80results about How to "Realize data fusion" patented technology

The invention discloses a mobile monitoring fusion platform based on a geographic information system (GIS), which belongs to the technical field of network communication. The platform comprises a monitoring center and a mobile monitoring station which are connected with each other in a wireless mode, wherein the mobile monitoring station comprises a monitoring unit, a moving posture sensor, a data acquisition plate, a data analysis module, a synchronous encoding module, a storage module, a global positioning system (GPS) module, a data processing module, a remote communication module and a central controller; the monitoring center comprises a remote communication module, a synchronous decoder, a monitoring databank, a data fusion module, a GIS geographic information system, a display module and a man-machine interaction control module. Monitoring data is transmitted to the monitoring center in the wireless mode after being acquired and treated, and the monitoring center receives and decodes the monitoring data and subsequently performs GIS fusion. The platform provides effective and visible services such as multi-source monitoring data fusion, storage, management and transferring, supports geographic information demonstration and live broadcast, and is applicable to fields of monitoring on ecologic environment, geological disaster areas and other emergency accidents.

The invention discloses a target identification method based on the data fusion of an airborne radar and an infrared imaging sensor, comprising the following steps: firstly, identifying a target and generating a corresponding elementary probability partition function M1 by the infrared imaging sensor; then identifying the target and generating a corresponding elementary probability partition function M2 by the airborne radar; fusing the identification results of the airborne radar and the infrared imaging sensor by a D-S evidence theory; finally, identifying the target, and making decisions. In the invention, the airborne radar adopts the target identification method based on fuzzy reasoning to identify the target, thereby overcoming the defect of the high nondeterminacy of data measured by a radar sensor and finally realizing the data fusion of the airborne radar and the infrared imaging sensor. A simulation result shows that the identification effect of an intelligent target algorithm based on the airborne radar and the infrared imaging sensor is much better than that of single radar or single infrared sensor.

The invention provides a mobile robot autonomous following system and method based on multi-sensor fusion and relates to the technical field of industrial automation. The system comprises an upper navigation unit, a bottom motion control unit and a power supply unit, wherein the upper navigation unit acquires positioning information about a target person by a sensor, plans and calculates tracks ofmotion of a robot towards the target person, selects a planned track with the optimal local path and sends a control command to the bottom motion control unit; the bottom motion control unit controlsthe robot to move towards the target person according to the control command of the upper navigation unit; the power supply unit supplies power to the whole system. Meanwhile, the invention further provides the specific mobile robot autonomous following method based on the system. By means of the mobile robot autonomous following system and method based on multi-sensor fusion, the mobile robot can realize a stable autonomous person following function in a dynamic shielding environment including the condition that the target person is shielded by an obstacle.

The embodiment of the invention discloses a data interworking method and data interworking equipment. The method is realized in the following steps: receiving first data and second data, wherein the first data comprises the first user identifier and the first data of a first platform, and the second data comprises the second user identifier and the second data of a second platform; if the first user identifier and the second user identifier have an association relationship, calcuating to obtain result data on the basis of the first data and the second data; and storing the result data and a uniform user identifier, wherein the uniform user identifier is a user identifier which has the association relationship with the first user identifier and the second user identifier. The embodiment of the invention can realize the data interworking of associated users under the account number systems of different platforms so as to realize data fusion.

The invention provides a method and device for automatic parking control based on data fusion. The method comprises the steps that (S1) a parking request is received, a current car speed is obtained, and whether the current car speed is lower than a preset car speed thread is judged, wherein a parking control system will be started if the current car speed is lower than the threshold; (S2) the angle of a parking camera is adjusted to a first preset angle, and a parking radar is started; (S3) a parking space is detected through methods based on image identification method and obstacle detection respectively; (S4) when the parking space is found first by the method based on the image identification, the next step will be executed, and otherwise the step S7 will be executed; (S5) a position LR of a right edge of a marking line of the parking space relative to the parking radar is computed; (S6) whether LR is bigger than 0 is judged, wherein the next step will be executed if LR is bigger than 0, and otherwise the step will be executed cyclically; (S7) the system is switched to a parking space type identification mode based on the obstacle detection; and (S8) a parking route is planned according to the method based on the obstacle detection. In this way, the automatic parking system based on radar detection can carry out parking according to information of the camera.

The invention provides a cross-domain knowledge graph construction method and device based on artificial intelligence, and relates to the field of artificial intelligence and cross-domain knowledge graph construction, and the device comprises a knowledge model which is used for describing a concept hierarchy system and is a concept template of knowledge in a knowledge base; a multi-source heterogeneity, data source management which supports multi-source heterogeneity, and comprises management of relational data sources, distributed Hdfs, Hbase, Hive and mongoDB data sources; knowledge data extraction, so that a user carries out data extraction on data of different data source types to form a series of high-quality fact expressions, and a foundation is laid for construction of an upper-layer mode layer; and cross-domain ontology fusion, and cross-domain ontology fusion can be carried out according to data of various industry fields, so that high quality of atlas data is always kept. Andmore accurate and reliable situation description in a fusion scene is obtained.

The invention discloses a steam turbine rotor fault diagnosis method based on LSTM, and belongs to the technical field of mechanical fault diagnosis. Firstly, multi-point acquisition sensors are deployed and controlled, and vibration signals of various typical turbine rotor faults are collected as a training set and a verification set. Secondly, the steam turbine rotor vibration signals are extracted from a power plant DCS system to serve as a testing set. Thirdly, the training set, the testing set and the verification set realize fusion of multi-point signal data and data enhancement throughsignal division, stacking and other operations. Fourthly, a neural network based on the LSTM is constructed, the training set and the verification set are used for completing training of the network,and finally, maintenance of a diagnostic model is achieved in cooperation with an actual diagnostic task, and finally the steam turbine rotor fault diagnosis is realized on the testing set.

The invention discloses an information processing method based on an environmental benefit monitoring sensing network, and a system and a device both adopting the information processing method. The information processing method is characterized by comprising the steps as follows: S1, a central node and a network switching node or a data transceiver module of a monitoring server can be in a continuous or intermittent monitoring mode according to preset programs or manual input commands or the commands that are transmitted through a network device or an active or passive wake-up circuit after the central node and the network switching node of a background processing system are installed or called or the monitoring server starts up, and then data transmitted by a monitoring unit on a wireless sensor network acquisition node is collected; and meanwhile, in S2, the central node and the network switching node or the monitoring server can transmit the commands to the wireless sensor network acquisition node when the background processing system sends the commands according to the preset programs or the manual input commands or the commands that are transmitted through the network device or the active or passive wake-up circuit. The system and the device can direct to open-ended commands and services. The technology is suitable for being applied to WSN (Wireless Sensor Network) projects such as Sensing China, Smarter Planet and the like.

The invention provides a ship-borne ship draught automatic detection system which comprises a ship-borne detection device and a remote management system.The ship-borne detection device is arranged on a detection ship and used for obtaining draught data of a ship to be detected.The ship-borne detection device comprises a high-density sonar array, a ship interval measuring device, a ship positioning device, an automatic ship recognizing device, a video evidence obtaining device and a data collecting device.According to the ship-borne ship draught automatic detection system, on the basis of the ship draught detection technology of an underwater sonar depth measuring system, active, digital and automatic ship draught detection is achieved, the requirement for manpower and material resources is effectively lowered, and the detection efficiency and precision are improved.It is ensured that shipping lanes are unblocked, the situation that sailing is blocked is avoided, normal and smooth industrial and agricultural production is ensured, and reliable material guarantees are provided for economic development.

The invention provides a data fusion method and system. The method comprises the steps of receiving a first to-be-fused data set sent by a first data processing system, wherein the first to-be-fused data set comprises a first content and an encrypted first keyword; receiving a second to-be-fused data set sent by a second data processing system, wherein the second to-be-fused data comprise a secondcontent and an encrypted second keyword, and the second keyword and the first keyword are same keywords respectively stored in the second data processing system and the first data processing system,and the encrypted second keyword and the encrypted first keyword are same encrypted keywords obtained by encrypting using a same encryption manner; and according to the same encrypted keywords, fusingthe first fusion data set and the second fusion data set to obtain a fused data set. According to the data fusion method and system, the security is improved.

The invention provides a signal processing platform applicable to a multi-sensor self-adaption optical system. The signal processing platform comprises a plurality of signal processing modules (1) with a bus switch, an interconnection rear panel (2) based on a CPCI (Compact Peripheral Component Interconnect) bus structure, an expansion I / O (Input / Output) board (3) and a monitoring host (4). With the adoption of the structure, the signal processing platform provided by the invention can be used for processing wave-front signals of a plurality of sensors in parallel and combining low-delay data among the sensors in parallel; the good expandability, the generality, the clipping property and the small size of the signal processing platform can be realized due to a dynamic bus structure with a switch and a changeable operation signal processing module structure; and the signal processing platform is applicable to processing the self-adaption optical wave-front signals in various applicable fields.

The invention discloses a double-filter fusion positioning method and system for park automatic driving, and aims to design the double-filter fusion positioning method and system for park automatic driving, so that by fusing a vehicle kinematics model and IMU integration through an extended Kalman filtering technology, a predicted pose of a vehicle is obtained. A particle filtering technology is used to fuse a vehicle prediction pose and a laser odometer so as to improve positioning precision and robustness of an automatic driving vehicle in a park environment.

The invention provides a ship draught detection method. The method includes the following steps of S1 obtaining the measurement frame data of water surface, side underwater part of a ship to be detected, and the riverbed; S2 obtaining the horizontal distance and height value of each wave beam corresponding detected point relative to a sonar source; S3 obtaining part of the measurement data groups of the side underwater part of the ship to be detected; S4 calculating the measurement data groups to obtain the draught measured value of the ship to be detected; and S5 performing data correction on the draught measured value obtained from S4 to obtain the draught value data of the ship to be detected in the measurement frame. Based on the ship draught detection technology of an underwater sonar sounding system, the ship draught detection method realizes active, digital and automatic ship draught detection, thereby effectively reducing the manpower and material requirements and improving the detection efficiency and precision. According to the invention, smooth channel is guaranteed, the situation of obstructing the navigation can be prevented, the normal and smooth operation of industrial and agricultural production is guaranteed, and reliable material guarantee is provided for economic development.

The invention provides a wireless sensor network data fusion method used for optimizing an RBF neural network based on a firefly algorithm. The method is characterized in that each sensor node carries out preliminary processing on collected data according to an input layer neuron function so as to obtain a small amount of characteristic data, and then sends the characteristic data to a cluster head node of a cluster where the sensor node is located; according to a hidden layer neuron function and an output layer neuron function, the characteristic data is further processed so as to obtain the fused data; and the cluster head node sends the fused data to an aggregation node. In the invention, a transmission quantity of information in the network can be effectively reduced, power consumption of the sensor node can be decreased and simultaneously a network communication bandwidth is saved.

A method for estimating the state of charge (SOC) of a battery through Kalman filtering is technologically characterized in that the SOC of a lithium battery pack is estimated by an extended Kalman filter (EKF) algorithm, a Vmin state space model of the lithium battery pack is built, and the minimum Vmin of the load voltage of the single batteries in the battery pack and the SOC of the battery pack as used as an observed variable and a state variable of the model. The adopted Kalman filter is an extended Kalman filter, SOC is worked out recursively by use of an ampere-hour integral method, SOC is substituted in an observation equation to get the estimated value of Vmin, the Kalman gain for every step is calculated, and the optimal estimate of SOC is worked out based on a state estimation observation renewal equation. The power battery SOC estimation method overcomes the defects of current error accumulation based on the ampere-hour integral method, and closed-loop estimation of state variable SOC is realized. As the effect of noise is taken into consideration in the process of calculation, the algorithm has a strong noise inhibition effect.

The invention discloses a fuzzy fusion-based multi-sensor crack damage comprehensive diagnosis method. The method comprises the steps of A, collecting a fiber grating spectrum, a Lamb signal and an intelligent coating resistance value; B, extracting characteristic parameters; C, respectively performing fuzzification processing on characteristic parameters of signals collected by fiber grating, piezoelectric and intelligent coating sensors by use of a Gaussian membership function to obtain three membership functions; D, performing fuzzy fusion by use of a Jaeger algorithm to obtain a comprehensive membership function and a fusion factor omega; E, defuzzifying the comprehensive membership function obtained by fuzzy fusion to obtain fused characteristic parameters; and F, performing relationfitting processing on data obtained by fuzzification processing by use of the fusion factor and fiber grating, piezoelectric and intelligent coating sensor data, and obtaining a relation between the fused characteristic parameters and a damage amount by prediction and verification. According to the fuzzy fusion-based multi-sensor crack damage comprehensive diagnosis method, the problem that information judgment and comprehensive diagnosis cannot be directly performed by use of the multi-sensor data at present can be effectively solved.

The invention discloses a method for obtaining mobile phone signaling track data with a label based on resident survey data. The method comprises the following steps: S1, obtaining basic attribute information of the mobile phone signaling track data and resident travel survey data; S2, converting the mobile phone travel OD into a traffic cell travel OD based on the spatial correspondence between the mobile phone base station and the traffic cell; S3, extracting feature sets of mobile phone signaling trajectory data and resident travel survey data for different dates, and performing preliminaryfusion on the data in time latitude; S4, based on the travel characteristic attribute of each part of resident travel survey data, performing preliminary screening matching on the mobile phone signaling track data under the set matching rule; and S5, in combination with the personal characteristic attributes of the traveler and the mobile phone user, completing final characteristic matching of the resident travel survey data and the mobile phone signaling track data, and obtaining the mobile phone signaling track data with the label. Data fusion of a mobile phone signaling data large sample and a resident travel survey data small sample is realized.

The invention discloses ship attitude measurement equipment and a ship attitude measurement method based on a Beidou system. The measurement equipment comprises a radio frequency interface module, twoBeidou double-antenna direction-finding receivers, an attitude measurement module, a data storage module and a communication interface module. Compared with the prior art, the Beidou dual-antenna direction-finding receiver board card has the advantages that an existing Beidou dual-antenna direction-finding receiver board card is used for measurement, the relative baseline length and direction between the two antennas can be accurately calculated through the mutual difference technology of the board card, and therefore dependence on a differential reference station is eliminated. Two Beidou double-antenna direction-finding receiver board cards are used for jointly measuring a public Beidou satellite navigation antenna, mutual connection of measurement data of the two Beidou double-antennadirection-finding receiver board cards and unification of measurement coordinate values are established, and therefore data fusion of the two Beidou double-antenna direction-finding receiver board cards is achieved.

The present invention discloses an ecological environment monitoring method based on an unmanned aerial vehicle, which an unmanned aerial vehicle with ultra-long endurance, a small unmanned aerial vehicle with long endurance, a mission payload system, a wireless link and a ground control station. The method is as follows: a working mode of the unmanned aerial vehicle with ultra-long endurance is selected based on the requirement that a target area need be uninterruptedly monitored for an ultra-long time; a working mode of the small unmanned aerial vehicle with long endurance is selected basedon the requirement for rapid response monitoring of the earth environment; and a cooperative working mode of the unmanned aerial vehicle with ultra-long endurance and the small unmanned aerial vehiclewith long endurance is selected based on the requirement of wide-area long-term monitoring of the earth environment and rapid response monitoring in key areas or emergency areas. The invention effectively makes up for the existing deficiencies such as low data credibility, slow update speed and overdue space-time reactions and responses under all-weather and all-weather conditions.

The invention discloses a gridding automatic monitoring system integrating augmented reality and geographic information. The gridding automatic monitoring system comprises rear-end intelligent software and front-end sensing equipment (an intelligent lamp post and the like). The rear-end intelligent software mainly refers to a gridding automatic monitoring system integrating geographic informationand augmented reality. The front-end sensing equipment (intelligent lamp poles) is installed in different areas of a park and used for video security, miniature meteorological stations, information publishing, broadcasting and the like, the front-end sensing equipment (intelligent lamp poles) sends all data information to the management server, and the management server provides a function of remotely calling data for a user side. The gridding automatic monitoring system integrating geographic information and augmented reality has the functions of multi-parameter fusion early warning, in-parkvehicle monitoring and on-site video transmission. The system can be used remotely through a public wide area network and can also be used through a local area network, and a powerful guarantee is provided for lack of a public network environment in an emergency occasion. The gridding automatic monitoring system integrating geographic information and augmented reality can be applied to industrialparks of petroleum, chemical engineering and the like; according to the functions of automatic patrol, electronic boundary, forbidden zone alarm, unattended operation and the like, potential safety hazards can be rapidly recognized and automatic early warning can be achieved under the unattended operation condition, and powerful information support is provided for safety management and emergency rescue.

The invention discloses a heterogeneous agricultural Internet of Things interconnection and interworking method. The method comprises steps that (1), heterogeneous network interface communication is realized through utilizing a loosely-coupled communication model taking an improved system base station as a middleware; (2), the XML technology is utilized to realize mutual understanding of data on the basis of heterogeneous network communication; and (3), through the decoupling Kalman filtering algorithm based on the stable gain, the received large-amount redundant data is integrated, and data accuracy is improved. The method is advantaged in that the shortcoming that heterogeneous networks of existing agricultural equipment cannot be integrated with each other in the prior art is overcome,heterogeneous data of the heterogeneous Internet of Things are integrated and filtered, and overall efficiency of the network is improved.

Embodiments of the invention disclose a data processing method and apparatus, which is used for performing secrecy processing on data used for user portrait drawing to facilitate realization of the user portrait drawing. The method comprises the steps of obtaining business system data; converting the business system data to obtain converted data, wherein the converted data is data having secrecy performance and capable of being used for performing the user portrait drawing; and sending the converted data to a portrait drawing platform, thereby enabling the portrait drawing platform to performthe user portrait drawing according to the converted data. After the business system data of a business system data party is subjected to the secrecy processing, the information leakage of sensitive data of the business system data can be avoided, and the converted data still can be used for the user portrait drawing; and in the portrait drawing platform, the user portrait drawing can be performedaccording to the converted data, and the fusion of the data of the business system data party and data of a portrait drawing service provider is facilitated, thereby realizing the user portrait drawing.

The invention relates to a portable industrial robot based on multi-dimensional sensor real-time data feedback fusion. The portable industrial robot comprises an industrial robot and a multi-dimensional sensor real-time feedback control system. The portable industrial robot has the beneficial effects of being based on multi-dimensional sensor real-time feedback and data fusion, the structure and the motion error compensation are completed, and accurate control which does not depend on the structure precision of the industrial robot is achieved; an industrial robot and a sensor real-time feedback control system generate accurate real-time control data through the real-time feedback data and data fusion technology to form a whole set of system, various working conditions can be realized through the single robot, the working conditions of the industrial robot can be adjusted in real time, and the working condition precision is ensured.

The embodiment of the invention discloses a lane recognition system and method and an autonomous car. The lane recognition system comprises at least one camera, at least one radar, a data processing unit and a lane recognition unit. The lane recognition unit is connected with the data processing unit and is used for confirming the weight proportion of the road edge features obtained by processingimage information to the road edge features obtained by processing radar information, and the lane position is recognized according to the weight proportion, the road edge features obtained by processing the image information, the road edge features obtained by processing the radar information and lane line features. According to the technical scheme, data fusion of camera and radar equipment on lane recognition is achieved, and therefore the system can handle with complex road conditions and environment interference better, dangerous accidents are avoided, and driving safety is improved.

The invention discloses a six-degree-of-freedom wideband integrated seismic monitoring system which at least comprises a sensor integration unit and a data fusion unit. The sensor integration unit comprises a GNSS receiver, a three-axis accelerator and a three-axis gyroscope which are arranged at a same sit and used for synchronously collecting GNSS observation data, acceleration observation dataand angular speed observation data and storing the same, the data fusion unit comprises a GNSS data positioning and processing module, an integrating, fusing and measuring data module and a comprehensive filter, the GNSS data positioning and processing module is used for positioning calculation according to the GNSS observation data, the integrating, fusing and measuring data module is used for fusing GNSS position information, the acceleration observation data and the angular speed observation data, and the comprehensive filter is used for calculating according to data after fusion. The system can realize single-station six-degree-of-freedom seismic monitoring, expand seismic monitoring bandwidth and improve seismic monitoring ability.

The invention provides a camera and laser radar synchronization method and system of an intelligent driving control system, and the method comprises the steps: calculating the synchronization delay time of the exposure time of the camera according to the frame rate and the installation angle of the camera, and enabling the exposure time to be just the same as the time when the laser radar scans the angle; at the moment, laser radar data and camera data can be subjected to fusion processing, and accurate time and space information is obtained.

The invention relates to a vehicle-mounted data acquisition device capable of extending ports and a data acquisition method. Two HDMI interfaces that are an upper chain port and a lower chain port arearranged on each vehicle-mounted acquisition instrument; a number of data acquisition instruments sequentially connect the lower chain ports and the upper chain ports of the adjacent data acquisitioninstruments through HDMI cables, and thus the number of data acquisition instruments are connected in series; multiple acquisition instruments automatically determine that a device that is not connected to the upper chain port is main equipment and other devices are slave equipment; the main equipment sends configuration information to the slave equipment through the HDMI interfaces and initiatestime synchronization instructions to the slave equipment; each slave equipment performs time synchronization based on the time of the main equipment; and after receiving a data acquisition start instruction, the main equipment sends the instruction to the slave equipment, the main and slave equipment synchronously acquires data, and the main equipment merges the received data of all slave equipment with the own acquired data and synchronously stores the merged data to a file. The scheme of the invention is simple in structure, and data acquisition and later data playback analysis can be performed conveniently and quickly.

The invention relates to a technology for applying all-digital technology integration platforms to digital communities. Various digital subsystems can be remotely controlled and can be connected with one another on the basis of the all-digital technology integration platforms comprising hardware platforms and software platforms, resources can be shared by the different subsystems by means of data management, and information can be integrated among the different subsystems by means of data management; the subsystems are connected with one another, can be in linkage with one another under abnormal conditions and can quickly respond. The various subsystems run independently, the integral performance of other subsystems and the platforms can be prevented from being affected when problems of a certain subsystem occur, and accordingly the reliability of the platforms can be guaranteed. The all-digital technology integration platforms can be conveniently and quickly managed in a centralized manner in the aspects including addition, modification, deletion, monitoring and configuration information management and safety, permission, round-tour scheme, video plan, warning linkage or defense organization and cancelation strategy management. The technology has the advantages of advanced system architecture, large-scale expansion capacity, innovative structured storage, global resource management and the like.

The invention provides an abnormal data processing method based on regulation and control matching data fusion. The abnormal data processing method comprises the steps of obtaining a system model andoperation data of a regulation and control cloud platform and a marketing, distribution and dispatching integrated platform; determining an association relationship between the regulation and controlcloud platform and a system model of the marketing, distribution and dispatching integrated platform based on a pre-trained model object; and performing data feature extraction and abnormal data processing on the operation data based on an association relationship of a system model and a pre-trained neural network model. The defect that the traditional cross-professional information sharing degreeis not enough is overcome, and meanwhile, the quality of data governance is improved by adopting a power grid data feature extraction and exception handling technology.

The invention relates to the technical field of distribution boxes, and particularly relates to a high-protection modular distribution box. The high-protection modular distribution box comprises a distribution box, a display module, an acquisition module, a security monitoring terminal, an alarm module, a data analysis module, a network server, a storage module, a database, a water level monitoring module, a feedback module, a distribution transformer and an infrared temperature measurement module, wherein the distribution box is connected with the display module, the distribution box is connected with the acquisition module, the acquisition module is connected with the security monitoring terminal, the security monitoring terminal is connected with the alarm module, the alarm module is connected with the distribution box, the distribution box is connected with the data analysis module, the distribution box is connected with the network server, the distribution box is connected with the storage module, and the storage module is connected with the database. The high-protection modular distribution box provided by the invention has the advantages of convenience in operation and maintenance information state analysis, monitoring and early warning, fault research and judgment, auxiliary decision, risk management and control and intelligent resource allocation.