customised software is developed to respond to external interrupts during the polishing procedure, implementing the active compliance control strategy on a hybrid mechanism. Through this method, the hybrid mechanism can adapt to the external environment with a given contact pressure automatically instead of relying on estimating the environment stiffness. Experimental results show that the proposed strategy adapts the unknown freeform surface without overshooting and improves the surface quality. The average surface roughness value decreases from 0.057 um to 0.027 um.

Simulation and Experimental Study of Non-Resonant Vibration-Assisted Lapping of SiCp/Al.

SiCp/Al is a difficult-to-machine material that makes it easy to produce surface defects during machining, and researchers focus on reducing the surface defects. Vibration-assisted machining technology is considered an effective method to reduce surface defects by changing the trajectory and contact mode of the abrasive. Aiming at the problem of SiCp/Al processing technology, a vibration-assisted lapping device (VLD) is designed, and elliptical motion is synthesized by a set of parallel symmetrical displacement output mechanisms. The working parameters of the device were tested by simulation and experiment, and the lapping performance was verified. Then, the effects of removal characteristics and process parameters on surface roughness and lapping force were analyzed by simulation and experiment. Simulation and experimental results show that frequency and amplitude that are too low or too high are not conducive to the advantages of NVL. The best surface quality was 54 nm, obtained at A = 8 μm and f = 850 Hz.

Early life adversity in primates: Behavioral, endocrine, and neural effects

BackgroundEvidence suggests that early life adversity is associated with maladaptive behaviors and is commonly an antecedent of stress-related psychopathology. This is particularly relevant to rearing in primate species as infant primates depend on prolonged, nurturant rearing by caregivers for normal development. To further understand the consequences of early life rearing adversity, and the relation among alterations in behavior, physiology and brain function, we assessed young monkeys that had experienced maternal separation followed by peer rearing with behavioral, endocrine and multimodal neuroimaging measures. Methods50 young rhesus monkeys were studied, half of which were rejected by their mothers and peer reared, and the other half were reared by their mothers. Assessments were performed at approximately 1.8 years of age and included: threat related behavioral and cortisol responses, cerebrospinal fluid (CSF) measurements of oxytocin and corticotropin releasing hormone (CRH), and multimodal neuroimaging measures (anatomical scans, resting functional connectivity, diffusion tensor imaging, and threat-related regional glucose metabolism). ResultsThe results demonstrated alterations across behavioral, endocrine, and neuroimaging measures in young monkeys that were reared without their mothers. At a behavioral level in response to a potential threat, peer reared animals engaged in significantly less freezing behavior (p = 0.022) along with increased self-directed behaviors (p < 0.012). Levels of oxytocin in the CSF, but not plasma, were significantly reduced in the peer reared animals (p = 0.019). No differences in plasma cortisol or CSF CRH were observed. Diffusion tensor imaging revealed significantly decreased white matter density across the brain. Exploratory correlational and permutation analyses suggest that the impact of peer rearing on behavior, endocrine and brain structural alterations are mediated by separate parallel mechanisms. ConclusionsTaken together, these results demonstrate in NHPs the importance of maternal rearing on the development of brain, behavior and hormonal systems that are linked to social functioning and adaptive responses. The findings suggest that the effects of maternal deprivation are mediated via multiple independent pathways which may account for the heterogeneity in behavioral and biological alterations observed in individuals that have experienced this early life adversity.

Design and optimization of a novel sagittal-plane knee exoskeleton with remote-center-of-motion mechanism

Powered exoskeleton is a wearable robot that can provide power-assisted motion for the human body. One of the challenges in the exoskeleton research is how to improve its kinematic synergy. In this paper, a novel knee exoskeleton robot is designed to improve the kinematic synergy between the exoskeleton and the human body. The novel exoskeleton adopts the sagittal-plane layout to reduce the bias moment, and uses a remote-center-of-rotation mechanism to ensure the coincidence of the rotation centers of the exoskeleton and the human body. To analyze the influence of the interaction error on the performance of the exoskeleton, this paper proposes a human-exoskeleton fusion model based on the virtual equivalent parallel mechanism method, and parameterize the interaction error through virtual kinematic chains. Finally, the assisting performance of the novel exoskeleton is analyzed and verified through simulation experiments, motion experiments, body motion experiments and human simulation experiments. The human-exoskeleton fusion model proposed in this paper has positive significance for the configuration design, biomechanical analysis and application of exoskeleton robots.

High-Throughput MPSoC Implementation of Sparse Bayesian Learning Algorithm

In the field of sparse signal reconstruction, sparse Bayesian learning (SBL) has excellent performance, which is accompanied by extremely high computational complexity. This paper presents an efficient SBL hardware and software (HW&amp;SW) co-implementation method using the ZYNQ series MPSoC (multiprocessor system-on-chip). Firstly, considering the inherent challenges in parallelizing iterative algorithms like SBL, we propose an architecture based on the iterative calculations implemented on the PL side (FPGA) and the iteration control and input management handled by the PS side (ARM). By adopting this structure, we can take advantage of task-level pipelines on the FPGA side, effectively utilizing time and space resources. Secondly, we utilize LDL decomposition to perform the inversion of the Hermitian matrix, which not only exhibits the lowest computational complexity and requires fewer computational resources but also achieves a higher level in the parallel pipeline mechanism compared with other alternative methods. Furthermore, the algorithm conducts iterations sequentially, utilizing the parameters derived from the previous dataset as prior information for initializing the subsequent dataset’s initial values. This approach helps to reduce the number of iterations required. Finally, with the help of Vitis HLS 2022.2 and Vivado tools, we successfully accomplished the development of a hardware design language and its implementation on the ZYNQ UltraScale+ MPSoC ZCU102 platform. Meanwhile, we have solved a direction of arrival (DOA) estimation problem using horizontal line arrays to verify the practical feasibility of the method.

2 DOF transformable wheel design based on geared 8 bar parallel linkage mechanism

This paper introduces a novel design and static optimization for a two-degrees-of-freedom transformable wheel based on a geared linkage mechanism. Overcoming obstacles, including stairs, with small wheels is a major challenge in the field of mobile robotics research. Among various robots, the transformable wheel, which can change the shape of the wheel to overcome steps and optimize the path, was presented and has undergone many improvements. Nevertheless, problems such as asymmetry and structural strength remain. Therefore, the design of this paper aims to address the structural inefficiencies identified in the previous research model, which were attributed to the asymmetric placement of the linear motion guide. Through the implementation of this mechanism, the linear motion of the lobe can be segregated, enabling each input motor to share the workload effectively. The optimization process focus on determining the optimal linkage length under static conditions, resulting in improved structural characteristics and force distribution of linkage within the designated workspace. As a result, asymmetry of motion is eliminated, required intervention angle of the driving motor and stress of linkage was reduced by 36.24% and 8.35%, respectively.

Ammonium Phosphotungstate Bonded on Imidazolized Activated Carbon for Selective Catalytic Rearrangement of α-Epoxypinane to Carveol

Carveol is a rare fine chemical with specific biological activities and functions in nature. The artificial synthesis of carveol from plentiful and cheap turpentine is expected to further improve development of pharmaceutical and industrial applications. A new green catalytic system for the preparation of high-value carveol from α-epoxypinane is presented. A novel ammonium salt solid acid (AC-COIMI-NH4PW) was obtained from phosphotungstic acid bonded with imidazole basic site on nitrogen-doped activated carbon which, after ammonia fumigation, presented an excellent catalytic performance for the selective rearrangement of α-epoxypinane to carveol in DMF as solvent under mild reaction conditions. At 90 °C for 2 h, the conversion of α-epoxypinane could reach 98.9% and the selectivity of carveol was 50.6%. The acidic catalytic sites exhibited superior durability and the catalytic performance can be restored by supplementing the lost catalyst. Based on the investigation of catalytic processes, a parallel catalytic mechanism for the main product was proposed from the rearrangement of α-epoxypinane on AC-COIMI-NH4PW.

Heter-Train: A Distributed Training Framework Based on Semi-Asynchronous Parallel Mechanism for Heterogeneous Intelligent Transportation Systems

Transportation big data (TBD) are increasingly combined with artificial intelligence to mine novel patterns and information due to the powerful representational capabilities of deep neural networks (DNNs), especially for anti-COVID19 applications. The distributed cloud-edge-vehicle training architecture has been applied to accelerate DNNs training while ensuring low latency and high privacy for TBD processing. However, multiple intelligent devices (e.g., intelligent vehicles, edge computing chips at base stations) and different networks in intelligent transportation systems lead to computing power and communication heterogeneity among distributed nodes. Existing parallel training mechanisms perform poorly on heterogeneous cloud-edge-vehicle clusters. The synchronous parallel mechanism may force fast workers to wait for the slowest worker for synchronization, thus wasting their computing power. The asynchronous mechanism has communication bottlenecks and can exacerbate the straggler problem, causing increased training iterations and even incorrect convergence. In this paper, we introduce a distributed training framework, Heter-Train. First, a communication-efficient semi-asynchronous parallel mechanism (SAP-SGD) is proposed, which can take full advantage of acceleration effect of asynchronous strategy on heterogeneous training and constrain the straggler problem by using global interval synchronization. Second, Considering the difference in node bandwidth, we design a solution for heterogeneous communication. Moreover, a novel weighted aggregation strategy is proposed to aggregate the model parameters with different versions. Finally, experimental results show that our proposed strategy can achieve up to <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$6.74 \times$</tex-math> </inline-formula> speedups on training time, with almost no accuracy decrease.

Parallel Elastic Self-Alignment Mechanism Enhances Energy Efficiency and Reduces Misalignment in a Powered Knee Exoskeleton.

This paper aims to enhance exoskeleton compliance during locomotion assistance by reducing misalignment and to improve energy efficiency by overcoming the limitations posed by the bulky structure of powered rigid exoskeletons. A novel compliant knee exoskeleton, featuring a parallel elastic self-alignment mechanism, has been developed and structurally optimized. The exoskeleton uses adaptive oscillators to determine the wearer's gait phase and provides real-time assistance to the knee joint. Bench tests demonstrate that the parallel elastic mechanism significantly reduces the driving torque of the knee exoskeleton. Performance evaluations reveal that, compared to a commercial orthosis, the root-mean-square of knee angle error, joint misalignment, and unexpected interaction forces are reduced by 16.5 ± 11.3%, 23.3 ± 4.9%, and 17.7 ± 1.3%, respectively. Gait intervention experiments show reductions in average and maximum muscle activity of the knee joint by 7.6 ± 4.9% and 23.2 ± 5.7%, respectively. Additionally, the exoskeleton decreases negative work performed by the knee joint and the total lower limb by 22.7% and 8.6%, respectively. The parallel elastic self-alignment mechanism effectively mitigates joint misalignment, while the parallel springs offer partial gravity compensation, thereby enhancing both the energy efficiency and locomotion assistance of the exoskeleton. The parallel elastic self-alignment mechanism effectively addresses both misalignment and energy efficiency challenges in powered exoskeletons, providing valuable insights for future design improvements.

Design of a novel two-degree-of-freedom translational-rotation low-frequency vibration isolation platform

Multi-dimensional vibration isolation platforms often use parallel mechanisms to achieve multi-dimensional vibration isolation control. However, due to the high stiffness of the parallel mechanism, its own natural frequency is high, and it has good performance when applied to high-frequency vibration isolation, but it is hard to achieve low-frequency vibration isolation. This paper aims at the problem that the actual polishing and grinding equipment is often subjected to axial and circumferential low-frequency disturbances during operation, a novel C/2-(2-RRR) RR two-degree-of-freedom (2-DOF) translational-rotation low-frequency vibration isolation platform is proposed based on the singular configuration of planar 2-RRR mechanism. The coupling dynamic model of the vibration isolation platform is established, and the amplitude-frequency curve and force transmissibility curve are analyzed. The simulation analysis and prototype experiment are carried out by using the independent external excitation in both translational and rotational directions, and the corresponding linear system is compared to verify the effectiveness of the low-frequency vibration isolation of the two-degree-of-freedom vibration isolation platform.

Analysis of Repetitive Positioning Accuracy of Parallel Mechanisms Based on Monte Carlo Method

Based on the inverse kinematics model of the Stewart parallel mechanism, a motion platform pose error model was constructed through differential transformation. The error model shows that the motion platform pose error comes from the length error of 6 struts and the position error of 12 hinge centers, with a total of 42 error terms. Monte Carlo simulation was used to analyze the effects of struts length error on the pose error of the moving platform. The results show that when the length error of the struts follows a normal distribution, the posture error of the moving platform approximately follows a normal distribution. Monte Carlo simulation is a highly effective error analysis method that can analyze the error distribution pattern of the mechanism before processing and assembly, which provides a method for error analysis of Stewart parallel mechanisms.

Optimal Parameter Design and Microrobotic Navigation Control of Parallel-Mobile-Coil Systems

In this work, we study the optimal parameter design and microrobotic navigation control of the parallel-mobile-coil system (PMCS) that consists of three mobile electromagnetic coils. With motion driven by a parallel mechanism, the three coils can move in 3D large space and keep as close as possible to the controlled microrobot for magnetic actuation. Although promising for microrobotic applications, how to design such a type of system for a specific workspace requirement is untackled. Regarding this issue, we propose a computational design method, by which one can calculate the structural parameters of a PMCS starting from a required cylindrical workspace. With the derived performance metrics for motion actuation and magnetic actuation of the PMCS, the system actuation performance (composed of motion and magnetic actuation) is optimized. Utilizing the design method, we optimally construct a prototype PMCS for microrobotic navigation. We then conduct experiments to validate the demanding field/force generation capability of the PMCS and demonstrate the navigation control of different types of magnetic microrobots. In particular, we design closed-loop motion controllers for both torque and force-driven microrobots, using which automated large-workspace and high-accuracy trajectory tracking is realized. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This work is motivated by the recent wide interest in magnetic microrobots. Driven by external magnetic fields, magnetic microrobots can navigate in a wireless manner for targeted delivery/therapy. To promote microrobot applications to the human body, a magnetic actuation system with large workspace is desirable. However, due to the fast decay of magnetic field, the commonly used stationary coil-based magnetic actuation systems have the workspace scalability problem. Thus, several mobile-coil-based systems have been designed. In this work, we propose an optimal design method for the PMCS, using which one can design a PMCS starting from a cylindrical workspace with performance being optimized. We construct a PMCS prototype with a workspace of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\Phi 230 \times 100$</tex-math> </inline-formula> mm <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$^3$</tex-math> </inline-formula> , and we then study the automated microrobotic navigation control methods for the PMCS. Controllers are designed for different types of magnetic microrobots, and experiments show that, using the controllers, the PMCS can perform automated large-workspace microrobotic navigation control with high accuracy.

Multi-Head Graph Convolutional Network for Structural Connectome Classification.

We tackle classification based on brain connectivity derived from diffusion magnetic resonance images. We propose a machine-learning model inspired by graph convolutional networks (GCNs), which takes a brain-connectivity input graph and processes the data separately through a parallel GCN mechanism with multiple heads. The proposed network is a simple design that employs different heads involving graph convolutions focused on edges and nodes, thoroughly capturing representations from the input data. To test the ability of our model to extract complementary and representative features from brain connectivity data, we chose the task of sex classification. This quantifies the degree to which the connectome varies depending on the sex, which is important for improving our understanding of health and disease in both sexes. We show experiments on two publicly available datasets: PREVENT-AD (347 subjects) and OASIS3 (771 subjects). The proposed model demonstrates the highest performance compared to the existing machine-learning algorithms we tested, including classical methods and (graph and non-graph) deep learning. We provide a detailed analysis of each component of our model.