Abstract: A method includes using at least one processor to detect that a tool coupled to an end effector of a robot having multiple joints is contacting a surface. The robot includes multiple joint motors configured to control multiple motions of the multiple joints. One or more control systems are configured to control each of the joint motors in a joint position mode. The method also includes identifying, via the at least one processor, a first joint of the multiple joints in response to detecting that the tool is contacting the surface. The method also includes sending, via the at least one processor, a command to at least one of the one or more control systems associated with a first joint motor of the multiple joint motors that corresponds to the first joint. The command is configured to cause the at least one of the one or more control systems to operate in a torque mode. The method also includes sending, via the at least one processor, a joint torque value to the at least one of the one or more control systems.

Abstract: A method for detecting defects in products from images thereof and an electronic device applying the method inputs a defect image repair data set into an autoencoder to train the autoencoder, and generates a reconstructed image, calculates a reference error value between the sample image and the reconstructed image by a preset error function, and set a threshold value based on the reference error value. The electronic device inputs an image possibly revealing a defect into the autoencoder and generates the reconstructed image corresponding to the image to be detected, and uses the preset error function to calculate the reconstruction error between the image and the reconstructed image, thereby determining whether the image being analyzed does reveal defects. When the reconstruction error is greater than the threshold value, a determination is made that a defect is revealed.

Abstract: A cleaning attachment and rotary tool for cleaning are provided. A cleaning attachment includes a housing, a plurality of bristles, and a drive shaft extending from the housing in an axial direction, the drive shaft being coupled to the housing such that rotation of the drive shaft causes rotation of the housing. The housing includes a reservoir, a valve for dispensing the fluid from the reservoir, and at least one resilient element surrounding an axis defined by the drive shaft. Translation of the drive shaft along the axial direction causes compression of the resilient element, and compression of the resilient element causes the valve to open to dispense fluid from the reservoir. A rotary tool for cleaning includes the cleaning attachment and a rotary tool having a motor, power source, and a chuck configured to receive the drive shaft of the cleaning attachment.

Abstract: A developing cartridge includes: a cartridge body for accommodating developers; a driving part at a first end in a first direction; a developing roller rotatable about a rotation axis extending along the first direction and positioned at a third end in a second direction; a storage medium for storing information; an electrical contact surface electrically connected to the storage medium and fixed relative to the cartridge body, and a lifting element configured to lift up the cartridge body in the third direction. The electrical contact surface is positioned at the first end in the first direction and is farther away from the developing roller relative to the driving part. The lifting element is farther away from the developing roller in the second direction away from the driving part.

Abstract: A method of a new energy centralized control station network based on dynamic IP can determine the occurrence frequency and the occurrence interval of the abnormal traffic, further analyze and determine the defense state of a new energy centralized control station based on this, and execute the corresponding dynamic optimization solution according to different defense states, so that the IP address and the firewall of the new energy centralized control station can be dynamically adjusted to ensure the security performance of the new energy centralized control station, reduce the risk of external intrusion, effectively resist malicious network reconnaissance such as scanning attacks and ensure the stable operation of a new energy power generation system.

Abstract: A system for identifying movers in an independent cart system includes movers having at least one magnet and sensors generating a feedback signal responsive to detecting a magnetic field from the magnet as each mover travels past the sensor. A memory stores an identifier and a corresponding digital fingerprint for each mover. The stored digital fingerprint is generated as a function of the magnetic field generated by the magnet on each mover. A controller receives the feedback signal from each sensor and determines a run-time digital fingerprint for each mover corresponding to the magnetic field generated by the magnet on each mover as a function of the feedback signal. The run-time digital fingerprint is matched to one of the stored digital fingerprints, and the identifier, corresponding to the stored digital fingerprint matching the run-time fingerprint, is read from memory.

Abstract: An independent cart system includes multiple track segments, multiple movers, multiple position sensors, and a controller. Each track segment includes multiple coils along a length of the track segment and a segment controller operative to selectively energize the coils to generate an electromagnetic field. Each mover includes a magnet array having multiple magnets, where the magnet array generates a magnetic field that interacts with the electromagnetic field generated by the coils to propel the mover along the track segments. The position sensors are spaced apart along the length of each track segment and generate a position feedback signal with a waveform as a function of the magnetic field generated by the magnet array on each mover. The controller is operative to receive the position feedback signal from each position sensor and to determine a unique identifier for each mover as a function of the waveform of the position feedback signal.

Abstract: Methods for adversarial training and/or for analyzing the impact of fine-tuning on deep learning models may include receiving a deep learning model comprising a set of parameters and a dataset of samples. A respective noise vector for a respective sample may be generated based on a length of the sample and a radius hyperparameter. For a target number of steps, the following may be repeated: adjusting the noise vector based on a step size hyperparameter, and projecting the respective noise vector to be within a boundary. The parameters of the deep learning model may be adjusted based on a gradient of a loss based on the noise vector. This may be repeated for each sample of the plurality of samples. A system and computer program product are also disclosed.

Abstract: A method includes extracting, by an analysis computer, a plurality of first datasets from a plurality of graph snapshots using a structural self-attention module. The analysis computer can then extract at least a second dataset from the plurality of first datasets using a temporal self-attention module across the plurality of graph snapshots. The analysis computer can then perform graph context prediction with at least the second dataset.

Abstract: The present disclosure is at least directed to utilizing air curtain devices to form air curtains to separate and isolate areas in which respective workpieces are stored from a transfer compartment within a workpiece processing apparatus. The transfer compartment of the workpiece processing apparatus includes a robot configured to transfer or transport ones of the workpieces to and from these respective storage areas through the transfer compartment and to and from a tool compartment. A tool is present in the tool compartment for processing and refining the respective workpieces. Clean dry air (CDA) may be circulated through the respective storage areas. The air curtains formed by the air curtain devices and the circulation of CDA through the respective storage areas reduces the likelihood of the generation of defects, damages, and degradation of the workpieces when present within the workpiece processing apparatus.

Abstract: Systems, methods, and computer program products are provided for saving memory during training of knowledge graph neural networks. The method includes receiving a training dataset including a first set of knowledge graph embeddings associated with a plurality of entities for a first layer of a knowledge graph, inputting the training dataset into a knowledge graph neural network to generate at least one further set of knowledge graph embeddings associated with the plurality of entities for at least one further layer of the knowledge graph, quantizing the at least one further set of knowledge graph embeddings to provide at least one set of quantized knowledge graph embeddings, storing the at least one set of quantized knowledge graph embeddings in a memory, and dequantizing the at least one set of quantized knowledge graph embeddings to provide at least one set of dequantized knowledge graph embeddings.

Abstract: In implementations of the subject matter as described herein, there is provided a method for forgery detection of a face image. Subsequent to inputting a face image, it is detected whether a blending boundary due to the blend of different images exists in the face image, and then a corresponding grayscale image is generated based on a result of the detection, where the generated grayscale image can reveal whether the input face image is formed by blending different images. If a visible boundary corresponding to the blending boundary exists in the generated grayscale image, it indicates that the face image is a forged image; on the contrary, if the visible boundary does not exist in the generated grayscale image, it indicates that the face image is a real image.

Abstract: Provided in embodiments of the present application are an occlusion detection method for medical imaging, a medical imaging method, and a medical imaging system. The occlusion detection method for medical imaging includes: acquiring an image sequence, the image sequence including a plurality of images of an object in a time dimension; and according to at least one among confidence level change information and depth change information of a keypoint of the object in the plurality of images, determining an occlusion state of the keypoint.

Abstract: A method for autonomous path planning triggered by freeway rubbernecking includes obtaining gaze directions of drivers of a plurality of connected vehicles in a region of a road. The method also includes determining whether an average speed of the plurality of connected vehicles in the region is less than a predetermined speed. The method further includes controlling mobility of one or more of the plurality of connected vehicles in response to determining that the gaze directions of one or more drivers of the plurality of connected vehicles deviate from the moving direction of the connected vehicles and determining that the average speed of the connected vehicles is less than the predetermined speed.

Abstract: A method and apparatus for determining the orientation of a subject during medical imaging, and a medical imaging system is provided. The method includes: acquiring an image sequence of a subject captured by a camera, the image sequence comprising a plurality of images based on a time sequence; determining, based on the image sequence, whether the state of the subject is abnormal; and in response to a determination result that the state of the subject is abnormal, determining the orientation of the subject by using a first orientation or a second orientation, wherein the first orientation is the orientation of the subject determined based on an image before a period when the state of the subject is determined to be abnormal, and the second orientation is the orientation of the subject determined based on an image after the period when the state of the subject is determined to be abnormal.

Abstract: A keypoint detection method is provided for medical imaging, a medical imaging method and a medical imaging system. The keypoint detection method for medical imaging includes: receiving an image sequence, the image sequence comprising a plurality of images of an object; performing keypoint detection on the plurality of images separately, and generating a keypoint image sequence; performing keypoint occlusion detection on the images in the image sequence; and determining keypoint distribution information according to the keypoint image sequence and a result of the keypoint occlusion detection.

Abstract: Disclosed herein are novel wavy, multi-component vascular grafts (WMVGs) with a wavy inner layer of rigid biopolymer fibers and an outer layer of elastic biopolymer fibers and a method for preparing WMVGs via electrospinning using a special assembled collector. The fabricated WMVGs closely mimic the non-linear tensile stress-strain relationship of native blood vessels and showed sufficient mechanical strength needed for implantation.

Abstract: EUV masks and methods of fabrication thereof are described herein. An exemplary method includes receiving an EUV mask having a multilayer structure, a capping layer disposed over the multilayer structure, a patterned absorber layer disposed over the capping layer, and a patterned hard mask disposed over the patterned absorber layer. The method further includes removing the patterned hard mask by performing a first etching process to partially remove the patterned hard mask and performing a second etching process to remove a remainder of the patterned hard mask. The first etching process uses a first etchant, and the second etching process uses a second etchant. The second etchant is different than the first etchant. In some embodiments, the first etchant is a halogen-based plasma (e.g., a Cl2 plasma), and the second etchant is a halogen-and-oxygen-based plasma (e.g., a Cl2+O2 plasma).

Abstract: A method for measuring a growth height of a plant, an electronic device, and a storage medium are provided. The method controls a camera device to obtain a color image and a depth image of a plant to be detected. The color image is detected by a detection model which is pre-trained, and a plurality of detection boxes which includes a plurality of plants to be detected is obtained. The color image and the depth image are aligned to create an alignment image. A plurality of target boxes is acquired from the alignment image, and depth values of the plurality of target boxes are determined. The quantity of the target boxes and a height of one or more plants to be detected are determined, no manual operations are required.

Abstract: Methods and systems for performing federated private anomaly detection are disclosed. An anomaly detector computer can collaborate with an aggregator computer and an account management computer in order to train machine learning models, which can be used to classify events as not fraudulent or fraudulent. The anomaly detector computer can obliviously use private information (e.g., account flags and account flag values) held by the account management computer to train and use the machine learning models, such that the anomaly detector does not become aware of the account flags or account flags values. Such a system can be used, for example, for the detection and prevention of financial crime. The anomaly detector computer can use the account flag information possessed by the account management computer to identify fraudulent events performed by customers of the organization operating the account management computer.