Top Read Articles

Published in last 1 year |

In last 2 years |

In last 3 years |

All

Please wait a minute...


For Selected: Download Citations EndNote Ris BibTeX Toggle Thumbnails

Select

Review of Power-Assisted Lower Limb Exoskeleton Robot HE Guisong (贺贵松), HUANG Xuegong (黄学功), LI Feng (李峰), WANG Huixing (汪辉兴) J Shanghai Jiaotong Univ Sci    2024, 29 (1): 1-15.   DOI: 10.1007/s12204-022-2489-3 Abstract （735）      PDF（pc） （1195KB）（205）       Knowledge map    Save Power-assisted lower limb exoskeleton robot is a wearable intelligent robot system involving mechanics,materials, electronics, control, robotics, and many other fields. The system can use external energy to provide additional power to humans, enhance the function of the human body, and help the wearer to bear weight that is previously unbearable. At the same time, employing reasonable structure design and passive energy storage can also assist in specific actions. First, this paper introduces the research status of power-assisted lower limb exoskeleton robots at home and abroad, and analyzes several typical prototypes in detail. Then, the key technologies such as structure design, driving mode, sensing technology, control method, energy management, and human-machine coupling are summarized, and some common design methods of the exoskeleton robot are summarized and compared. Finally, the existing problems and possible solutions in the research of power-assisted lower limb exoskeleton robots are summarized, and the prospect of future development trend has been analyzed. Reference | Related Articles | Metrics | Comments（0）

Select

Review of Key Technologies for Developing Personalized Lower Limb Rehabilitative Exoskeleton Robots TAO Jing, (陶璟), ZHOU Zhenhuan (周振欢) J Shanghai Jiaotong Univ Sci    2024, 29 (1): 16-28.   DOI: 10.1007/s12204-022-2452-3 Abstract （728）      PDF（pc） （1179KB）（471）       Knowledge map    Save Rehabilitative training and assistance to daily living activities play critical roles in improving the life quality of lower limb dyskinesia patients and older people with motor function degeneration. Lower limb rehabilitative exoskeleton has a promising application prospect in support of the above population. In this paper, critical technologies for developing lower limb rehabilitative exoskeleton for individualized user needs are identi- fied and reviewed, including exoskeleton hardware modularization, bionic compliant driving, individualized gait planning and individual-oriented motion intention recognition. Inspired by the idea of servitization, potentials in exoskeleton product-service system design and its enabling technologies are then discussed. It is suggested that future research will focus on exoskeleton technology and exoskeleton-based service development oriented to an individual’s physical features and personalized requirements to realize better human-exoskeleton coordination in terms of technology, as well as accessible and high-quality rehabilitation and living assistance in terms of utility. Reference | Related Articles | Metrics | Comments（0）

Select

Multi-Robot Task Allocation Using Multimodal Multi-Objective Evolutionary Algorithm Based on Deep Reinforcement Learning MIAO Zhenhua(苗镇华), HUANG Wentao(黄文焘), ZHANG Yilian(张依恋), FAN Qinqin(范勤勤) J Shanghai Jiaotong Univ Sci    2024, 29 (3): 377-387.   DOI: 10.1007/s12204-023-2679-7 Abstract （375）      PDF（pc） （975KB）（193）       Knowledge map    Save The overall performance of multi-robot collaborative systems is significantly affected by the multirobot task allocation. To improve the effectiveness, robustness, and safety of multi-robot collaborative systems,a multimodal multi-objective evolutionary algorithm based on deep reinforcement learning is proposed in this paper. The improved multimodal multi-objective evolutionary algorithm is used to solve multi-robot task allocation problems. Moreover, a deep reinforcement learning strategy is used in the last generation to provide a high-quality path for each assigned robot via an end-to-end manner. Comparisons with three popular multimodal multi-objective evolutionary algorithms on three different scenarios of multi-robot task allocation problems are carried out to verify the performance of the proposed algorithm. The experimental test results show that the proposed algorithm can generate sufficient equivalent schemes to improve the availability and robustness of multirobot collaborative systems in uncertain environments, and also produce the best scheme to improve the overall task execution efficiency of multi-robot collaborative systems. Reference | Related Articles | Metrics | Comments（0）

Select

Multi-Agent Path Planning Method Based on Improved Deep Q-Network in Dynamic Environments LI Shuyi (李舒逸), LI Minzhe (李旻哲), JING Zhongliang∗ (敬忠良) J Shanghai Jiaotong Univ Sci    2024, 29 (4): 601-612.   DOI: 10.1007/s12204-024-2732-1 Abstract （252）      PDF（pc） （1213KB）（146）       Knowledge map    Save The multi-agent path planning problem presents significant challenges in dynamic environments, primarily due to the ever-changing positions of obstacles and the complex interactions between agents’ actions. These factors contribute to a tendency for the solution to converge slowly, and in some cases, diverge altogether. In addressing this issue, this paper introduces a novel approach utilizing a double dueling deep Q-network (D3QN), tailored for dynamic multi-agent environments. A novel reward function based on multi-agent positional constraints is designed, and a training strategy based on incremental learning is performed to achieve collaborative path planning of multiple agents. Moreover, the greedy and Boltzmann probability selection policy is introduced for action selection and avoiding convergence to local extremum. To match radar and image sensors, a convolutional neural network - long short-term memory (CNN-LSTM) architecture is constructed to extract the feature of multi-source measurement as the input of the D3QN. The algorithm’s efficacy and reliability are validated in a simulated environment, utilizing robot operating system and Gazebo. The simulation results show that the proposed algorithm provides a real-time solution for path planning tasks in dynamic scenarios. In terms of the average success rate and accuracy, the proposed method is superior to other deep learning algorithms, and the convergence speed is also improved. Reference | Related Articles | Metrics | Comments（0）

Select

Review on Anti-Frost Technology Based on Microchannel Heat Exchanger YE Zhenhong(叶振鸿), WANG Wei(王炜), LI Xinhua(李新华), CHEN Jiangping(陈江平) J Shanghai Jiaotong Univ Sci    2024, 29 (2): 161-178.   DOI: 10.1007/s12204-022-2539-x Abstract （222）      PDF（pc） （4397KB）（100）       Knowledge map    Save Frosting is an inevitable adverse phenomenon in many fields such as industrial refrigeration, cryogenics, and heat pump air conditioning, which may influence the efficiency of the equipment and increase the energy consumption of the system. The complicated louvered-fin structure and fluid-channels arrangements of the microchannel heat exchanger (HEX) will affect the heat transfer performance and frosting characteristics. First, this article analyzes different factors such as refrigerant distribution, refrigerant flow pattern, and HEX surface temperature distribution. Further, combined with the features of the microchannel HEX, the existing anti-frosting technologies and various methods of surface treatment for anti-frosting are summarized. The review focuses on the preparation of superhydrophobic surfaces and their superior properties. Furthermore, the internal mechanism is analyzed in conjunction with the relevant research of our group. Superhydrophobic character has excellent anti-frosting performance and heat transfer performance, which is of great significance for improving energy-saving and system performance. Finally, the future development of superhydrophobic surface technology is analyzed and prospected. Reference | Related Articles | Metrics | Comments（0）

Select

Transfer Learning in Motor Imagery Brain Computer Interface: A Review LI Mingai1,2,3∗ (李明爱), XU Dongqin1 (许东芹) J Shanghai Jiaotong Univ Sci    2024, 29 (1): 37-59.   DOI: 10.1007/s12204-022-2488-4 Abstract （208）      PDF（pc） （1734KB）（103）       Knowledge map    Save Transfer learning, as a new machine learning methodology, may solve problems in related but different domains by using existing knowledge, and it is often applied to transfer training data from another domain for model training in the case of insufficient training data. In recent years, an increasing number of researchers who engage in brain-computer interface (BCI), have focused on using transfer learning to make most of the available electroencephalogram data from different subjects, effectively reducing the cost of expensive data acquisition and labeling as well as greatly improving the learning performance of the model. This paper surveys the development of transfer learning and reviews the transfer learning approaches in BCI. In addition, according to the “what to transfer” question in transfer learning, this review is organized into three contexts: instance-based transfer learning, parameter-based transfer learning, and feature-based transfer learning. Furthermore, the current transfer learning applications in BCI research are summarized in terms of the transfer learning methods, datasets, evaluation performance, etc. At the end of the paper, the questions to be solved in future research are put forward, laying the foundation for the popularization and in-depth research of transfer learning in BCI. Reference | Related Articles | Metrics | Comments（0）

Select

Journal's online articles can be found at SpringerLink Online publication is equivalent to paper publication which can also be indexed in the database of EI Village approximately one month after the online date. Paper publications will be printed within around one year after online publication (content cannot be changed). The link is Online First Articles in our journal: https://link.springer.com/journal/12204/online-first J Shanghai Jiaotong Univ Sci    Accepted: 21 December 2023

Select

Anti-Occlusion Object Tracking Algorithm Based on Filter Prediction CHEN Kun(陈坤), ZHAO Xu(赵旭), DONG Chunyu(董春玉), DI Zichao(邸子超), CHEN Zongzhi(陈宗枝) J Shanghai Jiaotong Univ Sci    2024, 29 (3): 400-413.   DOI: 10.1007/s12204-022-2484-8 Abstract （183）      PDF（pc） （5510KB）（73）       Knowledge map    Save Visual object tracking is an important issue that has received long-term attention in computer vision.The ability to effectively handle occlusion, especially severe occlusion, is an important aspect of evaluating theperformance of object tracking algorithms in long-term tracking, and is of great significance to improving therobustness of object tracking algorithms. However, most object tracking algorithms lack a processing mechanism specifically for occlusion. In the case of occlusion, due to the lack of target information, it is necessary to predict the target position based on the motion trajectory. Kalman filtering and particle filtering can effectively predict the target motion state based on the historical motion information. A single object tracking method, called probabilistic discriminative model prediction (PrDiMP), is based on the spatial attention mechanism in complex scenes and occlusions. In order to improve the performance of PrDiMP, Kalman filtering, particle filtering and linear filtering are introduced. First, for the occlusion situation, Kalman filtering and particle filtering are respectively introduced to predict the object position, thereby replacing the detection result of the original tracking algorithm and stopping recursion of target model. Second, for detection-jump problem of similar objects in complex scenes, a linear filtering window is added. The evaluation results on the three datasets, including GOT-10k, UAV123 and LaSOT, and the visualization results on several videos, show that our algorithms have improved tracking performance under occlusion and the detection-jump is effectively suppressed. Reference | Related Articles | Metrics | Comments（0）

Select

Online Multi-Object Tracking Under Moving Unmanned Aerial Vehicle Platform Based on Object Detection and Feature Extraction Network LIU Zengmin (刘增敏), WANG Shentao(王申涛), YAO Lixiu(姚莉秀), CAI Yunze(蔡云泽) J Shanghai Jiaotong Univ Sci    2024, 29 (3): 388-399.   DOI: 10.1007/s12204-022-2540-4 Abstract （179）      PDF（pc） （1105KB）（76）       Knowledge map    Save In order to solve the problem of small object size and low detection accuracy under the unmanned aerial vehicle (UAV) platform, the object detection algorithm based on deep aggregation network and high-resolution fusion module is studied. Furthermore, a joint network of object detection and feature extraction is studied to construct a real-time multi-object tracking algorithm. For the problem of object association failure caused by UAV movement, image registration is applied to multi-object tracking and a camera motion discrimination model is proposed to improve the speed of the multi-object tracking algorithm. The simulation results show that the algorithm proposed in this study can improve the accuracy of multi-object tracking under the UAV platform, and effectively solve the problem of association failure caused by UAV movement. Reference | Related Articles | Metrics | Comments（0）

Select

Unsupervised Oral Endoscope Image Stitching Algorithm HUANG Rong (黄荣), CHANG Qing∗ (常青), ZHANG Yang （张扬） J Shanghai Jiaotong Univ Sci    2024, 29 (1): 81-90.   DOI: 10.1007/s12204-022-2513-7 Abstract （165）      PDF（pc） （5774KB）（63）       Knowledge map    Save Oral endoscope image stitching algorithm is studied to obtain wide-field oral images through registration and stitching, which is of great significance for auxiliary diagnosis. Compared with natural images, oral images have lower textures and fewer features. However, traditional feature-based image stitching methods rely heavily on feature extraction quality, often showing an unsatisfactory performance when stitching images with few features. Moreover, due to the hand-held shooting, there are large depth and perspective disparities between the captured images, which also pose a challenge to image stitching. To overcome the above problems, we propose an unsupervised oral endoscope image stitching algorithm based on the extraction of overlapping regions and the loss of deep features. In the registration stage, we extract the overlapping region of the input images by sketching polygon intersection for feature points screening and estimate homography from coarse to fine on a three-layer feature pyramid structure. Moreover, we calculate loss using deep features instead of pixel values to emphasize the importance of depth disparities in homography estimation. Finally, we reconstruct the stitched images from feature to pixel, which can eliminate artifacts caused by large parallax. Our method is compared with both feature-based and previous deep-based methods on the UDIS-D dataset and our oral endoscopy image dataset. The experimental results show that our algorithm can achieve higher homography estimation accuracy, and better visual quality, and can be effectively applied to oral endoscope image stitching. Reference | Related Articles | Metrics | Comments（0）

Select

Time-Resolved Imaging in Short-Wave Infrared Region XU Yang (徐杨), LI Wanwan∗ (李万万) J Shanghai Jiaotong Univ Sci    2024, 29 (1): 29-36.   DOI: 10.1007/s12204-022-2547-x Abstract （152）      PDF（pc） （810KB）（52）       Knowledge map    Save Compared with the conventional first near-infrared (NIR-I, 700—900 nm) window, the short-wave infrared region (SWIR, 900—1 700 nm) possesses the merits of the increasing tissue penetration depths and the suppression of scattering background, leading to great potential for in vivo imaging. Based on the limitations of the common spectral domain, and the superiority of the time-dimension, time-resolved imaging eliminates the auto-fluorescence in the biological tissue, thus supporting higher signal-to-noise ratio and sensitivities. The imaging technique is not affected by the difference in tissue composition or thickness and has the practical value of quantitative in vivo detection. Almost all the relevant time-resolved imaging was carried out around lanthanide-doped upconversion nanomaterials, owing to the advantages of ultralong luminescence lifetime, excellent photostability, controllable morphology, easy surface modification and various strategies of regulating lifetime. Therefore, this review presents the research progress of SWIR time-resolved imaging technology based on nanomaterials doped with lanthanide ions as luminescence centers in recent years. Reference | Related Articles | Metrics | Comments（0）

Select

Performance and Optimization of Air Source Heat Pump Water Heater with Cyclic Heating LI Fan(李凡), LU Gaofeng(陆高锋), DING Yunxiao(丁云霄), ZHENG Chunyuan(郑春元), LI Bin(李斌), ZHAI Xiaoqiang(翟晓强) J Shanghai Jiaotong Univ Sci    2024, 29 (2): 179-187.   DOI: 10.1007/s12204-022-2500-z Abstract （146）      PDF（pc） （1349KB）（60）       Knowledge map    Save A new type of microchannel condenser applied in the air source heat pump water heater (ASHPWH) with cyclic heating was proposed in this study. The operating performance of the ASHPWH was first tested. Then,the structure of the microchannel condenser was optimized with the implement of vortex generators. Finally, a numerical model of the ASHPWH was established and the optimized microchannel condenser was studied. The experimental results showed that the average coefficient of performance (COP) of the 1 HP (735 W) ASHPWH reached 3.48. In addition, the optimized microchannel condenser could be matched with a 3 HP (2 430 W) ASHPWH with an average heating capacity of 10.30 kW, and achieving an average COP of 4.24, 14.6% higher than the limit value in the national standard. Reference | Related Articles | Metrics | Comments（0）

Select

A Novel Cable-Driven Soft Robot for Surgery LI Ru1 (李茹), CHEN Fang2 (陈方), YU Wenwei3 (俞文伟), IGARASH Tatsuo3,4, SHU Xiongpeng1 (舒雄鹏), XIE Le1,5,6∗ (谢叻) J Shanghai Jiaotong Univ Sci    2024, 29 (1): 60-72.   DOI: 10.1007/s12204-022-2497-3 Abstract （145）      PDF（pc） （2939KB）（81）       Knowledge map    Save Robot-assisted laparoscopic radical prostatectomy (RARP) is widely used to treat prostate cancer. The rigid instruments primarily used in RARP cannot overcome the problem of blind areas in surgery and lead to more trauma such as more incision for the passage of the instrument and additional tissue damage caused by rigid instruments. Soft robots are relatively flexible and theoretically have infinite degrees of freedom which can overcome the problem of the rigid instrument. A soft robot system for single-port transvesical robot-assisted radical prostatectomy (STvRARP) is developed in this study. The soft manipulator with 10 mm in diameter and a maximum bending angle of 270? has good flexibility and dexterity. The design and mechanical structure of the soft robot are described. The kinematics of the soft manipulator is established and the inverse kinematics is compensated based on the characteristics of the designed soft manipulator. The master-slave control system of soft robot for surgery is built and the feasibility of the designed soft robot is verified. Reference | Related Articles | Metrics | Comments（0）

Select

Psychological Impact of the 2022 Round COVID-19 Pandemic on China’s College Students HONG Dongyang1,3 (洪冬羊), WANG Jinxia2,3 (王金霞), ZHANG Hongyang2,3 (张虹洋), CAO Ziyang2,3 (曹紫阳), YAN Zijun 2,3 (晏紫君), ZOU Lin2,3∗ (邹琳) J Shanghai Jiaotong Univ Sci    2024, 29 (1): 141-149.   DOI: 10.1007/s12204-022-2557-8 Abstract （142）      PDF（pc） （194KB）（25）       Knowledge map    Save In response to the new round of COVID-19 outbreaks since March 2022, universities with high outbreak rates around the country have taken quarantine measures to contain the epidemic. Evidence from previous coronavirus outbreaks has shown that people under quarantine are at risk for mental health disorders. To better understand the impacts of this round of COVID-19 quarantine on domestic college students and their responses, we conducted a systematic survey to assess the stress and anxiety, and to evaluate effective measurements in this population. We searched relevant documents and literature, and designed a questionnaire from six aspects, including psychological status, epidemic situation, study, daily life, sports, and interpersonal communication, with 51 items in total. We sent the questionnaire on the Wenjuanxing Web platform, from April 2 to 8, 2022. We evaluated the mental status according to parts of the Generalized Anxiety Disorder-7 (GAD-7) and Depression Anxiety Stress Scales-21 (DASS-21), and investigated the influencing risk factors and countermeasures. Statistical analysis was performed by using the Chi-square test and multi-variable logistic regression. In total, 508 college respondents were recruited in our survey, and the pooled prevalence of mild anxiety (GAD score  5, or DASS-21 anxiety score 8) or stress (DASS-21 pressure score 14) caused by the new round of COVID-19 pandemic quarantine was 19.69% (100/508). The prevalence of the anxiety or stress in college students with COVID-19 quarantine between different genders, regions, and majors was not significantly different. Independent risk factors for the mild anxiety or stress of undergraduates by COVID-19 quarantine included learning efficiency or duration [OR = 1.36, 95%CI (1.14—1.62), P = 0.001], based on the combined analysis of Chi-square test analysis with multi-variable logistic regression analysis. Interestingly, the mental well-beings before COVID-19 epidemic quarantine [OR = 0.22, 95%CI (0.13—0.36), P < 0.000 1], more low-intensity exercise [OR = 0.36, 95%CI (0.15—0.87), P = 0.02, high-intensity exercise as reference], and good sleep quality [OR = 0.14, 95%CI (0.07—0.30), P < 0.000 1: OR = 0.42, 95%CI (0.30—0.59), P < 0.000 1] are protective factors for alleviating the quarantinecaused anxiety or stress in Chinese college students for this round of COVID-19 epidemic quarantine. During the round of COVID-19 epidemic quarantine in 2022, a small number of college students have mild anxiety, affected by decreased learning efficiency or duration, which could be mitigated with low-intensity exercise and good sleep quality. Reference | Related Articles | Metrics | Comments（0）

Select

Working Fluid Distribution and Charge Regulation Control in Organic Rankine Cycle YE Zhenhong(叶振鸿), WANG Wei(王炜), LI Xinhua(李新华), CHEN Jiangping(陈江平) J Shanghai Jiaotong Univ Sci    2024, 29 (2): 188-201.   DOI: 10.1007/s12204-022-2538-y Abstract （139）      PDF（pc） （1116KB）（35）       Knowledge map    Save Charge-based studies, in particular investigations of mass distribution, are still almost absent, although the efficiency of the organic Rankine cycle (ORC) has attracted a great deal of scholarly attention. This paper aims to provide a new perspective on the intrinsic relationship among the mass distribution, phase-zone distribution in the heat exchanger (HEX), charge of working fluid (WF), rotation speed of the pump (RSP), and system performance. A comprehensive ORC simulation model is presented by linking each component’s sub-models, including the independent models for HEX, pump, and expander in an object-oriented fashion. The visualization study of mass distribution of the WF in the system is investigated under different working conditions. Furthermore, the volume and mass of the gas phase, two-phase and liquid phase of WF in the HEX and their variation rules are analyzed in-depth. Finally, the strategies of charge reduction considering HEX areas and pipe sizes are investigated. The results show that the model based on the interior-point method provides high levels of accuracy and robustness. The mass ratio of the WF is concentrated in the liquid receiver, especially in the regenerator, which is 32.9% and 21.9% of the total mass, respectively. Furthermore, 2.4 kg (6.9%) WF in the system gradually migrates to the hightemperature side as the RSP increases while 6.1 kg (17.4%) WF migrates to the low-temperature side, especially to the condenser, as the charge in the system increases. Output power and efficiency both decrease gradually after the peak due to changes in RSP and charge. Last, reducing heat transfer areas of the condenser and regenerator is the most effective way to reduce WF charge. Reference | Related Articles | Metrics | Comments（0）

Select

Performance Improvement of Multiband Triangular Microstrip Patch Antenna Using Frequency Selective Surface MAHENDRAN Krishnakumar∗, GAYATHRI Rajaraman J Shanghai Jiaotong Univ Sci    2024, 29 (2): 316-321.   DOI: 10.1007/s12204-022-2492-8 Abstract （125）      PDF（pc） （798KB）（21）       Knowledge map    Save Today’s antennas have to operate in multiple resonant frequencies to satisfy the need of recent advances in communication technologies. This paper presents split ring resonator based triangular multiband antenna whose antenna performance is enhanced with the help of frequency selective surfaces (FSSs). The antenna has multiple resonances at S, C, and X bands. An array of 4 × 3 crisscross-shaped unit cells are arranged to form the FSS layer. The antenna is fed with a microstrip line feeding technique. The proposed antenna operates at 3.5 GHz, 4.1 GHz, 5.5 GHz, 9.4 GHz, and 9.8 GHz with a better return loss and gain. Simulated and measured results yield a good match. Reference | Related Articles | Metrics | Comments（0）

Select

Social Network Analysis of COVID-19 Research and the Changing International Collaboration Structure QIN Ye1 (秦野), CHEN Rongrong2∗ (陈蓉蓉) J Shanghai Jiaotong Univ Sci    2024, 29 (1): 150-160.   DOI: 10.1007/s12204-022-2558-7 Abstract （123）      PDF（pc） （3127KB）（31）       Knowledge map    Save Research in Information Science and interdisciplinary areas suggested the formation of a growing network of international research collaboration. The massive transmission of COVID-19 worldwide especially after the identification of the Omicron variant could fundamentally alter the factors shaping the network’s development. This study employs network analysis methods to analyze the structure of the COVID-19 research collaboration from 2020 to 2022, using two major academic publication databases and the VOSviewer software. A novel temporal view is added by examining the dynamic changes of the network, and a fractional counting method is adopted as methodological improvements to previous research. Analysis reveals that the COVID-19 research network structure has undergone substantial changes over time, as collaborating countries and regions form and re-form new clusters. Transformations in the network can be partly explained by key developments in the pandemic and other social-political events. China as one of the largest pivots in the network formed a relatively distinct cluster, with potential to develop a larger Asia-Pacific collaboration cluster based on its research impact. Reference | Related Articles | Metrics | Comments（0）

Select

Explosion Hazard Analysis of Liquefied Petroleum Gas Transportation GAO Sida1 (高思达),HAO Lin 1* (郝琳), ZHU Zhenxing2* (朱振兴), WEI Hongyuan1 (卫宏远) J Shanghai Jiaotong Univ Sci    2024, 29 (2): 252-260.   DOI: 10.1007/s12204-022-2536-0 Abstract （115）      PDF（pc） （1310KB）（43）       Knowledge map    Save This paper presents a quantitative risk analysis of liquefied petroleum gas (LPG) transportation. An accident that happened on June 13, 2020, on the highway near Wenling, China is studied as a case. In this accident, LPG carried by a tank truck on the highway leaked and caused a large explosion, which led to 20 deaths. Different methods are combined to calculate the consequence of the accident. Multi-energy model and rupture of vessel model are employed to calculate the overpressure; the simulation result of the multi-energy model is closer to the damage caused by the accident. The safety distances in accidents of LPG transport storage tanks of different capacities are calculated in this study; the results show that the damage of explosion will increase with the filling degree of the tank. Even though the filling degree is 90% (value required by law), the 99% fatality rate range will reach 42 m, which is higher than regulated distance between road and building. The social risk of the tank truck has also been calculated and the results show that the risk is not acceptable. The calculating method used in this study could evaluate the risk of LPG tanker more accurately, which may contribute to the establishment of transportation regulation so that losses from similar accidents in the future could be reduced. Reference | Related Articles | Metrics | Comments（0）

Select

Prediction of Pediatric Sepsis Using a Deep Encoding Network with Cross Features CHEN Xiao1,2 (陈潇), ZHANG Rui1,2 (张瑞), TANG Xinyi1,2 (汤心溢), QIAN Juan3∗ (钱娟) J Shanghai Jiaotong Univ Sci    2024, 29 (1): 131-140.   DOI: 10.1007/s12204-022-2499-1 Abstract （114）      PDF（pc） （1153KB）（33）       Knowledge map    Save Sepsis poses a serious threat to health of children in pediatric intensive care unit. The mortality from pediatric sepsis can be effectively reduced through in-time diagnosis and therapeutic intervention. The bacilliculture detection method is too time-consuming to receive timely treatment. In this research, we propose a new framework: a deep encoding network with cross features (CF-DEN) that enables accurate early detection of sepsis. Cross features are automatically constructed via the gradient boosting decision tree and distilled into the deep encoding network (DEN) we designed. The DEN is aimed at learning sufficiently effective representation from clinical test data. Each layer of the DEN filtrates the features involved in computation at current layer via attention mechanism and outputs the current prediction which is additive layer by layer to obtain the embedding feature at last layer. The framework takes the advantage of tree-based method and neural network method to extract effective representation from small clinical dataset and obtain accurate prediction in order to prompt patient to get timely treatment. We evaluate the performance of the framework on the dataset collected from Shanghai Children’s Medical Center. Compared with common machine learning methods, our method achieves the increase on F1-score by 16.06% on the test set. Reference | Related Articles | Metrics | Comments（0）

Select

Retinal Vessel Segmentation via Adversarial Learning and Iterative Refinement GU Wen (顾闻), XU Yi∗ (徐奕) J Shanghai Jiaotong Univ Sci    2024, 29 (1): 73-80.   DOI: 10.1007/s12204-022-2479-5 Abstract （110）      PDF（pc） （914KB）（40）       Knowledge map    Save Retinal vessel segmentation is a challenging medical task owing to small size of dataset, micro blood vessels and low image contrast. To address these issues, we introduce a novel convolutional neural network in this paper, which takes the advantage of both adversarial learning and recurrent neural network. An iterative design of network with recurrent unit is performed to refine the segmentation results from input retinal image gradually. Recurrent unit preserves high-level semantic information for feature reuse, so as to output a sufficiently refined segmentation map instead of a coarse mask. Moreover, an adversarial loss is imposing the integrity and connectivity constraints on the segmented vessel regions, thus greatly reducing topology errors of segmentation. The experimental results on the DRIVE dataset show that our method achieves area under curve and sensitivity of 98.17% and 80.64%, respectively. Our method achieves superior performance in retinal vessel segmentation compared with other existing state-of-the-art methods. Reference | Related Articles | Metrics | Comments（0）

Select

Ship Pipe Layout Optimization Based on Improved Particle Swarm Optimization LIN Yan1, 2(林焰), BIAN Xuanyi1(卞璇屹), DONG Zongran3(董宗然) J Shanghai Jiaotong Univ Sci    2024, 29 (5): 737-746.   DOI: 10.1007/s12204-022-2530-6 Abstract （109）      PDF（pc） （1456KB）（85）       Knowledge map    Save Ship pipe layout optimization is one of the difficulties and hot spots in ship intelligent production design. A high-dimensional vector coding is proposed based on the research of related pipe coding and ship pipe route features in this paper. The advantages of this coding method are concise structure, strong compatibility, and independence from the gridding space. Based on the proposed coding, the particle swarm optimization algorithm is implemented, and the algorithm is improved by the pre-selected path strategy and the branch-pipe processing strategy. Finally, two simulation results reveal that the proposed coding and algorithm have feasibility and engineering practicability. Reference | Related Articles | Metrics | Comments（0）

Select

Identification of Steady State and Transient State YU Sheng (于生), LI Xiangshun∗ (李向舜) J Shanghai Jiaotong Univ Sci    2024, 29 (2): 261-270.   DOI: 10.1007/s12204-022-2516-4 Abstract （109）      PDF（pc） （1612KB）（22）       Knowledge map    Save Identification of steady state and transient state plays an important role in modeling, control, optimization, and fault detection of industrial processes. Many existing methods for state identification are not satisfactory in practical applications due to problems of ideal hypothesis, too many parameters, and poor robustness. In this paper, a novel state identification approach is proposed. The problem of state identification is transformed into finding the noise band of differential signal. For practical application, automatic selection of noise band amplitude is proposed to make the method convenient to be used. Problems of gross errors, low signal-to-noise ratio and online identification are considered. And comparison with other two methods shows that the proposed method has better identification performance. Simulations and experiments also prove the effectiveness and practicability of the proposed method. Reference | Related Articles | Metrics | Comments（0）

Select

Multi-Channel Based on Attention Network for Infrared Small Target Detection ZHANG Yanjun(张彦军), WANG Biyun(王碧云)，CAI Yunze (蔡云泽) J Shanghai Jiaotong Univ Sci    2024, 29 (3): 414-427.   DOI: 10.1007/s12204-023-2616-9 Abstract （102）      PDF（pc） （1697KB）（48）       Knowledge map    Save Infrared detection technology has the advantages of all-weather detection and good concealment,which is widely used in long-distance target detection and tracking systems. However, the complex background,the strong noise, and the characteristics of small scale and weak intensity of targets bring great difficulties to the detection of infrared small targets. A multi-channel based on attention network is proposed in this paper, aimed at the problem of high missed detection rate and false alarm rate of traditional algorithms and the problem of large model, high complexity and poor detection performance of deep learning algorithms. First, given the difficulty in extracting the features of infrared multiscale and small dim targets, the multiple channels are designed based on dilated convolution to capture multiscale target features. Second, the coordinate attention block is incorporated in each channel to suppress background clutters adaptively and enhance target features. In addition, the fusion of shallow detail features and deep abstract semantic features is realized by synthesizing the contextual attention fusion block. Finally, it is verified that, compared with other state-of-the-art methods based on the datasets SIRST and MDFA, the proposed algorithm further improves the detection effect, and the model size and computational complexity are smaller. Reference | Related Articles | Metrics | Comments（0）

Select

Weighted Heterogeneous Graph-Based Incremental Automatic Disease Diagnosis Method TIAN Yuanyuan (田圆圆), JIN Yanrui (金衍瑞), LI Zhiyuan (李志远), LIU Jinlei (刘金磊), LIU Chengliang∗ (刘成良) J Shanghai Jiaotong Univ Sci    2024, 29 (1): 120-130.   DOI: 10.1007/s12204-022-2537-z Abstract （101）      PDF（pc） （1081KB）（27）       Knowledge map    Save The objective of this study is to construct a multi-department symptom-based automatic diagnosis model. However, it is difficult to establish a model to classify plenty of diseases and collect thousands of diseasesymptom datasets simultaneously. Inspired by the thought of “knowledge graph is model”, this study proposes to build an experience-infused knowledge model by continuously learning the experiential knowledge from data, and incrementally injecting it into the knowledge graph. Therefore, incremental learning and injection are used to solve the data collection problem, and the knowledge graph is modeled and containerized to solve the large-scale multi-classification problems. First, an entity linking method is designed and a heterogeneous knowledge graph is constructed by graph fusion. Then, an adaptive neural network model is constructed for each dataset, and the data is used for statistical initialization and model training. Finally, the weights and biases of the learned neural network model are updated to the knowledge graph. It is worth noting that for the incremental process, we consider both the data and class increments. We evaluate the diagnostic effectiveness of the model on the current dataset and the anti-forgetting ability on the historical dataset after class increment on three public datasets. Compared with the classical model, the proposed model improves the diagnostic accuracy of the three datasets by 5%, 2%, and 15% on average, respectively. Meanwhile, the model under incremental learning has a better ability to resist forgetting. Reference | Related Articles | Metrics | Comments（0）

Select

Federated Approach for Privacy-Preserving Traffic Prediction Using Graph Convolutional Network LONARE Savita1，2* , BHRAMARAMBA Ravi2 J Shanghai Jiaotong Univ Sci    2024, 29 (3): 509-517.   DOI: 10.1007/s12204-021-2382-5 Abstract （99）      PDF（pc） （525KB）（37）       Knowledge map    Save Existing traffic flow prediction frameworks have already achieved enormous success due to large traffic datasets and capability of deep learning models. However, data privacy and security are always a challenge in every field where data need to be uploaded to the cloud. Federated learning (FL) is an emerging trend for distributed training of data. The primary goal of FL is to train an efficient communication model without compromising data privacy. The traffic data have a robust spatio-temporal correlation, but various approaches proposed earlier have not considered spatial correlation of the traffic data. This paper presents FL-based traffic flow prediction with spatio-temporal correlation. This work uses a differential privacy (DP) scheme for privacy preservation of participant’s data. To the best of our knowledge, this is the first time that FL is used for vehicular traffic prediction while considering the spatio-temporal correlation of traffic data with DP preservation. The proposed framework trains the data locally at the client-side with DP. It then uses the model aggregation mechanism federated graph convolutional network (FedGCN) at the server-side to find the average of locally trained models. The results of the proposed work show that the FedGCN model accurately predicts the traffic. DP scheme at client-side helps clients to set a budget for privacy loss. Reference | Related Articles | Metrics | Comments（0）

Select

Active Magnetic Compensation Based on Parametric Resonance Magnetometer GUO Yang1 (郭阳), LI Shaoliang2 (李绍良), HUANG Yiming1 (黄艺明), LUO Manruo1 (骆曼箬), LIU Hua1* (刘华) J Shanghai Jiaotong Univ Sci    2024, 29 (2): 280-289.   DOI: 10.1007/s12204-022-2524-4 Abstract （98）      PDF（pc） （1976KB）（24）       Knowledge map    Save Based on the parametric resonance magnetometer (PRM) theory, this paper establishes an experimental system of PRM. The experimental results are consistent with the theoretical predictions. A PRM has been developed with sensitivity of 0.5 pT/Hz1/2, which can detect the magnitude of residual magnetic field; furthermore, a proportion-integration-differentiation (PID) closed-loop magnetic compensation system of the residual magnetic field also has been realized. Compared with open-loop compensation, the PID closed-loop compensation reduces the average value of the residual magnetic field in the z-axis direction from 0.024 4 nT to −0.002 3 nT, and the mean-square error from 0.208 3 nT to 0.069 1 nT. In the same way, the average value of the residual magnetic field in the y-axis direction is reduced from 0.081 6 nT to −0.004 2 nT, and the mean-square error from 0.131 6 nT to 0.046 1 nT. The magnitude of residual magnetic fields in both directions is decreased to the order of picotesla (pT). In addition, based on the signal waveforms of the magnetometer, a method of verifying the effect of magnetic compensation is proposed. Reference | Related Articles | Metrics | Comments（0）

Select

Numerical Investigation on Dynamic Response Characteristics of Fluid-Structure Interaction of Gas-Liquid Two-Phase Flow in Horizontal Pipe WANG Zhiwei(王志伟), HE Yanping(何炎平), LI Mingzhi（李铭志）, QIU Ming(仇明), HUANG Chao(黄超), LIU Yadong（亚东），WANG Zi(王梓) J Shanghai Jiaotong Univ Sci    2024, 29 (2): 237-244.   DOI: 10.1007/s12204-022-2469-7 Abstract （97）      PDF（pc） （1576KB）（36）       Knowledge map    Save Fluid-structure interaction (FSI) of gas-liquid two-phase flow in the horizontal pipe is investigated numerically in the present study. The volume of fluid model and standard k-ε turbulence model are integrated to simulate the typical gas-liquid two-phase flow patterns. First, validation of the numerical model is conducted and the typical flow patterns are consistent with the Baker chart. Then, the FSI framework is established to investigate the dynamic responses of the interaction between the horizontal pipe and gas-liquid two-phase flow. The results show that the dynamic response under stratified flow condition is relatively flat and the maximum pipe deformation and equivalent stress are 1.8 mm and 7.5 MPa respectively. Meanwhile, the dynamic responses induced by slug flow, wave flow and annular flow show obvious periodic fluctuations. Furthermore, the dynamic response characteristics under slug flow condition are maximum; the maximum pipe deformation and equivalent stress can reach 4 mm and 17.5 MPa, respectively. The principal direction of total deformation is different under various flow patterns. Therefore, the periodic equivalent stress will form the cyclic impact on the pipe wall and affect the fatigue life of the horizontal pipe. The present study may serve as a reference for FSI simulation under gas-liquid two-phase transport conditions. Reference | Related Articles | Metrics | Comments（0）

Select

Numerical Simulation of Radial Ultrasonic Assisted MIG Welding Arc HONG Lei1 (洪蕾), XIAO Hao1 (肖皓), YE Jia2 (叶佳), MA Guohong1* (马国红) J Shanghai Jiaotong Univ Sci    2024, 29 (2): 330-338.   DOI: 10.1007/s12204-021-2380-7 Abstract （97）      PDF（pc） （1906KB）（25）       Knowledge map    Save The numerical simulation of arc was carried out for both conventional melt inert gas (MIG) welding and ultrasonic assisted melt inert gas (U-MIG) welding. Based on the model established by Fluent, the arc shape, temperature field, and potential distribution were simulated. The study found that the shape of the arc changed when ultrasonic was added radially; the high-temperature area of the arc stretched, and the temperature peak increased. But as the current increased, the increase in temperature decreased. In addition, under the same conditions, the potential of U-MIG decreased and the pressure on the workpiece increased. To verify the accuracy of the simulation results, welding experiments under identical conditions were carried out, and a high-speed camera was used to collect dynamic pictures of the arc. The simulation results were in a favorable agreement with the experimental results, which provided a certain reference value for ultrasonic assisted arc welding. Reference | Related Articles | Metrics | Comments（0）

Select

Wind Speed Short-Term Prediction Based on Empirical Wavelet Transform, Recurrent Neural Network and Error Correction ZHU Changsheng(朱昶胜), ZHU Lina∗ (朱丽娜) J Shanghai Jiaotong Univ Sci    2024, 29 (2): 297-308.   DOI: 10.1007/s12204-022-2477-7 Abstract （94）      PDF（pc） （1282KB）（35）       Knowledge map    Save Predicting wind speed accurately is essential to ensure the stability of the wind power system and improve the utilization rate of wind energy. However, owing to the stochastic and intermittent of wind speed, predicting wind speed accurately is difficult. A new hybrid deep learning model based on empirical wavelet transform, recurrent neural network and error correction for short-term wind speed prediction is proposed in this paper. The empirical wavelet transformation is applied to decompose the original wind speed series. The long short term memory network and the Elman neural network are adopted to predict low-frequency and highfrequency wind speed sub-layers respectively to balance the calculation efficiency and prediction accuracy. The error correction strategy based on deep long short term memory network is developed to modify the prediction errors. Four actual wind speed series are utilized to verify the effectiveness of the proposed model. The empirical results indicate that the method proposed in this paper has satisfactory performance in wind speed prediction. Reference | Related Articles | Metrics | Comments（0）

Select

Tree Detection Algorithm Based on Embedded YOLO Lightweight Network LV Feng(吕峰), WANG Xinyan* (王新彦), LI Lei(李磊), JIANG Quan(江泉), YI Zhengyang(易政洋) J Shanghai Jiaotong Univ Sci    2024, 29 (3): 518-527.   DOI: 10.1007/s12204-022-2451-4 Abstract （94）      PDF（pc） （2443KB）（36）       Knowledge map    Save To avoid colliding with trees during its operation, a lawn mower robot must detect the trees. Existing tree detection methods suffer from low detection accuracy (missed detection) and the lack of a lightweight model. In this study, a dataset of trees was constructed on the basis of a real lawn environment. According to the theory of channel incremental depthwise convolution and residual suppression, the Embedded-A module is proposed, which expands the depth of the feature map twice to form a residual structure to improve the lightweight degree of the model. According to residual fusion theory, the Embedded-B module is proposed, which improves the accuracy of feature-map downsampling by depthwise convolution and pooling fusion. The Embedded YOLO object detection network is formed by stacking the embedded modules and the fusion of feature maps of different resolutions. Experimental results on the testing set show that the Embedded YOLO tree detection algorithm has 84.17% and 69.91% average precision values respectively for trunk and spherical tree, and 77.04% mean average precision value. The number of convolution parameters is 1.78 × 106, and the calculation amount is 3.85 billion float operations per second. The size of weight file is 7.11 MB, and the detection speed can reach 179 frame/s. This study provides a theoretical basis for the lightweight application of the object detection algorithm based on deep learning for lawn mower robots. Reference | Related Articles | Metrics | Comments（0）

Select

Distributed Photovoltaic Real-Time Output Estimation Based on Graph Convolutional Networks CHEN Liyue1 (陈利跃), HONG Daojian2 (洪道鉴), HE Xing3* (何星), LU Dongqi2 (卢东祁), ZHANG Qian2 (张乾), XIE Nina2 (谢妮娜), XU Yizhou2 (徐一洲), YING Huanghao2 (应煌浩) J Shanghai Jiaotong Univ Sci    2024, 29 (2): 290-296.   DOI: 10.1007/s12204-022-2522-6 Abstract （91）      PDF（pc） （1160KB）（27）       Knowledge map    Save The rapid growth of distributed photovoltaic (PV) has remarkable influence for the safe and economicoperation of power systems. In view of the wide geographical distribution and a large number of distributed PV power stations, the current situation is that it is difficult to access the current dispatch data network. According to the temporal and spatial characteristics of distributed PV, a graph convolution algorithm based on adaptive learning of adjacency matrix is proposed to estimate the real-time output of distributed PV in regional power grid. The actual case study shows that the adaptive graph convolution model gives different adjacency matrixes for different PV stations, which makes the corresponding output estimation algorithm have higher accuracy. Reference | Related Articles | Metrics | Comments（0）

Select

Ensemble Attention Guided Multi-SEANet Trained with Curriculum Learning for Noninvasive Prediction of Gleason Grade Groups from MRI SHEN Ao1,2‡ (沈傲), HU Jisu 2,3‡ (胡冀苏), JIN Pengfei4 (金鹏飞), ZHOU Zhiyong2 (周志勇), QIAN Xusheng 2,3 (钱旭升), ZHENG Yi2 (郑毅), BAO Jie 4 (包婕), WANG Ximing4∗ (王希明), DAI Yakang1,2∗ (戴亚康) J Shanghai Jiaotong Univ Sci    2024, 29 (1): 109-119.   DOI: 10.1007/s12204-022-2502-x Abstract （87）      PDF（pc） （1407KB）（25）       Knowledge map    Save The Gleason grade group (GG) is an important basis for assessing the malignancy of prostate cancer, but it requires invasive biopsy to obtain pathology. To noninvasively evaluate GG, an automatic prediction method is proposed based on multi-scale convolutional neural network of the ensemble attention module trained with curriculum learning. First, a lesion-attention map based on the image of the region of interest is proposed in combination with the bottleneck attention module to make the network more focus on the lesion area. Second, the feature pyramid network is combined to make the network better learn the multi-scale information of the lesion area. Finally, in the network training, a curriculum based on the consistency gap between the visual evaluation and the pathological grade is proposed, which further improves the prediction performance of the network. Experimental results show that the proposed method is better than the traditional network model in predicting GG performance. The quadratic weighted Kappa is 0.471 1 and the positive predictive value for predicting clinically significant cancer is 0.936 9. Reference | Related Articles | Metrics | Comments（0）

Select

Pressure Pulse Response of High Temperature Molten Salt Check Valve Hit by Crystal Particles LI Shuxun (李树勋), SHEN Hengyun* (沈珩云), LIU Bincai (刘斌才),HU Yinggang (胡迎港), MA Tingqian (马廷前) J Shanghai Jiaotong Univ Sci    2024, 29 (2): 271-279.   DOI: 10.1007/s12204-023-2601-3 Abstract （86）      PDF（pc） （2060KB）（38）       Knowledge map    Save In view of the problem that crystalline particles cause wall vibration at a low temperature, based on two-phase flow model, computational fluid dynamics is used to conduct the numerical simulation of internal flows when the valve openings are 20%, 60% and 100% respectively. The molten salt flow may be changed under strict conditions and produce forced vibration of the inner parts of molten salt particle shock valve body. Euler two-phase flow model is used for different molten salt sizes to extract temporal pressure pulse information and conduct statistical data processing analysis. The influence of the molten salt crystallization of molten salt particles on the flow and pressure pulse strength is analyzed. The results show that the crystallization of molten salt has a serious impact on the vibration of the valve body, especially in the throttle rate. The valve oscillation caused by the pressure pulsation mostly occurs from the small opening rate. As the opening increases, the pressure pulse threshold and its change trend decrease. Reference | Related Articles | Metrics | Comments（0）

Select

Event-Triggered Fixed-Time Consensus of Second-Order Nonlinear Multi-Agent Systems with Delay and Switching Topologies XING Youjing1 (邢优靖), GAO Jinfeng1∗ (高金凤), LIU Xiaoping1, 2 (刘小平), WU Ping1 (吴平) J Shanghai Jiaotong Univ Sci    2024, 29 (4): 625-639.   DOI: 10.1007/s12204-024-2695-2 Abstract （86）      PDF（pc） （1059KB）（52）       Knowledge map    Save To address fixed-time consensus problems of a class of leader-follower second-order nonlinear multiagent systems with uncertain external disturbances, the event-triggered fixed-time consensus protocol is proposed. First, the virtual velocity is designed based on the backstepping control method to achieve the system consensus and the bound on convergence time only depending on the system parameters. Second, an event-triggered mechanism is presented to solve the problem of frequent communication between agents, and triggered condition based on state information is given for each follower. It is available to save communication resources, and the Zeno behaviors are excluded. Then, the delay and switching topologies of the system are also discussed. Next, the system stabilization is analyzed by Lyapunov stability theory. Finally, simulation results demonstrate the validity of the presented method. Reference | Related Articles | Metrics | Comments（0）

Select

Numerical Study on Effect of Suction Slot Geometric Parameters on Airflow Field in Compact Spinning LIN Huiting1,2 (杨娜), WANG Jun1∗ (张淑霞), ZHANG Yongfa3 (白牡丹) J Shanghai Jiaotong Univ Sci    2024, 29 (2): 245-251.   DOI: 10.1007/s12204-022-2521-7 Abstract （84）      PDF（pc） （1054KB）（26）       Knowledge map    Save The airflow field in the condensing zone is crucial as it affects the fiber condensing, additional twists, and consequently yarn properties. Parameters of spinning and suction slot geometric were found to be key factors influencing the airflow characteristics. To develop a better understanding of the complex airflow field within the pneumatic compact spinning system with lattice apron, a 3D numerical simulation model was built and the influence of negative pressure and geometric of suction slot was investigated. The results reveal that the accelerating air from the top of the suction slot generates transverse condensing force and downward pressure on the fiber strand. The inclination angle has a small effect on airflow velocity. The absolute z-velocity and x-velocity in the positive x-axis were both increased with increasing the slot width from 1.0 mm to 1.5 mm. An arc suction slot increased the absolute z-velocity and x-velocity compared with a linear one, thus benefiting fiber condensing. By decreasing the outlet negative pressure to −3 kPa, the airflow velocity increased significantly. Reference | Related Articles | Metrics | Comments（0）

Select

Fast Four-Stage Local Motion Planning Method for Mobile Robot HUANG Shan(黄山), HUANG Hongzhong(黄洪钟), ZENG Qi(曾奇) J Shanghai Jiaotong Univ Sci    2024, 29 (3): 428-435.   DOI: 10.1007/s12204-022-2423-8 Abstract （79）      PDF（pc） （1810KB）（32）       Knowledge map    Save Mobile robot local motion planning is responsible for the fast and smooth obstacle avoidance, which is one of the main indicators for evaluating mobile robots’ navigation capabilities. Current methods formulate local motion planning as a unified problem; therefore it cannot satisfy the real-time requirement on the platform with limited computing ability. In order to solve this problem, this paper proposes a fast local motion planning method that can reach a planning frequency of 500 Hz on a low-cost CPU. The proposed method decouples the local motion planning as the front-end path searching and the back-end optimization. The front-end is composed of the environment topology analysis and graph searching. The back-end includes dynamically feasible trajectory generation and optimal trajectory selection. Different from the popular methods, the proposed method decomposes the local motion planning into four sub-modules, each of which aims to solve one problem. Combining four submodules, the proposed method can obtain the complete local motion planning algorithm which can fast generate a smooth and collision-free trajectory. The experimental results demonstrate that the proposed method has the ability to obtain the smooth, dynamically feasible and collision-free trajectory and the speed of the planning is fast. Reference | Related Articles | Metrics | Comments（0）

Select

Fault-Tolerant Dynamical Consensus of Double-Integrator Multi-Agent Systems in the Presence of Asynchronous Self-Sensing Function Failures WU Zhihai∗ (吴治海), XIE Linbo (谢林柏) J Shanghai Jiaotong Univ Sci    2024, 29 (4): 613-624.   DOI: 10.1007/s12204-024-2716-1 Abstract （78）      PDF（pc） （540KB）（58）       Knowledge map    Save Double-integrator multi-agent systems (MASs) might not achieve dynamical consensus, even if only partial agents suffer from self-sensing function failures (SSFFs). SSFFs might be asynchronous in real engineering application. The existing fault-tolerant dynamical consensus protocol suitable for synchronous SSFFs cannot be directly used to tackle fault-tolerant dynamical consensus of double-integrator MASs with partial agents subject to asynchronous SSFFs. Motivated by these facts, this paper explores a new fault-tolerant dynamical consensus protocol suitable for asynchronous SSFFs. First, multi-hop communication together with the idea of treating asynchronous SSFFs as multiple piecewise synchronous SSFFs is used for recovering the connectivity of network topology among all normal agents. Second, a fault-tolerant dynamical consensus protocol is designed for doubleintegrator MASs by utilizing the history information of an agent subject to SSFF for computing its own state information at the instants when its minimum-hop normal neighbor set changes. Then, it is theoretically proved that if the strategy of network topology connectivity recovery and the fault-tolerant dynamical consensus protocol with proper time-varying gains are used simultaneously, double-integrator MASs with all normal agents and all agents subject to SSFFs can reach dynamical consensus. Finally, comparison numerical simulations are given to illustrate the effectiveness of the theoretical results. Reference | Related Articles | Metrics | Comments（0）

Select

Multi-Objective Loosely Synchronized Search for Multi-Objective Multi-Agent Path Finding with Asynchronous Actions DU Haikuo1,2 (杜海阔), GUO Zhengyu3，4(郭正玉), ZHANG Lulu1,2(章露露), CAI Yunze1,2∗ (蔡云泽) J Shanghai Jiaotong Univ Sci    2024, 29 (4): 667-677.   DOI: 10.1007/s12204-024-2744-x Abstract （77）      PDF（pc） （1177KB）（38）       Knowledge map    Save In recent years, the path planning for multi-agent technology has gradually matured, and has made breakthrough progress. The main difficulties in path planning for multi-agent are large state space, long algorithm running time, multiple optimization objectives, and asynchronous action of multiple agents. To solve the above problems, this paper first introduces the main problem of the research: multi-objective multi-agent path finding with asynchronous action, and proposes the algorithm framework of multi-objective loose synchronous (MOLS) search. By combining A∗ and M∗, MO LS-A∗ and MO-LS-M∗ algorithms are respectively proposed. The completeness and optimality of the algorithm are proved, and a series of comparative experiments are designed to analyze the factors affecting the performance of the algorithm, verifying that the proposed MO-LS-M∗ algorithm has certain advantages. Reference | Related Articles | Metrics | Comments（0）

Select

Toughening Mechanism of Large Heat Input Weld Metal for Marine Engineering Extra-Thick Plate LENG Junjie1 (冷俊杰), DI Xinjie,2*1 (邸新杰), LI Chengning1,2 (利成宁), CHENG Shanghua3 (程尚华) J Shanghai Jiaotong Univ Sci    2024, 29 (2): 349-360.   DOI: 10.1007/s12204-023-2638-3 Abstract （76）      PDF（pc） （4684KB）（32）       Knowledge map    Save In order to study the latest designed large heat input welding material of marine engineering extrathick plate, EH36 steel was joined by using twin-wire submerged arc welding with heat inputs of 85, 100 and 115 kJ/cm separately. Meanwhile, the microstructure and mechanical properties were evaluated to explore the toughening mechanism of weld metal. Results show that a lot of active inclusions are obtained in the weld metal due to the design idea of low carbon and oxide metallurgy, which contributes to the generation of numerous fine and interlocking acicular ferrite. The acicular ferrite volume ratio of weld metal exceeds 60%. Moreover, the impact energy at −40 ◦C surpasses 115 J and the crack tip opening displacement value at −10 ◦C is more than 0.2 mm under three heat inputs owing to the role of acicular ferrite, of which 85 kJ/cm is the best. The martensiteaustenite constituents are minor in size and the microstructure of the weld metal in reheated zone is dominated by small massive equiaxed ferrite, without impairing the toughness. As the heat input increases, the content of acicular ferrite drops and then rises; the impact toughness and fracture toughness first worsen consequently and then stabilize on account of the dramatic expansion of the proeutectoid ferrite size. Reference | Related Articles | Metrics | Comments（0）

Select

New Lite YOLOv4-Tiny Algorithm and Application on Crack Intelligent Detection SONG Liboa (宋立博), FEI Yanqiongb (费燕琼) J Shanghai Jiaotong Univ Sci    2024, 29 (3): 528-536.   DOI: 10.1007/s12204-022-2504-8 Abstract （76）      PDF（pc） （1358KB）（26）       Knowledge map    Save Conforming to the rapidly increasing market demand of crack detection for tall buildings, the idea of integrating deep network technology into wall climbing robot for crack detection is put forward in this paper. Taking the dependence and hardware requirements when deployed on such edge devices as Raspberry Pi into consideration, the Darknet neural network is selected as the basic framework for detection. In order to improve the inference efficiency on edge devices and avoid the possible premature over-fitting of deep networks, the lite YOLOv4-tiny algorithm is then improved from the original YOLOv4-tiny algorithm and its structure is illustrated using Netron accordingly. The images downloaded from Internet and taken from the buildings in campus are processed to form crack detection data sets, which are trained on personal computer with the AlexeyAB version of Darknet to generate weight files. Meanwhile, the AlexeyAB version of Darknet accelerated by NNpack package is deployed on Raspberry Pi 4B, and the crack detection experiments are carried out. Some characteristics, e.g., fast speed and lower false detection rate of the lite YOLOv4-tiny algorithm, are confirmed by comparison with those of original YOLOv4-tiny algorithm. The innovations of this paper focus on the simple network structure, fewer network layers, and earlier forward transmission of features to prevent over-fitting, showing the new lite neural network exceeds the original YOLOv4-tiny network significantly. Reference | Related Articles | Metrics | Comments（0）