555 results about "Intelligent computing" patented technology

The invention discloses an intelligent water resource control platform based on cloud computing and an expert system. The platform comprises a site measurement and control station, an internet of things data transmission system and a cloud computing service platform, and is characterized in that the control platform carries out automatic real-time collection on water resource information of a water supply water source, underground water, a water intake, a boundary control fracture surface and a sewage draining exit and site equipment operating condition information through the site measurement and control station, and real-time data preprocessing is achieved by means of an additionally installed site intelligent controller; internet of things data transmission equipment is adopted in the internet of things data transmission system to sent data of the measurement and control station to the cloud computing platform, and data support is provided for water resource management and advanced applications; the cloud computing service platform achieves real-time monitoring on water resources, monitoring on operating condition and remote on-line maintenance on site monitoring equipment and applications on water resource service, and a water resource decision is utilized to support the expert system to achieve assessment, early warning and emergency commanding of the water resources. The cloud computing and the Internet of things technology are applied to the platform, so measurement and control data are timely and efficiently processed, the intelligent computing technology is applied so that intelligent assessment and early warning of the water resources can be achieved, and safety of the water resources is guaranteed to a great extent.

A collision detection and data synchronization mechanism operates to expand the speed and capability of distributed applications. The execution environment employs collision detection and data synchronization to distribute data objects to appropriate network nodes to support local processing. The collision detection and data synchronization mechanism defines the set of processes and algorithms used to track distributed data objects and to manage changes to the distributed objects throughout a network of intelligent computing devices. Data and code are distributed to local computing devices necessitating tracking and synchronizing updates. The system and method ensures that data updates received during the course of a local transaction do not unwillingly affect the results of other transactions by notifying the local computing devices of changes to the data that are subject of the transaction. These conflicts are then resolved in the applications, and notification of the resolution is sent to the remaining intelligent computing devices.

The present invention discloses methods and apparatus for developing, analyzing, investigating, and advising healthcare and well-being related decisions. In particular, the present invention relates to the architecture of systems in either stand-alone or distributed / collaborative / pervasive settings, the components of the systems and their underlying processes and couplings, the computational techniques built into the methods, input data sources integrated into and output results produced and distributed by the systems, as well as the apparatus for carrying out the corresponding user interaction, data access and collection, data integration and processing, data-driven inferences and simulation, intelligent computations, decision analytics, and decision support to generating solutions to various healthcare analytics and decision-making problems. This invention also relates to two working illustrations of the methods and apparatus that present the embodiment illustrations of the present invention.

A method for displaying a user-defined artifact in a debugger view is provided. The user-defined artifact is defined in a software application that is developed by a computer software application developer. The software application is included in a packaged application that includes instructions, with each instruction being either a user-defined instruction or a non-user defined instruction. A breakpoint is defined at a desired instruction of the packaged application. The method includes initiating debugging of the packaged application and hitting the breakpoint defined at the desired instruction. Also included is identifying the user-defined artifact in the software application and obtaining data to populate the debugger view. Further included is filtering-in data associated with the user-defined artifact and displaying the debugger view showing the filtered-in data. The method also includes marking data not associated with the user-defined artifact and stepping through the user-defined instruction.

An accurate grammar analyzer based on a so-called POST (part-of-speech tagged) parser and a learners' model for use in automated language learning applications such as the template-based ICALL (intelligent computer assisted language learning) system.

A system and apparatus to provide remote interactive access to multiple databases and, specifically, related to a system and apparatus to provide remote interactive access to real estate information related databases using a portable, voice interactive computing device with GPS capabilities and “smart” computer features. The system combines access to search engines configured to retrieve real estate listings, publicly available statistics about a specific geographic area, and retrieval of stored forms specific to completing an offer for real estate. The device can have several components in various combinations including a touch sensitive screen, a keyboard, a processor, GPS circuitry, a microphone, an audio device, wireless circuitry, cellular circuitry, a data jack, a wireless remote, and a camera. Software utilizing those components maximizes efficiency for a real estate agent in the field.

The invention provides a behavior recognition method of depth supervised convolution neural network based on training feature fusion, belonging to the artificial intelligence computer vision field. This method extracts multi-layer convolution features of target video, designs local evolutionary pooling layer, and maps video convolution features to a vector containing time information by using local evolutionary pooling layer, thus extracts local evolutionary descriptors of target video. The local evolutionary descriptors of target video are extracted by using local evolutionary pooling layer.By using VLAD coding method, multiple local evolutionary descriptors are coded into meta-action based video level representations. Based on the complementarity of the information among the multiple levels of convolution network, the final classification results are obtained by integrating the results of the multiple levels of convolution network. The invention fully utilizes the time information to construct the video level representation, and effectively improves the accuracy of the video behavior recognition. At the same time, the performance of the whole network is improved by integrating the multi-level prediction results to improve the discriminability of the middle layer of the network.

A system and process for directing a laser beam within a space is presented. The system includes a pointing device which periodically outputs orientation data indicative of the direction it is pointing and a cursor device which projects a laser beam. The orientation data is used to compute the direction the pointing device is pointing in terms of yaw and pitch angles. The laser beam is directed to locations in the space relative to the amount of movement of the pointing device. In an absolute mode, the pointed device and the laser beam of the cursor device are pointed at the same location in the space, whereas in a relative mode, the pointing device does not point at the same location as the laser beam. The position of the laser can be used to select a variety of hardware devices known to be in a room for future control actions.

The invention discloses a rotary type automatic goods storing and taking device and method. The device comprises a system closing, supporting and assisting unit, a mechanical arm, a rotary type goods storing and taking device, a control unit, an identity recognition unit and the like. The space utilization rate of the device is greatly increased, the goods secrecy is improved, modulation assembling and controlling are achieved, and the device is convenient to transport, easy to operate and convenient to maintain; goods can be stored, taken and conveyed through the mechanical arm, the repeated labor sum is reduced, and the storing and taking efficiency is improved. A goods storing mode, a goods taking mode, a fault mode and an initialization mode are involved in the rotary type automatic goods storing and taking method and correspond to the four operation modes or states of goods storing, goods taking, faults and initialization, automation of goods storing, taking and conveying is achieved, human factors are eliminated basically, man-made interference is reduced greatly, the goods storing and taking positions are calculated intelligently, authentication is needed in the storing and taking process, and the accuracy of goods storing and taking is ensured.

The invention relates to an automatic driving vehicle intersection scheduling system and method based on vehicle road cooperation, and belongs to the technical field of automatic driving. The system and method solve the problem about vehicle road cooperation of automatic driving vehicles at intersections. The system comprises a vehicle-mounted communication terminal, a roadside measurement and control communication terminal, an intelligent computing center terminal and a cloud traffic scheduling platform; vehicle running state data transmitted by the vehicle-mounted communication terminal, real-time traffic data collected and transmitted by the roadside measurement and control communication terminal and manual scheduling instructions transmitted by the cloud traffic scheduling platform form a vehicle scheduling information set through data fusion, traffic light timing calculation, lane guiding and lane switching setting conducted by the intelligent computing center terminal, and the vehicle scheduling information set is sent to the vehicle-mounted communication terminal to achieve driving control over the automatic driving vehicles. The system and method increase the intersection road utilization rate to the largest extent, and real-time dynamic high-precision data updating is realized.

The invention discloses an intelligent computing unloading method under a vehicle-mounted network, which combines a deep reinforcement learning algorithm with vehicle edge computing, studies the optimization problem of task scheduling and resource allocation in the vehicle-mounted network, and gives a computing unloading method. On the basis of the limited Markov chain, the intelligent unloading model under the vehicle-mounted network is constructed, and the service experience quality of a user is considered while the profit of a network operator is maximized. According to the invention, the joint optimization problem of initial task scheduling and resource allocation is decomposed into two sub-optimization problems. For the first sub-problem, a utility function is designed to measure theservice experience quality grade of a user, and then a task scheduling algorithm based on a bilateral matching model is provided to solve the sub-problem. Based on an improved deep reinforcement learning algorithm, a reinforcement learning model under a vehicle-mounted network unloading system is constructed, and a resource allocation algorithm is provided to solve a second sub-problem. Experimental results prove the high efficiency of the algorithm provided by the invention.

The invention relates to a modeling and graphical displaying method for a causal relationship reasoning model of unusual working conditions of a chemical process. The method mainly solves the problems that in the prior art, the root causes of the unusual working conditions are not judged accurately and a graphical displaying effect is poor. According to the modeling and graphical displaying method for the causal relationship reasoning model of the unusual working conditions of the chemical process, real-time data of a control system are collected, an expert rule model which is closely related to the judging process of the unusual working conditions is established, a fault tree logical reasoning model for root cause analysis of the unusual working conditions is established according to a root cause analysis mode of the unusual working conditions, an intelligent computing method is adopted, through the fact that threshold value judgment and feature extraction are carried out on key safety technical parameters, feature matching is carried out on the established expert rule model and the established fault tree analysis model, intelligent monitoring and early warning on the unusual working conditions of a production process and root cause analysis are achieved, the problems are solved better through the technical solutions, and the method can be applied to treatment of the unusual working conditions of the chemical process.

The invention discloses a system for calculating a feasible operation range of a BHKW under the condition of ordering power by heat. The system comprises a data collecting and analyzing module, a working condition changing calculating module, a working condition graph module and a peak adjusting load reading module which are sequentially connected, wherein the data collecting and analyzing module reads operation data from a power plant DCS and obtains a representation actual operation benchmark working condition point through the clustering methodology method, and the working condition changing calculating module carries out working condition changing calculation based on the equivalent enthalpy drop method, extracts a working condition changing characteristic according to the optimum square approach method of a continuous function and obtains an actual operating working condition curve of the BHKW. The working condition graph module is obtained according to a working condition changing characteristic curve. The peak adjusting load reading module intelligently calculates the feasible operation range of a peak adjusting load from a working condition graph based on a database. The invention further provides a method for calculating the feasible operation range through the system. The system and method can be used for carrying out operation calculation of a large combined heat and power generation single drawing unit and a large combined heat and power generation double-drawing unit, and a heat supply unit is arranged to adjust the peak, so that thermal power plant and power grid dispatching is greatly facilitated.

The invention discloses an optimal deployment method of a large-scale industrial wireless sensor network based on differential evolution algorithm, which ensures the system reliability through carrying out optimization deployment on nodes, and relates to two fields of industrial wireless sensor network and intelligent computation. The method comprises the following steps of: carrying out automatic coordination on spaces according to actual spaces of industrial sites, obstacles, wireless sensor power and accurate requirement; and using the total number of nodes and minimum load standard difference of cluster heads as targets, wherein a node deployment model is established for restriction conditions based on the redundancy requirement, '1' represents arrangement of the cluster heads corresponding to mesh points, and '0' represents no arrangement. The invention provides a new binary differential evolution algorithm for optimizing and solving the model. By using a new probability prediction operator, the population is updated by a generated binary variation individual. The method can ensure the system reliability, and can reduce the construction cost of the system at the same time, balances the system energy consumption and prolongs the network life cycle through the optimization deployment of the nodes.

A modular umbrella system, comprises one or more shading elements and a support assembly, coupled to the one or more shading elements, to provide support for the one or more shading elements. The modular umbrella system further comprises a base assembly, coupled to the support assembly, to provide contact with a surface. The base assembly comprises one or more microphones to capture audio commands, one or more processors, one or more memory modules and one or more communication transceivers. Computer-readable instructions stored in the one or more memory modules are executed by a processor to convert the captured audio commands into audio files and communicate the captured audio commands to an external computing device utilizing the one or more communication transceivers. The computer-readable instructions are part of an artificial intelligence application programming interface that communicates with an external artificial intelligence computing device to process the communicated audio file.

The invention relates to a real-time searching and combining technology of a mobile end three-dimensional city model based on Internet. The technology realizes a three-dimensional real city scene of shooting and obtaining an image of a mobile terminal including an intelligent cell phone and the like, namely a front face image of an interested city building can be shot by the mobile terminal; a size unchanged characteristic conversion descriptor is extracted and a non-linear encoding matrix is used for encoding and compressing; a server is connected through a wireless Internet and the three-dimensional city model is automatically searched from the Internet; the three-dimensional model which does not meet the requirements is automatically filtered and an optimal model is intelligently calculated; if the related three-dimensional model cannot be searched, the related image is searched; based on an image reconstruction three-dimensional model, the three-dimensional model and attribute information are automatically transferred to the mobile terminal to be overlapped with a two-dimensional map; and therefore, a user can browse and inquire region space information of the user in real time. The real-time searching and combining technology disclosed by the invention has a wide application prospect on the aspects of city navigation, space position service and the like.

The invention relates to a wearable continuous blood pressure estimating system and a method based on dynamic compensation of diastolic blood pressure.A wearable electrocardio-sensing unit and a volume pulse wave sensing unit are used for measuring an electrocardiosignal and a volume pulse wave signal, both of which are connected to a signal acquisition and transmission unit by means of a connection converting unit at the neck portion and synchronously collected and sent wirelessly to a portable intelligent computing and processing unit. The portable intelligent computing and processing unit provides functions of blood pressure calibration and continuous estimation, more specifically, comprises real-time wireless reception of signals, pre-processing, detection of feature points, display and storage and real-time computation and display of systolic blood pressure and diastolic blood pressure.A pulse transit time method is adopted by an estimation of systolic blood pressure. A way of thinking for dynamic compensation based on volume pulse amplitude value and time parameters is utilized by an estimation of diastolic blood pressure. Therefore, estimating accuracy is improved. The wearable continuous blood pressure estimating system is characterized by being small in size and power consumption and low in cost and is suitable for being worn everyday without influencing upon daily activities.

A technical hotline resource management method and system provides a business solution for contact centers which receive contacts both through telephone-based inquiries and web-based inquiries. The disclosed method and system provides, among other things, intelligent calculation of projected resource requirements, a common schedule template for all groups engaged in providing technical assistance, simple input and maintenance screens, a readily-viewed calendar of employee schedules, detailed employee profiles (which identify employee skills), and options for scheduling different work types during given increments of time. Having such information readily available the technical hotline resource manager can thus readily review lists of available employees, groups, and technical requests and can route the inquiries efficiently and quickly. Using the disclosed method and system the resource manager is able to schedule different work types over the contact information, thus enabling employees to be assigned to various work types, including responsibility for different types of inquiries (telephone-based and web-based) as well as for training and meetings, avoiding unnecessary work beyond standard work schedules.

The invention provides a method for intelligently computing the personalized rational medication time. The method comprises the steps of building a medication time related multi-dimensional attribute item dictionary, extracting medication time related element information, carrying out multi-dimensional processing on the medication time related element information, setting priorities of the medication time, extracting medication related element information of a patient, carrying out multi-dimensional processing on the medication related element information of the patient, setting the work and rest time of the patient, generating the rational medication time of a single drug, and generating the rational medication time of whole drugs of the patient. By implementation of the intelligent medication guiding method, system and equipment, massive medication data can be analyzed and processed through an information technology, and an intelligent medication guidance mode is provided.

The invention provides an LED cloud street lamp intelligent control and decision making system based on Internet of Things. The LED cloud street lamp intelligent control and decision making system based on Internet of Things is characterized in that novel efficient energy saving light source LED lamps are used based on a current street lamp control system; an intelligent chip is arranged in each street lamp to name and identify the street lamp; the Internet of Things wireless sensing technology is used to perform data communication between lamps so as to connect the lamps to form a controllable street lamp network, and then the controllable street lamp network is connected with a cloud computing center through a stable Internet of Things communication technology, and automatically sets a optimal illumination scheme by means of background data processing and a control program and according to the factors, such as the local latitude and longitude, the climate, the light and the traffic, so that remote single lamp control, real-time control, adaptive adjusting, early warning of faults and the like can be realized; and the data, such as the environment data, acquired by the street lamps, can be transmitted to the cloud computing platform in the wireless mode, and then an optimal illumination adjusting scheme is generated through intelligent computing analysis and judgment of the acquired data, so that the Illumination As Required adaptive management function can be realized.

The present disclosure relates to a neural network quantization parameter determination method and related products. A board card in the related products includes a memory device, an interface device, a control device, and an artificial intelligence chip, in which the artificial intelligence chip is connected with the memory device, the control device, and the interface device respectively. The memory device is configured to store data, and the interface device is configured to transmit data between the artificial intelligence chip and an external device. The control device is configured to monitor the state of the artificial intelligence chip. The board card can be used to perform an artificial intelligence computation.

The invention provides an automatic production scheduling optimization management information model based on a big data analysis and processing technology. The model based on the construction method comprises a customer requirement information collection module, a customer information processing module, a production plan scheduling module and an inventory classification module. The concrete steps include that customer requirement information is established; data are summarized to a data exchanger center to perform requirement classification processing on the customer information; current inventory rate is intelligently calculated, and corresponding customer order tasks are established; a production plan is issued; production is organized; product offline inventory classification (to be delivered) is performed; and the customer order production and processing tasks and product offline inventory classification are completed and the detailed data are returned to the data exchanger center to be stored, and the corresponding tasks are eliminated. The standby inventory coefficient is intelligently calculated, product demanding requirements of users are predicted and reasonable standby inventory is performed according to the product demanding habit of the users and various other factors so that the production and processing plan meeting the product demanding requirements of the users can be automatically arranged under the premise of guaranteeing supply.

The invention discloses an intelligent computer. The intelligent computer comprises a hybrid neural network application layer, an HnetCP interface layer, a LabGrid middleware layer and a software and hardware resource layer. In the hybrid neural network application layer, a customer realizes a hybrid neural network system and develops a visual hybrid neural network application development environment through an interface of the HnetCP interface layer. The HnetCP interface layer defines various interfaces for operating a hybrid neural network, a top-layer application is an application in the hybrid neural network application layer, and bottom-layer network middleware is the LabGrid middleware layer. The LabGrid middleware layer provides a grid operation environment for the upper-layer application. The software and hardware resource layer is located at the bottom, and software resources comprise kinds of software supporting the upper-layer application. According to the intelligent computer, knowledge stored in the computer and experiential knowledge of people are integrated, and the overall advantages of a computer system are exerted.

The present disclosure relates to a neural network quantization parameter determination method and related products. A board card in the related products includes a memory device, an interface device, a control device, and an artificial intelligence chip, in which the artificial intelligence chip is connected with the memory device, the control device, and the interface device respectively. The memory device is configured to store data, and the interface device is configured to transmit data between the artificial intelligence chip and an external device. The control device is configured to monitor the state of the artificial intelligence chip. The board card can be used to perform an artificial intelligence computation.

The invention discloses a cloud intelligent invoice identification and examination system and method based on a scanner. Paper-based information of an invoice is converted into an electronic image through a scanner. The electronic image of the invoice is subjected to intelligent processing and identification through an Internet cloud intelligent computing center. Intelligent comparison and automatic query are carried out through corresponding content in tax bureau website query systems and tax bureau end invoice databases, an authentication examination result is generated, the result is sent back to an examiner, rapid, simple and accurate query of invoice authenticity is achieved for an examiner, the accuracy reaches 99.99% effectively, a plurality of invoice scanning images can be uploaded at the same time, and the query efficiency is high. The invoice query operation is simplified, the efficiency is raised, the system facilitates supervision of functional departments, illegal behaviors of producing and selling false invoices or the like are restrained effectively, and great loss of the nation is avoided.

The invention provides a mental disorder diagnosis and treatment system based on artificial intelligence. The system comprises a patient end system, a therapist end system and an intelligent auxiliarysystem. The patient end system collects patient information and sends it to the intelligent auxiliary system for communication between the patient and the therapist. The intelligent auxiliary systemestablishes data interaction between the patient end system and the therapist end system, obtains the initial diagnosis result according to the patient information and sends the related data to the patient end system and the therapist end system. In the process of communication between the patient and the therapist, the intelligent auxiliary system recognizes the communication content, collects the patient information for the second time and processes the data collected twice to obtain the secondary diagnosis result and obtains the recommended treatment scheme according to the relevant data and sends the secondary diagnosis result and the recommended treatment scheme to the therapist end system. The therapist end system displays the patient information and the intelligent computing resultto the therapist for the therapist to communicate with the patient and sends the final diagnosis result and the treatment scheme to the patient. The diagnosis and treatment system can reduce the workload of the therapist.

The embodiments of the invention disclose a greenhouse crop planting environment Internet of things intelligent regulation and control system and method. An Internet of things cloud platform receivesa first data packet which includes standardized crop planting environment data and is sent by an agricultural Internet of things gateway, and carries out data processing and data analysis on the standardized crop planting environment data according to a preset intelligent computing model, and determines that a control command is sent to the agricultural Internet of things gateway according to dataprocessing and data analysis results. Each Internet of things control device or a servo subsystem receives the various control commands of the agricultural Internet of things gateway, executes regulation and control motion, and completes the adjustment of corresponding parameters so that each parameter associated with a current crop planting environment tends to be each parameter expected value associated with a preset crop planting environment. In the scheme of the invention, for different planting areas, climates, seasons, crops, and different growers, targeted, intelligent, and personalized intelligent regulation and control can be performed.

Disclosed are an integrated authoring system for an electronic technical manual and a paper technical manual using logistics support analysis data, which is capable of authoring an interactive electronic technical manual (IETM) of Class IV level and a paper technical manual at the same time. The integrated authoring system for an electronic technical manual and a paper technical manual using logistics support analysis data comprises: an iCAST-LSA (intelligent computer based acquisition support tool-logistics support analysis) module; a TMM (Technical Manual Management) module for performing a technical manual management function by importing from the DB of the iCAST-LSA module or exporting to the DB of the iCAST-LSA module; an illustrated drawing manager module; and a TMA (Technical Manual Author) module for performing a manual author manager function for sending and receiving illustrated drawings (ID) or CGM data to and from the illustrated drawing manager module.

The invention discloses a quadrotor autonomous obstacle-avoidance unmanned aerial vehicle (UAV) in an urban community environment. The quadrotor autonomous obstacle-avoidance UAV includes an airbornesystem platform, a ground station platform, a data transmission link, an urban community environment model and an intelligent three-dimensional path planning algorithm module based on the urban community environment model. Aimed at the urban community building environment model, by integrating a binocular CCD, a 2D lidar, an ultrasonic radar, an airborne IMU, a GPS and other perceptual detection,positioning navigation sensors, and through a UAV three-dimensional obstacle-avoidance path planning method based on intelligent computing, the invention designs and implements a quadrotor obstacle-avoidance UAV platform based on binocular vision and laser radar data airborne real-time online processing, which can meet the requirements of autonomous obstacle-avoidance flight of the UAV in the urban community environment space and lays a solid foundation for a wide range of applications of intelligentized UAVs to the anti-terrorism, street fighting, battlefield intelligence detection, parcel delivery, power inspection, disaster inspection, natural environment remote sensing and the like.