Nanjing University of Aeronautics and Astronautics

Huawei Technologies Co., Limited

Telefonaktiebolaget Lm Ericsson (publ)

Nanjing University

Shenzhen Yinwang Intelligent Technologies Co., Ltd.

Nanjing University Yancheng Environmental Protection Technology and Engineering Research Institute

Nanjing University 5d Technology Co., Ltd.

Jiangsu Green River Environmental Technology Co., Ltd.

Est Water and Technologies Co., Ltd

Jun Wang

Zhongde Shan

Dawei Li

Anyi Huang

Xixian Chen

Qian Xie

Yuanpeng Liu

Zikuan Li

Yuan Zhang

Chao Li

James Jianfeng Weng

Haoqin Yang

Yingde Liu

Yaoyao Wang

Cheng Yi

Yuxiang Wu

Tianchi Zhong

Kun Xiao

Rong Li

Huazi Zhang

Shengchen Dai

Aimin Li

Xiang Li

Jiajie Tong

Xianbin Wang

Chendong Shuang

Wei Li

Li-Te Kuo

Lin Wang

Feng Yan

Xun Cao

Honghua Chen

Peng Chen

Haomin Li

Yeli Jiang

Xi Zhu

Haibo Li

Xinyao Jin

Zeyong Wei

Chao Zhang

Zhaolian Zhu

Zengping Xing

Weisong Pan

Xun Chen

Zhijian Yao

Changming Wang

Yiming Zhang

Hao Tang

Yabin Xu

Fuzheng Chen

Hongli Ji

Jinhao Qiu

Zhoutao Wang

Chongcong Tao

Li Dai

Jianping Yang

Zhengshui Kang

Hangbin Zeng

Zhaomin Wang

Qifan Tu

Kaijun Zhang

Shiyan Hua

Shuyi Jia

Hao Yan

Xiaokang Ma

Zhengyuan Wei

Jiajia Dai

Begood Chen

Guidong Liu

Shan Li

Yu Huang

Xiaowei Sun

Bojun Wang

Haiou Song

Gang Tao

The present disclosure provides a method for analyzing a minor defect based on a progressive segmentation network, including: acquiring an original image for a surface of a component, and cropping the original image into a plurality of patches; inputting each of the patches to a minor defect feature extraction network to extract an image feature; classifying the patch into a defective image or a non-defective background image according to an extracted image feature; inputting an extracted image feature of the defective image to a defect segmentation network to obtain a segmentation mask image of a corresponding defect; and quantitatively analyzing the defect according to the segmentation mask image to obtain information such as an area, a length and a width of the defect.

Method for analyzing minor defect based on progressive segmentation network

The present disclosure disclose a method for feature detection of complex defects based on multimodal data, including feature extraction of multimodal data, multimodal feature cross-guided learning, multimodal feature fusion, and defect classification and regression. Feature extraction networks for multimodal two-dimensional data are constructed first, and a defect data set is sent to the networks for training; during training, cross-guided learning is implemented by using a multimodal feature cross-guidance network; then feature fusion is performed by using a weight adaptive method; and finally a defect detection task is implemented by using a classification subnetwork and a regression subnetwork. In the present disclosure, fusion of the multimodal data in a process of feature detection of the complex defects can be implemented efficiently, a capability of detecting the complex defects in an industrial environment can be improved more effectively, and production efficiency in an industrial manufacturing process is ensured.

Method for feature detection of complex defects based on multimodal data

The present disclosure relates to a method and system for detecting a structural defect in additive manufacturing. The method includes: layering a three-dimensional model of an additive manufacturing test piece to obtain a two-dimensional contour of an interface of each layer, and generating a machining path; arranging a non-contact sensor at a fixed measuring point of the additive manufacturing test piece, and acquiring an ultrasonic signal at each machining point when a pulse laser conducts machining point by point along the machining path; forming a visual ultrasonic field based on all the ultrasonic signals, and determining ultrasonic field data; determining, based on the ultrasonic field data, a curve of a peak of an incident wave changing with the machining path; and determining whether a machining defect exists at the machining points based on the curve of the peak of the incident wave changing with the machining path.

Method and system for detecting structural defect in additive manufacturing

The present disclosure provides a point cloud denoising method based on multi-level attention perception, including the following steps: constructing a data set of point cloud denoising; constructing a point cloud denoising neural network, including a patch feature encoder, a global level perception module, a global level attention module, and a multi-offset decoder module, and training a network model by using the data set of point cloud denoising; for input point cloud, separately obtaining a neighborhood patch of a point of each original data point, and inputting coordinates of each data point in the neighborhood patch of a point to a trained denoising neural network to obtain a location offset of each original point; and separately adjusting, based on the obtained location offset, a location corresponding to each original data point in the input point cloud, to complete point cloud denoising.

Point cloud denoising method based on multi-level attention perception

A method for measuring a distance between trace lines of a 3D braided material, including: (S) establishing a vision data acquisition system using a vision sensor; (S) acquiring, by the vision data acquisition system, a training data of the trace lines of the 3D braided material; (S) constructing a deep learning model for recognizing the trace lines of the 3D braided material; and inputting the training data acquired in step (S) to the deep learning model to obtain a trained deep learning model; and (S) positioning a location of the trace lines of the 3D braided material in batch images according to the trained deep learning model obtained in step (S); and measuring a distance between adjacent trace lines.

Method for measuring distance between trace lines of three-dimensional braided material

This application relates to a drive assembly, a motor, and a terminal. The drive assembly includes a stator and a rotor. The stator includes an excitation part, a vibration part, a first fixed part, and a pushing part. The vibration part is connected to the excitation part. The first fixed part is connected to the vibration part, and the vibration part is located between the excitation part and the first fixed part along a first direction L. The pushing part is connected to the vibration part and the rotor. The excitation part is capable of vibrating, and the excitation part is capable of driving the vibration part to act. Under limiting by the first fixed part, the vibration part is capable of vibrating at least along the first direction L and a second direction W.

Drive assembly, motor, and terminal

Establishing, by a mobile terminal, a near field communication link to a vehicle, where the vehicle corresponds to the in-vehicle system; negotiating a login binding code with the vehicle through the near field communication link; and providing the login binding code to a network side device, where the login binding code is used by the network side device to verify whether the in-vehicle system can be logged in to with the first user account.

Method and related apparatus for logging in to in-vehicle system

A device for automatically detecting a through-hole rate of a honeycomb sandwich composite-based acoustic liner, including a customized tooling, a data acquisition system, a motion mechanism and a data processing system. The data acquisition system is configured to acquire a surface three-dimensional (3D) point cloud data of an acoustic liner using a two-dimensional (2D) laser profile sensor in a manner of parallel movement shooting, and connected with a graphics workstation. The motion mechanism includes an industrial robot, and the 2D laser profile sensor is fixed at an end of the industrial robot. The motion mechanism is configured to support the data acquisition system to perform translational scanning. The data processing system includes the graphics workstation, and plays a role of path planning and data storage. A method for automatically detecting a through-hole rate of a honeycomb sandwich composite-based acoustic liner is also provided.

Device and method for automatically detecting through-hole rate of honeycomb sandwich composite-based acoustic liner

An encoding method includes obtaining to-be-encoded information and a mother code length N. The to-be-encoded information includes K information bits. The method also includes determining, based on K and N, a set I corresponding to subchannels of the information bits and a set F corresponding to subchannels of frozen bits. Information bits corresponding to subchannel sequence numbers in the set I are distributed in X outer component codes, a code length of each outer component code is B, and the X outer component codes includes a first-type outer component code and a second-type outer component code or the X outer component codes include a first-type outer component code, a second-type outer component code, and one third-type outer component code. Different types of component codes have different code rates. The method additionally includes performing polarization encoding based on the set I and the set F.

Polarization encoding method and apparatus

A method for generating a scanning path of a machining feature surface of an aircraft panel, including: acquiring a main direction and a triangular mesh model of the aircraft panel; dividing the triangular mesh model into multiple regions; recognizing the machining feature surface according to the main direction; projecting the machining feature surface to a 2D coordinate system thereof; extracting a 2D scanning path for the machining feature surface; and mapping the 2D scanning path to a 3D space to generate a 3D scanning path.

Method for generating scanning path of machining feature surface of aircraft panel

A method in a first network node, which comprises a primary Radio Link Control (RLC) entity is provided. The method comprises: receiving packets from a PDCP entity and sending packets to a PDCP entity; separating the received packets into at least a first group and a second group; assigning sequence numbers to the packets of the first group from a sequence number range allocated to at least one the secondary RLC entity in a second network node, the at least one secondary RLC connected to the primary RLC entity through a RLC channel; sending the packets of the first group with the assigned sequence numbers to the at least one secondary RLC entity.

Network nodes and methods supporting multiple connectivity

A feature-guided scanning trajectory optimization method for a 3D measurement robot, including: building a 3D digital model of an aircraft surface; obtaining a size of the 3D digital model; extracting features to be measured; classifying the features to be measured; calculating a geometric parameter of each type of features to be measured; generating an initial scanning trajectory of each type of features to be measured; building a constraint model of the 3D measurement robot; optimizing the initial scanning trajectory into a local optimal scanning trajectory; and planning a global optimal scanning trajectory of each type of features to be measured on the aircraft surface by using a modified ant colony optimization algorithm.

Feature-guided scanning trajectory optimization method for three-dimensional measurement robot

A method for analyzing fuselage profile based on measurement data of an aircraft, including: acquiring point-cloud data of an aircraft via a laser scanner; selecting point-cloud data of a fuselage component from the point-cloud data of the aircraft; based on a weighted locally optimal projection (WLOP) operator and Lmedian curve-skeleton concept of point cloud, extracting a medial axis from the point-cloud data of the fuselage component; uniformly sampling the medial axis into a plurality of skeleton points; extracting a discrete point set of a cross-section contour of the fuselage component; performing circle fitting on the discrete point set to obtain a fitted circle and parameters thereof; calculating a deformation displacement measurement indicator μ of the cross-section of the fuselage component to evaluate cross-section contour of the fuselage.

Method for analyzing fuselage profile based on measurement data of whole aircraft

A method in a User Equipment (UE) for activation or deactivation of a carrier in a communication network supporting carrier aggregation is provided. The method comprises: receiving a control signal for a transmission of a carrier activation or deactivation message command, the control signal including an indication of a first resource for the UE to transmit a feedback message to a network node; receiving a carrier activation or deactivation message; wherein the carrier activation or deactivation message comprises an indication of a second resource for the UE to transmit the feedback message to the network node; decoding the received carrier activation or deactivation message; and sending the feedback message using one of the first resource and the second resource based at least on a result of the decoding of the carrier activation or deactivation message. A UE for performing this method is also provided.

Methods and nodes for activation or deactivation of a carrier in a communication network supporting carrier aggregation

A semantic segmentation method for aircraft point cloud based on voxelization and three views, including: filtering a collected point cloud followed by centralization to obtain a centralized point cloud; inputting the centralized point cloud into a T-Net rotation matrix network; rotating the centralized point cloud to a front side followed by voxelization to obtain a voxelized point cloud; subjecting the voxelized point cloud to voxel filling to obtain a voxel-filled point cloud; calculating thickness maps of three views of the voxel-filled point cloud, followed by sequentially stitching and inputting to the point cloud semantic segmentation network to train the point cloud semantic segmentation network; inputting the collected point cloud into the trained point cloud semantic segmentation network; and predicting a result semantic segmentation of a 3D point cloud of the aircraft.

Semantic segmentation method for aircraft point cloud based on voxelization and three views

Growing community of inventors

Nanjing, China

Jun Wang