117 results about "Local machine" patented technology

The systems and methods for improving pose estimation models are disclosed herein. Digital image data of an environment can be obtained and provided to a first machine learning model. A first confidence metric can be computed for the image. The first confidence metric can be compared with a threshold value and provided to a second machine learning model. A second confidence metric can be generated for training of machine learning models for pose estimation. A generic machine learning model can be updated using model parameters from trained local machine learning models.

The invention discloses a multi-robot path coordination system based on reinforcement learning, and the system comprises a state collection module which is used for collecting the operation state and environment perception information of each robot, and generating an original data set; the state modeling module is used for constructing local state representation for each robot in each decision period; the opponent perception module is used for executing opponent learning perception training and outputting opponent parameter estimation vectors; the improved LOLA module is used for generating a final estimation result on the basis of self-strategy updating of the traditional LOLA; the strategy updating module is used for correcting the updating direction of the reinforcement learning strategy of the local machine and generating a feasible action set; the arbitration decision module is used for executing arbitration solution in combination with the candidate path action distribution to generate a scheduling result; and the execution feedback module is used for issuing the control action to a robot executor for execution, generating a path coordination result and storing the path coordination result. According to the invention, path coordination of multiple robots is realized.

The invention provides an unmanned aerial vehicle group obstacle avoidance method and system based on least square and distributed arbitration, and belongs to the field of multi-unmanned aerial vehicle path planning, and the method comprises the steps: maintaining the historical observation data of a dynamic obstacle; a trajectory prediction model is constructed based on historical observation data, and a future trajectory prediction set is generated through least square fitting; selecting an optimal prediction model from the trajectory prediction set through a backtracking verification strategy; on the basis of a preset distributed arbitration rule, whether the local machine obtains the release right of the obstacle prediction information or not is judged; packaging the trajectory information corresponding to the optimal prediction model into a standard message by using the unmanned aerial vehicle which obtains the release right, and broadcasting the standard message to the cluster; and incorporating the obtained predicted trajectory message into collision avoidance constraint of a planner, performing local trajectory planning, and generating a collision-free flight path. According to the method, the safety, the collaboration and the passing efficiency of the cluster system in a real complex scene are remarkably improved.

A computing device, that is configured to configure a global machine learning model, performs respective electronic risk audits of client devices configured to train respective local machine learning models that correspond to a global machine learning model. Based on respective electronic risk scores of one or more of the client devices, determined via the respective electronic risk audits, the computing device implements one or more parameter privacy adjustment methods on respective parameters received from the client devices prior to using the respective parameters to configure the global machine learning model, wherein respective client devices determined to have higher electronic risk scores have more of the parameter privacy adjustment methods applied than other respective client devices determined to have lower electronic risk scores. The computing device provides, to the client devices, the global machine learning model configured according to the respective parameters as adjusted.

This invention discloses a data processing method, system, device, and medium based on a user-selected local model. The method includes: acquiring a local machine learning model file in response to a user operation; dynamically parsing model metadata to determine input and output specifications; acquiring multimodal target data and preprocessing and adapting it according to the input specifications; calling local hardware resources to execute model inference; parsing the inference results according to the output specifications and outputting them to the target business system. This invention overcomes the closed nature of traditional software-built-in AI models, achieving full utilization of local AI computing power and effective privacy protection of sensitive data. Through a highly abstract underlying adaptation architecture, this invention is not dependent on specific application scenarios and can be widely integrated as a general-purpose underlying technology into various industry software such as video restoration, medical image analysis, professional image retouching, and industrial inspection, giving the software flexible AI capability scalability.

The invention discloses a crystal oscillator-free Bluetooth chip pairing method and device and a communication system, and the method comprises the steps: switching a preset frequency offset value for polling between a paging scanning state and a query scanning state, so as to search an ID packet sent by host equipment in the query state; when the ID packet is received, a slave equipment information packet of the host is sent to the host equipment; receiving a host equipment information packet sent by the host equipment in response to the slave equipment information packet, so that the host equipment can establish Bluetooth connection with a local machine; and sending a connection request to the host device according to the address information of the host device loaded in the host device information packet in the paging state to establish a Bluetooth connection with the host device so as to realize Bluetooth pairing between the local machine and the host device. For low-power-consumption equipment, the Bluetooth chip without the crystal oscillator is ensured to be paired with other equipment to establish connection. The power consumption of low-power-consumption equipment is reduced, the development difficulty, development cost and period of products are reduced, and the equipment is more miniaturized.

Methods and systems for automating surgical procedures in model organisms. The system includes a user input panel, a trained computer vision system, a server, and articulated robotic arm. The system may further include an analytical scale, an anesthesia chamber, a tube labeling system, a tissue freezing system, and a biohazard waste disposal system. The computer vision system communicates with the robotic arm to coordinate tissue collections and common research procedures such as injections via recognition models related to detection, identification, localization, and classification. The central server provides rapid synchronization of data between the local machine and a cloud account system for real-time data analytics. The anesthesia chamber provides standardized administration of anesthesia, the sample labeling system provides 2D barcoded labels according to user input, and the tissue freezing and waste disposal systems enable storage of samples and removal of carcasses following tissue collection, fully automating the necropsy.

This invention discloses a federated averaging algorithm based on residual networks, characterized by high accuracy and efficient communication: Step 1: The server partitions the MNIST dataset into IID and Non-IID categories and distributes them to various clients; Step 2: Each client locally constructs a residual network model; Step 3: The number of communication rounds R between the server and clients is set. In each round, the server randomly selects K clients to participate in the communication and sends the parameters of the global model; Step 4: The selected clients download the parameters of the global model and perform local machine learning training to obtain the trained model parameters; Step 5: The K clients upload the trained model parameters to the server; Step 6: The server updates the global model parameters according to a weighted average aggregation strategy and repeats steps 3-6 until the final global model parameters are obtained in the Rth round of communication. This invention significantly improves accuracy and communication efficiency compared to existing technologies.

Techniques for training a global model for use at a host of oilfield application sites based on raw data obtained from the application sites without direct exposure of the data to the global model. The techniques include developing and distributing a global model with a predetermined set of parameter weights. The model is then locally employed at each application site by a local computer which maintains the integrity of the acquired data during performance of the oilfield application. The data is used to update the parameter weights based on real-time circumstances. Thus, the parameter weights may be transmitted to the centralized computer for updating of the global model. Further, the updated global model may continue to direct other applications and the process continued in a beneficial feedback loop manner.

This invention discloses a method, apparatus, medium, and product for calculating bonding interactions in molecules. The method includes: allocating bonding tasks using a three-dimensional interconnected computing architecture, and constructing a bonding atom set and a bonding task index table matching the allocated bonding tasks within each computing module. This ensures that identical atoms do not need to be repeatedly stored when the computing module executes all bonding tasks, reducing the storage overhead of atomic information; dynamically acquiring and filling atomic information from the local machine or other computing modules into the corresponding storage locations of the bonding atom set, extracting atomic information according to the index table, calculating bonding components in parallel, and accumulating the bonding components of identical atoms to obtain the resultant bonding force; and updating the resultant bonding force and atomic information through inter-module interaction, and re-storing the atomic information according to the updated spatial arrangement. This reduces the amount of data transmission between modules while ensuring the accuracy of molecular bonding interaction simulation, thereby improving the stability and efficiency of the overall molecular dynamics simulation.

A biological sample analysis system and a biological sample analysis method. The biological sample analysis system includes a plurality of imaging devices, a plurality of local computing devices, and a cloud computing device. Each imaging device captures a series of images from a biological sample. Actions performed by each local computing device include: receiving the series of images from a corresponding one of the imaging devices; and processing the series of images using a local machine learning model to generate an analysis result. The cloud computing device performs actions including: training a cloud machine learning model based on the series of images and the analysis result from the local computing devices to generate an optimized machine learning model; and transmitting the optimized machine learning model to each local computing device to replace or update the local machine learning model.

Proposed are a test and measurement system for evaluating learning models of one or more test devices, a test and measurement procedure, and a computer program. A test and measurement system (100) for evaluating learning models of one or more test devices (102; 104; 106; 108), comprising a test and measurement device (10), includes one or more interfaces (12) configured for data communication with the one or more test devices (102; 104; 106; 108). The test and measurement device (10) further comprises one or more computing units (14) configured to generate stimulus data for the one or more test devices (102; 104; 106; 108).The one or more computing units (14) are configured to provide the stimulus data to the one or more test devices (102; 104; 106; 108) via the one or more interfaces (12), to train one or more local machine learning models by the one or more test devices (102; 104; 106; 108) based on the stimulus data, and to receive training model data from the one or more trained machine learning models via the one or more interfaces (12) from the one or more test devices (102; 104; 106; 108). The one or more computing units (14) are configured to evaluate the quality of the one or more trained machine learning models based on the training model data.

A privacy-preserving authentication method for anti-counterfeiting QR codes using Vision Transformer- (ViT-) based Federated Learning (FL) is provided. In the method, an individual client authenticates an anti-counterfeiting QR code as captured in an image presented to the individual client. A local machine-learning (ML) model of the individual client determines authenticity of the anti-counterfeiting QR code as captured in an image presented to the individual client. The local ML model is initialized as a pretrained ViT-based model, pretrained on the large-scale ImageNet dataset for processing an input image to determine authenticity of the anti-counterfeiting QR code as captured in the input image. The plurality of clients performs a cyclic weight transfer FL process to update respective local ML models of the plurality of clients according to instant pluralities of training data respectively owned by different clients in the plurality of clients while preserving training-data privacy among the different clients.

The invention provides a centralized storage method and device for an airborne multi-relation table database. A transaction management unit obtains a transaction operation instruction and sends the transaction operation instruction to a data mirror image table processing unit; the data mirror image table processing unit locally searches whether a corresponding data relation list exists or not based on the transaction operation instruction, if yes, a data table corresponding to the data relation list is locally obtained, and operation indicated by the transaction operation instruction is carried out on the obtained data table; after executing all transaction operation instructions, the data mirror image table processing unit temporarily stores the updated data relation list and the updated data table locally; when a preset condition is met, the updated data relation list and the updated data table are sent to a multi-relation-table storage unit to be stored, and the original data relation list and the original data table of the multi-relation-table storage unit are covered; the system performance and the resource utilization rate can be improved; the method is used for managing data tables in a disk in an airborne embedded database.

The utility model discloses a handheld nondestructive testing machine, which relates to the technical field of nondestructive testing, and comprises a chest hanger, one side of the chest hanger is fixedly connected with a support column, the inside of the support column is rotatably connected with a first ball screw, bearings are arranged in the support column and are positioned at the two ends of the first ball screw, and the first ball screw is rotatably connected with the first ball screw. The outer side of the first ball screw is in threaded connection with a push column, one end of the push column extends to the outer side of the supporting column and is fixedly connected with two side plates, a bearing is arranged on one side of each side plate, a rotating shaft is mounted in each bearing, and a mounting plate is fixedly connected between the two rotating shafts. By arranging the supporting column, the mounting plate and the main machine, the main machine can be hung in front of the chest of a user, the distance between the main machine and the chest of the user and the angle of the main machine can be adjusted, so that the user does not need to hold the main machine to operate the main machine, and the ultrasonic flaw detection work can be conveniently carried out by the machine.

The invention discloses a gradient tube structure reverse design framework construction method for inhibiting initial peak force, which comprises the following steps of: constructing a global machine learning agent model, reversely designing a stable section section, namely parametrically defining a thin-walled tube section of a gradient tube structure, sampling and simulating to obtain training data to train the global machine learning agent model, and constructing a stable section section; reversely solving optimal stable section design parameters by taking the target platform force as input; constructing a local machine learning agent model inverse design and an optimized gradient section, namely training the local machine learning agent model by using the section shape of the optimal stable section, and reversely solving optimal wall thickness and length distribution parameters of the gradient section by using a multi-objective optimization algorithm with smooth transition and target platform force as constraints; and combining the optimal stable section design parameter and the optimal wall thickness and length distribution parameter of the gradient section for simulation verification. According to the method, the mechanical response in the crushing process is accurately controlled, and effective suppression of peak force and maximum energy absorption efficiency are realized.

The application relates to a low-voltage interconnection system software detection and master-slave switching method. The low-voltage interconnection system comprises a plurality of low-voltage areas, each of which is provided with an area intelligent terminal, the method is applied to each area intelligent terminal, and the method comprises the following steps: establishing a shared memory area for storing system running state data; sending a heartbeat message at a set time interval, and simultaneously listening to heartbeat messages of other area intelligent terminals to determine the running state of master-slave node; according to the running state of the master-slave node and the node priority order, running the local machine as a master node or a slave node; when running as a master node, collecting the system running state data and storing the system running state data into the shared memory area, and sending change data to the slave node; when running as a slave node, receiving the change data sent by the master node and saving the change data into the shared memory area. The application can realize data synchronization and real-time detection of the body and software running state of the plurality of area intelligent terminals.

System and method for a data environment using linked chains of data packages and logical packages to allow for identification of server data from a local machine without allowing access to unencrypted data by the server. The local machine may identify all data packages and all logical packages in the data environment without downloading all data packages and logical packages providing for faster remote data use with improved security. Data packages and logical packages may be organized by their unique checksums. Header information contained within the data packages and logical packages allows a local machine to identify all data packages in the data environment while only downloading and decrypting a minimal number of data packages or logical packages and data.

The invention discloses a high-real-time virtual PLC implementation method and system for wind power control, and relates to the technical field of industrial control, and the method comprises the steps: deploying a real-time patch host machine operation system, generating a Hypervisor layer, and creating a real-time virtual machine based on a wind power plant private cloud server; by modifying system guide parameters, exclusive binding of a specific CPU on a physical server and a virtual CPU of a virtual PLC instance is isolated, a large-page memory is distributed for the virtual PLC instance, and memory locking operation is executed; configuring an SR-IOV network card, transmitting the SR-IOV network card to a virtual PLC instance in a PCIe unvarnished transmission mode, deploying a user mode real-time protocol stack, and realizing extremely low delay and high-precision time synchronization of network I / O; according to the method, an IEC 61131-3 standard control logic program is followed, a local machine code is compiled by an LLVM compiler, a dynamic shared library is generated for a virtual PLC real-time task to be directly loaded and run, a double-cycle task scheduling architecture is set to monitor task execution jitter online, and real-time performance self-diagnosis and operation and maintenance analysis are supported. The method has the beneficial effect that the downtime is shortened.

The application provides a machine learning model processing method and device and a storage medium. A UE creates a local machine learning model of a target application in advance according to a global machine learning model provided by a first network function entity, determines local training data related to the target application, trains the local machine learning model according to the local training data, and sends local model parameters of the trained local machine learning model to the first network function entity to update the global machine learning model by the first network function entity. Federated learning between the UE and the first network function entity for providing the model can be implemented, the performance of sharing, transmitting and training the machine learning model between the UE and the network is improved, and the rapidly developing communication service and application requirements are met.

The invention provides an ACAS X anti-collision decision-making method, system and equipment based on probability prediction, and relates to the technical field of aviation safety, and the method comprises the steps: when a local machine detects a target machine, setting the relative motion state information of the two machines at an initial moment as an observation value at the initial moment, setting the variance at the initial moment as 0, and setting the variance at the initial moment as 0; constructing a state vector at an initial moment; establishing a discrete time system state equation for describing time evolution of the state vector; for each action in the candidate action set, on the basis of the state vector at the initial moment and a system state equation, online predicting future probability distribution of the state vector after the local machine executes the current action; based on the future probability distribution of the state vector, calculating the conflict probability of the local machine and the target machine in the future after the local machine executes the current action; and the action with the minimum conflict probability is determined as the optimal anti-collision decision instruction from the candidate action set, so that the problems of poor flexibility, difficulty in updating and insufficient real-time performance of an existing anti-collision system are solved.

This application discloses a distributed data service sharing method, which relates to the field of data sharing technology. The method framework keeps the original data locally on the data provider's local machine, and only supports data users to discover data by registering the metadata information of the shared data. At the same time, it combines technical mechanisms such as hybrid encryption, authorized lists and distributed ledgers to achieve secure and efficient data sharing.

The present application provides a detection method for SpringBoot Actuator vulnerabilities, comprising: inputting a target IP address list to a scanning module, and detecting the network reachability of each IP address in the IP address list; if any IP address is reachable in a network, detecting all ports of the IP address, and recording open port numbers; performing service classification by identifying services running on open ports; if a service is classified to as a SpringBoot Web-type service, determining whether an Actuator path of the service is accessible; if the access is successful, using a crawler module to crawl a Web page under the Actuator path; if the crawled content contains a Heapdump binary file, downloading the Heapdump file to a local computer; parsing the downloaded Heapdump file by means of a parsing module to extract sensitive information from the file; and a detection report module generating a SpringBoot Actuator vulnerability detection report on the basis of the extracted sensitive information.

1. Name of the product in this design: Operation area of ​​the graphical user interface for remote communication of electronic equipment. 2. Purpose of this design: An electronic device. 3. The key design features of this product are the graphical user interface of the part for which protection is sought. 4. The picture or photo that best illustrates the key design points: Design 1 front view. 5. Design 1 is designated as the basic design. 6. Purpose of the graphical user interface: The operation area and the overall interface are used to realize remote communication. 7. Human-computer interaction method of graphical user interface: The main view of Design 1 is the remote communication interface, including: the Communications operation box, the Control operation box, and the Send message to host operation box. At this time, it is in the Local Online state. The green line in the Control operation box indicates that the local machine and the factory end maintain a remote connection. The green box indicates that the operator is the local machine, and the local machine can send information to the factory end (FA). Clicking the Remote button in the Control operation box switches to the change state diagram of Design 1. At this time, it is in the Remote Online state. The green box indicates that the operator is the factory end, and the local machine can receive information sent by the factory end. Design 2's main view is the remote communication interface, including: the Communications operation box, the Control operation box, and the Send message to host operation box. At this time, it is in the Local Online state. The green line in the Control operation box indicates that the local machine and the factory side maintain a remote connection. The green box indicates that the operator is the local machine, and the local machine can send information to the factory (FA). Clicking the Offline button in the Control operation box switches to the Design 2 change state diagram. At this time, it is in the Local Offline state. The green box indicates that the operator is the local machine, and local operations can be performed. Design 3's main view is the remote communication interface, including: the Communications operation box, the Control operation box, and the Send message to host operation box. At this time, it is in the Local Online state. The green line in the Control operation box indicates that the local machine and the factory are maintaining a remote connection. The green box indicates that the operator is the local machine, and the local machine can send information to the factory (FA). Clicking the Connect button in the Communications operation box switches to the Design 3 change state diagram, which indicates that the local machine and the factory are disconnected from the remote connection. 8. Other situations requiring explanation: In each view interface, the area covered by the red semi-transparent color block is the part that does not require protection.

This invention discloses a distributed task offloading method and apparatus suitable for AI inference on mobile devices. The method involves a performance evaluator that obtains the execution time of the AI ​​model on both an edge server and the mobile device, and comprehensively assesses the offloading gain of offloading the AI ​​model's execution task to the edge server, considering the difference in inference accuracy. A workload predictor constructs a load prediction model based on historical load data periodically pushed by the edge server through polynomial fitting, and predicts the future workload of the edge server using this model, obtaining a load prediction result. A distributed task scheduler combines the offloading gain and the load prediction result to construct a stochastic optimization problem, and determines the optimal execution position of the AI ​​model on the local machine or the edge server by solving the stochastic optimization problem. This invention meets the high performance and low overhead requirements of AI inference on mobile devices.

This invention provides a method and system for dynamically generating and exporting Excel files, comprising: Step 1: defining a data structure for the content of the Excel file to be exported; Step 2: querying the data to be written to the Excel file and writing the data; Step 3: setting the style attributes of the Excel file; Step 4: downloading the generated Excel file to the local machine. This invention improves the efficiency of report file generation and export without affecting the normal operation of the system; it has strong scalability, achieving dynamic generation of Excel files through dynamically maintaining the defined data structure.

A flexible digital radiographic detector assembly includes a flexible sleeve enclosing a photosensor array supported by a flexible substrate. Integrated circuit readout electronics are coupled to the photosensor array. The digital radiographic detector assembly communicates with a local computer system or with a cloud server to transmit image data captured in the photosensor array, to display the image data on a screen, and to analyze defects that may be present in the image.

The application provides a grouting density and flow detection device and method, relates to the technical field of grouting monitoring, and comprises a control module, a density data acquisition module, a flow data acquisition module, a data processing module, a data transmission module, a data storage module and a display control module.The control module is used for exciting a sensor; the density data acquisition module is used for applying ultrasonic waves to a density detection pipe section and converting the ultrasonic waves into electric signals; the flow data acquisition module is used for applying ultrasonic waves to a flow detection pipe section and converting the ultrasonic waves into electric signals; the data processing module is used for analyzing cement slurry density and flow after filtering sound wave data; the data transmission module is used for outputting analyzed data to a control end; the data storage module is used for storing backup data locally; and the display control module is used for monitoring a working state and adjusting parameters on the local machine.The application can monitor cement slurry density and flow parameters flowing in a pipe.

The invention relates to distributed machine learning, such as federated learning. A client device outputs a learned local weight set after training a local machine learning model using a local data set. Each learned local weight is of a Boolean data type. The client device is configured to send the learned local set of weights to a server. After receiving the plurality of sets of learned local weights from the plurality of client devices, the server is used to aggregate corresponding local weights and determine a global set of weights based thereon. The server is then used to send the global weights to one or more client devices. Each global weight is of a Boolean data type. By exchanging a one-bit binary local weight and a one-bit binary global weight between each client device and the server, the communication overhead of distributed machine learning can be reduced.

A method for training a local machine learning model is provided. The method includes receiving a scanning electron microscope (SEM) image of semiconductor features. The method additionally includes determining a location and dimensions of a bounding box within the SEM image. The method yet further includes determining, whether a defect feature exists within the bounding box, based on an unsupervised object detection process. The method also includes, if the defect feature exists within the bounding box, receiving positive rewards. The method also includes, if the defect feature does not exist within the bounding box, receiving negative rewards.