This paper presents an improved support vector machine （SVM） algorithm, which employs invariant moments-based edge extraction to obtain feature attribute. A heuristic attribute reduction algorithm based on rough set...This paper presents an improved support vector machine （SVM） algorithm, which employs invariant moments-based edge extraction to obtain feature attribute. A heuristic attribute reduction algorithm based on rough set＇s discernible matrix is proposed to identify and classify micro-targets. To avoid the complicated calibration for intrinsic parameters of camera, an improved Broyden＇s method is proposed to estimate the image Jacobian matrix which employs Chebyshev polynomial to construct a cost function to approximate the optimization value. Finally, a visual controller is designed for a robotic micromanipulation system. The experiment results of micro-parts assembly show that the proposed methods and algorithms are effective and feasible.展开更多
Kinematic semantics is often an important content of a CAD model(it refers to a single part/solid model in this work)in many applications,but it is usually not the belonging of the model,especially for the one retriev...Kinematic semantics is often an important content of a CAD model(it refers to a single part/solid model in this work)in many applications,but it is usually not the belonging of the model,especially for the one retrieved from a common database.Especially,the effective and automatic method to reconstruct the above information for a CAD model is still rare.To address this issue,this paper proposes a smart approach to identify each assembly interface on every CAD model since the assembly interface is the fundamental but key element of reconstructing kinematic semantics.First,as the geometry of an assembly interface is formed by one or more adjacent faces on each model,a face-attributed adjacency graph integrated with face structure fingerprint is proposed.This can describe each CAD model as well as its assembly interfaces uniformly.After that,aided by the above descriptor,an improved graph attention network is developed based on a new dual-level anti-interference filtering mechanism,which makes it have the great potential to identify all representative kinds of assembly interface faces with high accuracy that have various geometric shapes but consistent kinematic semantics.Moreover,based on the abovementioned graph and face-adjacent relationships,each assembly interface on a model can be identified.Finally,experiments on representative CAD models are implemented to verify the effectiveness and characteristics of the proposed approach.The results show that the average assembly-interface-face-identification accuracy of the proposed approach can reach 91.75%,which is about 2%–5%higher than those of the recent-representative graph neural networks.Besides,compared with the state-of-the-art methods,our approach is more suitable to identify the assembly interfaces(with various shapes)for each individual CAD model that has typical kinematic pairs.展开更多
An automotive body is composed of compliant sheet metal parts.Fast and exactly diagnosing variation sources is very important when assembly variations happen.This paper proposes a diagnosis method of multi fixture var...An automotive body is composed of compliant sheet metal parts.Fast and exactly diagnosing variation sources is very important when assembly variations happen.This paper proposes a diagnosis method of multi fixture variations based on the variation model of compliant sheet metal assembly.The assembly variation model is obtained by using the method of influence coefficients(MIC) and considering the manufacturing variations of compliant parts and multi fixture variations.The measurement point variations induced by part manufacturing variations are firstly removed from the measurement data.The variation patterns of multi fixture variations are constructed by column vectors of fixture variation sensitivity matrix.This method is proved to be feasible for exactly diagnosing the fixture variations and has higher diagnosis efficiency than designated component analysis(DCA).展开更多
The rapid development of high-throughput sequencing technologies has led to a dramatic decrease in the money and time required for de novo genome sequencing or genome resequencing projects, with new genome sequences c...The rapid development of high-throughput sequencing technologies has led to a dramatic decrease in the money and time required for de novo genome sequencing or genome resequencing projects, with new genome sequences constantly released every week. Among such projects, the plethora of updated genome assemblies induces the requirement of versiondependent annotation files and other compatible public dataset for downstream analysis. To handlethese tasks in an efficient manner, we developed the reference-based genome assembly and annotation tool（RGAAT）, a flexible toolkit for resequencing-based consensus building and annotation update. RGAAT can detect sequence variants with comparable precision, specificity, and sensitivity to GATK and with higher precision and specificity than Freebayes and SAMtools on four DNAseq datasets tested in this study. RGAAT can also identify sequence variants based on cross-cultivar or cross-version genomic alignments. Unlike GATK and SAMtools/BCFtools, RGAAT builds the consensus sequence by taking into account the true allele frequency. Finally, RGAAT generates a coordinate conversion file between the reference and query genomes using sequence variants and supports annotation file transfer. Compared to the rapid annotation transfer tool（RATT）,RGAAT displays better performance characteristics for annotation transfer between different genome assemblies, strains, and species. In addition, RGAAT can be used for genome modification,genome comparison, and coordinate conversion. RGAAT is available at https://sourceforge.net/projects/rgaat/and https://github.com/wushyer/RGAAT;2 at no cost.展开更多
Multi-laser powder bed fusion(ML-PBF)adopts multiple laser-scanner systems to increase the build envelope and build speed,but its calibration is an iterative and time-consuming process.In particular,multiple large-sca...Multi-laser powder bed fusion(ML-PBF)adopts multiple laser-scanner systems to increase the build envelope and build speed,but its calibration is an iterative and time-consuming process.In particular,multiple large-scale scan fields have a complex distortion in the overlap area,challenging the calibration process.In this study,owing to the enormous workload and alignment problems in the calibration of multiple scan fields,a novel calibration system is designed in this study to realize in situ auto-detection of numerous laser spots in the build chamber to ensure high efficiency and accuracy.Moreover,because the detectable area could not cover the entire build area and the detection data still contained errors,a virtual laser-scanner system was established by identifying the assembly defects and galvo nonlinearities of the ML-PBF system from the detection data,which served as the system's controller to improve calibration accuracy.The multi-field alignment error was less than 0.012%,which could avoid the intersection and separation of scan paths in multi-laser scanning and therefore meet the requirements for high-precision ML-PBF.Finally,the reliability of the method was verified theoretically using principal component analysis.展开更多
基金supported by National Natural Science Foundation of China (No.60873032)National High Technology Research and Development Program of China (863 Program) (No.2008AA8041302)
文摘This paper presents an improved support vector machine （SVM） algorithm, which employs invariant moments-based edge extraction to obtain feature attribute. A heuristic attribute reduction algorithm based on rough set＇s discernible matrix is proposed to identify and classify micro-targets. To avoid the complicated calibration for intrinsic parameters of camera, an improved Broyden＇s method is proposed to estimate the image Jacobian matrix which employs Chebyshev polynomial to construct a cost function to approximate the optimization value. Finally, a visual controller is designed for a robotic micromanipulation system. The experiment results of micro-parts assembly show that the proposed methods and algorithms are effective and feasible.
基金supported by the National Natural Science Foundation of China[61702147]the Zhejiang Provincial Science and Technology Program in China[2021C03137].
文摘Kinematic semantics is often an important content of a CAD model(it refers to a single part/solid model in this work)in many applications,but it is usually not the belonging of the model,especially for the one retrieved from a common database.Especially,the effective and automatic method to reconstruct the above information for a CAD model is still rare.To address this issue,this paper proposes a smart approach to identify each assembly interface on every CAD model since the assembly interface is the fundamental but key element of reconstructing kinematic semantics.First,as the geometry of an assembly interface is formed by one or more adjacent faces on each model,a face-attributed adjacency graph integrated with face structure fingerprint is proposed.This can describe each CAD model as well as its assembly interfaces uniformly.After that,aided by the above descriptor,an improved graph attention network is developed based on a new dual-level anti-interference filtering mechanism,which makes it have the great potential to identify all representative kinds of assembly interface faces with high accuracy that have various geometric shapes but consistent kinematic semantics.Moreover,based on the abovementioned graph and face-adjacent relationships,each assembly interface on a model can be identified.Finally,experiments on representative CAD models are implemented to verify the effectiveness and characteristics of the proposed approach.The results show that the average assembly-interface-face-identification accuracy of the proposed approach can reach 91.75%,which is about 2%–5%higher than those of the recent-representative graph neural networks.Besides,compared with the state-of-the-art methods,our approach is more suitable to identify the assembly interfaces(with various shapes)for each individual CAD model that has typical kinematic pairs.
基金the National Natural Science Foundation of China (No. 50705056)the National High Technology Research and Development Program (863) of China (No.2006AA04Z148)
文摘An automotive body is composed of compliant sheet metal parts.Fast and exactly diagnosing variation sources is very important when assembly variations happen.This paper proposes a diagnosis method of multi fixture variations based on the variation model of compliant sheet metal assembly.The assembly variation model is obtained by using the method of influence coefficients(MIC) and considering the manufacturing variations of compliant parts and multi fixture variations.The measurement point variations induced by part manufacturing variations are firstly removed from the measurement data.The variation patterns of multi fixture variations are constructed by column vectors of fixture variation sensitivity matrix.This method is proved to be feasible for exactly diagnosing the fixture variations and has higher diagnosis efficiency than designated component analysis(DCA).
基金supported by grants from the Strategic Priority Research Program of the Chinese Academy of Sciences (Grant No. XDA08020102)National Natural Science Foundation of China (Grant Nos. 81701071, 31501042, 31271385, and 31200957)+2 种基金Shenzhen Science and Technology Program (Grant No. JCYJ20170306171013613),ChinaKing Abdulaziz City for Science and Technology (KACSTGrant No. 1035-35),Kingdom of Saudi Arabia
文摘The rapid development of high-throughput sequencing technologies has led to a dramatic decrease in the money and time required for de novo genome sequencing or genome resequencing projects, with new genome sequences constantly released every week. Among such projects, the plethora of updated genome assemblies induces the requirement of versiondependent annotation files and other compatible public dataset for downstream analysis. To handlethese tasks in an efficient manner, we developed the reference-based genome assembly and annotation tool（RGAAT）, a flexible toolkit for resequencing-based consensus building and annotation update. RGAAT can detect sequence variants with comparable precision, specificity, and sensitivity to GATK and with higher precision and specificity than Freebayes and SAMtools on four DNAseq datasets tested in this study. RGAAT can also identify sequence variants based on cross-cultivar or cross-version genomic alignments. Unlike GATK and SAMtools/BCFtools, RGAAT builds the consensus sequence by taking into account the true allele frequency. Finally, RGAAT generates a coordinate conversion file between the reference and query genomes using sequence variants and supports annotation file transfer. Compared to the rapid annotation transfer tool（RATT）,RGAAT displays better performance characteristics for annotation transfer between different genome assemblies, strains, and species. In addition, RGAAT can be used for genome modification,genome comparison, and coordinate conversion. RGAAT is available at https://sourceforge.net/projects/rgaat/and https://github.com/wushyer/RGAAT;2 at no cost.
基金This study was supported by the National High Technology Research and Development Program of China(863 Program)(Grant No.2015AA042503)the K.C.Wong Education Foundation.
文摘Multi-laser powder bed fusion(ML-PBF)adopts multiple laser-scanner systems to increase the build envelope and build speed,but its calibration is an iterative and time-consuming process.In particular,multiple large-scale scan fields have a complex distortion in the overlap area,challenging the calibration process.In this study,owing to the enormous workload and alignment problems in the calibration of multiple scan fields,a novel calibration system is designed in this study to realize in situ auto-detection of numerous laser spots in the build chamber to ensure high efficiency and accuracy.Moreover,because the detectable area could not cover the entire build area and the detection data still contained errors,a virtual laser-scanner system was established by identifying the assembly defects and galvo nonlinearities of the ML-PBF system from the detection data,which served as the system's controller to improve calibration accuracy.The multi-field alignment error was less than 0.012%,which could avoid the intersection and separation of scan paths in multi-laser scanning and therefore meet the requirements for high-precision ML-PBF.Finally,the reliability of the method was verified theoretically using principal component analysis.
