47 results about "Local algorithm" patented technology

The invention provides a multi-sensor-based autonomous obstacle avoidance navigation system. The multi-sensor-based autonomous obstacle avoidance navigation system comprises a sensor processing layer, a local algorithm processing layer, an error detection layer and a central processing layer, wherein the sensor processing layer comprises a plurality of sensor processing nodes; the local algorithm processing layer comprises a plurality of algorithm calculation nodes; the error detection layer is used for detecting hardware and software faults of a detection system and shielding nodes with faults when determining that the sensor processing nodes and/or the algorithm calculation nodes have faults; and the central processing layer is used for carrying out overall optimization on results of the plurality of algorithm calculation nodes in the local algorithm processing layer. The multi-sensor-based autonomous obstacle avoidance navigation system adopts a non-centralized type fusion design so that the robustness and the fault tolerance of the system are greatly improved.

The invention relates to a computer binocular vision matching method based on global and local algorithms, which comprises the following steps: (1) obtaining a layered matching result from a left image and a right image which are obtained by a binocular three-dimensional vision sensor by using an image cutting method; (2) obtaining different local matching results in a local matching result by using a window matching method with different windows; (3) putting the local matching results into a markov random field with a second-order smooth energy function for iterating and optimizing to obtain the matching result. The computer binocular vision matching method can integrate global optimized and local characteristics and obtain a more accurate matching result.

The invention provides a gray level fluctuation threshold segmentation method of an image with non-uniform illumination. Compared with the existing local algorithm, the method regards the image as a three dimensional terrain of the gray level, the non-uniform illumination is equivalent to changing of the terrain landform, the grey value fluctuates up or down within a local area, thus a larger-scale crest point and trough in each gray scale fluctuation curve are searched along the horizontal and vertical directions in an iterative way, a reasonable threshold between each pair of crest point and trough point is calculated to realize the division of image goal pixels and background pixels, and finally, threshold images obtained in the two directions are subjected to integration to obtain a final segmentation image. By applying the method, the influence of non-uniform illumination can be effectively reduced, the instantaneity of the algorithm is improved, and an effective pretreatment technology base for target detection of the image under a complex background is provided.

The invention provides a leakage positioning method for urban non-metallic pipelines based on an inverse transient model, which uses Fluent software to simulate the pressure, flow rate, velocity distribution and other parameters of the non-metallic pipeline in the leakage state and the leakage-free state, and finds out the regularity of each parameter varying with time and space under the two states; the leakage simulation tests of non-metallic buried pipelines are carried out in laboratory, and the flow rate, velocity and pressure parameters of the pipelines are measured under leakage and non-leakage conditions respectively. The inverse transient model of non-metallic pipeline leakage state is established and the experimental data are analyzed according to the variation law of parametersobtained by simulation software. By using global and local algorithms, the best convergence target parameters Jk and tau k are determined, and the pressure measuring head calculated by the transient solver is obtained. The calculated pressure measuring head and the experimental pressure measuring head are substituted into the inverse transient model to determine the size and positioning of leakage. Compared with the prior art, the method only needs to measure pipeline pressure and flow parameters, and is easy to realize.

The invention discloses a region scale soil erosion modeling-oriented terrain factor parallel computing method and belongs to the technical field of computers and information. The method comprises the following steps: (1) determining a data partitioning strategy, and performing equal-row partitioning on original terrain data by a row buffer zone partitioning method according to the number of parallel nodes, wherein each sub data block comprises three elements of a data zone, a buffer zone and a control point; (2) performing slope parallel computing, water flow direction computing, flow accumulation computing, river network extraction computing, slope length computing and slope-slope length factor computing based on preprocessed DEM (digital elevation model) data according to the data partitioning strategy. According to the method, the computing characteristics of a serial algorithm are fully considered in the design process of a parallel algorithm; different parallel strategies are designed for a local algorithm and a global algorithm respectively; the parallel efficiency of the algorithms is effectively improved; terrain factors based on mass terrain data can be quickly and accurately computed.

The invention provides a visual logic editing runtime engine. The engine comprises an algorithm library, an algorithm library management module, a logic runtime library, and an editing interface, wherein the algorithm library provides algorithms for communication connection, protocol analysis, data processing, parameter exporting, device control and data display; the algorithm library management module is used for automatically managing a local algorithm library, and comprises algorithms provided by the visual logic editing runtime engine and an algorithm for application system expansion; the logic runtime library performs an operational instruction of an application system; and the editing interface is used for realizing field editing of algorithms in the logic runtime library and algorithm association relationships, and providing a visual graphic interface for editing a logic process. The visual logic editing runtime engine provides a visual business process and an object logic editing function, so that what you see is what you get, and the engine is visual and easy to use; and the engine provides an instant running function, so that users can modify processes or visual effects and logics of an existing system without the need for writing codes, and the change can be immediately showed in the visual application system.

The invention relates to the field of production line scheduling, in particular to a production line risk scheduling method for sudden large faults of equipment. According to the method, the influence of sudden large faults of the production line equipment is considered during the evaluation process of the production line equipment fault risk, and the health state of the sudden large faults is introduced, so that the adopted risk evaluation scheme can more accurately evaluate the production line equipment risk. By establishing a production line equipment health state model, the weight of the sudden large fault state of the production line equipment in the whole production line equipment health state is determined, and the production line equipment risk is comprehensively evaluated through the linear weighting; an improved differential local algorithm maintains the advantages of the original search method, increases the diversity of individuals, and effectively prevents the search from falling into a local optimal solution. According to the present invention, the risks caused by the sudden large faults of the production line equipment can be considered, and the multi-target risk scheduling of the production line is achieved.

The invention belongs to the technical field of intelligent robots, and discloses a mobile laundry assistant robot based on image scanning, recognition and route planning. The initial overall route ofthe robot is planned out through the improved ant colony algorithm; when the robot meets obstacles on the way, the Morphin local algorithm is called, and the robot avoids obstacles and runs to the front overall route; and if the Morphin local algorithm cannot find out a feasible route, the improved ant colony algorithm is called to replan the route, and remote control and voice control over the robot are achieved. According to the mobile laundry assistant robot based on image scanning, recognition and route planning, time and manpower are saved, and the situation that clothes are collected and put to be washed manually with low efficiency is avoided; and the function of remotely controlling the robot to collect clothes and using washing machines for washing can be achieved. Washing machines can be used in a staggered manner, and meanwhile people disabled in action or the disabled can be helped to put clothes into washing machines.

The invention provides a face snapshot, screening and storage method and system for real estate transactions, the system comprises a device end, a local end and a cloud end, the device end collects and snapshots face data of a visiting customer and cuts a video stream, the local end and the cloud end respectively store a local algorithm and a cloud end algorithm for screening, processing and storing snapshot images, and carrying out duplicate removal processing on the retrieved data according to a preset point location and preset time, wherein uploading and storage of the data are processed in a multi-thread mode. According to the method, the problems that customers are difficult to track, data management is disordered and the like when the visitor flow rate of a real estate sales site is excessive are solved, accurate positioning of the customers is realized based on face snapshot, disordered and redundant data stored by the customers are avoided, the workload of workers is reduced, and statistics and analysis of the customers are facilitated.

The invention relates to an oil field management system. The oil-field management system comprises an oil well management module and a cloud-end management module, wherein the oil well management module is abstracted into corresponding Web services, Web services of single oil well are combined to form an interoperation model of the single oil well; the interoperation model comprises a data acquisition unit for acquiring oil well data, a data transmission unit for storing and transmitting the oil-well data to a cloud-end management module and updating a local algorithm library according to an optimized control strategy, and a control unit for adjusting a production strategy of the oil well in real time according to the updated local algorithm library; and the cloud-end management module comprises a data management unit for receiving and storing the oil-well data sent from the interoperation model, and a strategy computing unit for calculating an optimization strategy of the oil well according to the received oil well data and sending the optimization strategy to the interoperation model. According to the oil field management system disclosed by the application, the purposes of controlling the oil well production and adjusting the oil-recovery strategy can be achieved and the implementation is simple and efficient.

The invention discloses an AGV path planning method based on an improved DWA algorithm in a dynamic environment, and belongs to the technical field of AGV path planning. According to the method, firstly, a local DWA algorithm is designed by obtaining local map information around the AGV, dynamic obstacles in a local map are classified, the speed of the dynamic obstacles is evaluated, the obstacle avoidance judgment capability of the AGV on the dynamic obstacles is improved, and the degree that the AGV deviates from the shortest path is reduced; secondly, aiming at the problem that the direction needs to be adjusted when the DWA algorithm moves to the path point, the adjustment time of the AGV at the path point is shortened by optimizing a DWA evaluation function, and the transportation efficiency of the AGV is improved by optimizing the inflection point of the global path of the AGV through the improved DWA algorithm; and finally, simulation verification is carried out, and a simulation result shows that the improved DWA algorithm provided by the invention can ensure real-time obstacle avoidance of the AGV in an environment with a complex dynamic condition, and meanwhile, the transportation time of the AGV is shortened, and the transportation efficiency is improved.

The invention discloses an approach judgment method, device and apparatus aand a storage medium. The approach judgment method comprises the following steps: firstly, acquiring signal intensity and acquisition time of a plurality of acquisition points of a current user terminal; processing the signal intensity and the acquisition time of the plurality of acquisition points based on an action behavior model to obtain action behavior parameters; and finally, judging the approaching behavior or leaving behavior of the current user terminal based on the action behavior parameters. The algorithm adopted by the method is relatively simple, and the data processing amount is small. The algorithm is a local algorithm and does not need a network. The identity of a visitor can be automatically judgedby acquiring the wireless signal physical ID of the user terminal, for example, the access frequency of the user terminal is determined through address information, whether the visitor is a return customer or a new customer is determined, and the access frequency of the return customer does not need to be judged by people. Generally speaking, the response is quick, the delay is short, and the userexperience is good.

The invention discloses a distributed vector processing method, device and system and a medium, the system comprises at least two clusters, each cluster comprises a plurality of nodes, and the number of the nodes in each cluster in the at least two clusters is the same. In each cluster, the plurality of nodes comprise an aggregation node and a computing node. Specifically, the aggregation node in each cluster is used for managing the data in the cluster where the aggregation node is located, and the computing node is used for storing and computing the data of at least one partition. According to the scheme provided by the invention, each computing node in each cluster can store and compute the data in parallel, and a local algorithm can be quickly transplanted without a distributed foundation, so that distributed computing is realized, and the development efficiency of a federated algorithm can be accelerated.

The invention discloses a basic algorithm architecture supporting distributed parallel computing and a computing method, and the basic algorithm architecture comprises a local algorithm library which is arranged in an application computer and comprises a plurality of programming language algorithm libraries; the CPU equipment node agent program is in communication connection with the CPU computing node of the application computer; the GPU equipment node agent program is in communication connection with the GPU computing node of the application computer; the FPGA equipment node agent program is in communication connection with the FPGA computing node of the application computer; the virtual computer resource management program is in communication connection with the local algorithm library, the GPU equipment node agent program, the CPU equipment node agent program and the FPGA equipment node agent program, and the virtual computer resource management program is used for selecting different equipment node agent programs for calculation according to the complexity of the calculation task of the application program. The system can support multiple programming languages, not only supports a CPU, but also can support GPU and FPGA computing platforms, and is high in computing efficiency.

An embodiment of the disclosure relates to a method for providing an aggregate algorithm for processing medical data and a method for processing the medical data. A multitude of local algorithms is trained by machine learning, wherein the training of each respective local algorithm is performed on a respective local system using respective local training data, wherein a respective algorithm dataset concerning the respective local algorithm is transferred to an aggregating system that generates the aggregate algorithm based on the algorithm datasets.