100 results about "Parallel search" patented technology

The invention discloses an intelligent-based multi-task self-adaptive scheduling system for a pension accompanying robot, and belongs to the technical field of multi-task self-adaptive scheduling. Comprising a multi-modal task perception fusion module, a dynamic task evaluation module, a dynamic task priority generation module, a quantum causal reasoning module, a quantum optimization task scheduling module, a multi-task collaborative optimization module and a man-machine collaborative interaction module. Task features are mapped to a multi-dimensional weight space, a time attenuation coefficient is introduced to dynamically adjust the influence of historical causal association, a dynamic adjustment mechanism ensures timely execution of high-priority tasks, a quantum superposition state is utilized to parallelly search a global optimal solution through a quantum optimization task scheduling module, and the task scheduling efficiency is improved. Through a dynamic priority queue and a conflict solution feedback mechanism, the scheduling scheme is iteratively optimized, and the feasibility and the resource utilization rate of the scheduling scheme are remarkably improved.

Systems and methods for machine learning-informed response and augmentation of incomplete queries and query artifacts that include executing parallel search engines in a machine learning pipeline based on an input of a query to retrieve one or more corpora of candidate digital artifacts, constructing a merged corpus of candidate digital artifacts based on the retrieved corpora of digital artifacts, ranking each candidate digital artifact of the merged corpus of candidate digital artifacts based on a computed relevance to the input query, and returning, via a user interface, a response to the input query based on a subset of prioritized candidate digital artifacts most relevant to the input query.

The invention discloses a competitive evolution multi-task optimization method, system and equipment, and aims to solve the problem that an existing method is difficult to carry out collaborative optimization on fuel cost and gas emission under the condition that power balance and unit capacity constraint conditions are met. The method comprises the following steps: constructing main and auxiliary double-population parallel search; state indexes representing population convergence, diversity and feasibility are calculated in real time; based on the current state, a cooperation mode and evolution operator combined action is adaptively selected through a deep reinforcement learning agent; rewards are calculated according to improvement of the population on cost, emission and constraint satisfaction after action execution; and training the intelligent agent by using the state, the action and the reward, and iteratively optimizing the decision strategy until the Pareto optimal scheduling scheme meeting the constraint is output. According to the method, adaptive intelligent guidance of the evolutionary process is realized, and the optimization quality, the convergence speed and the algorithm robustness of the economic emission scheduling scheme of the power system are remarkably improved.

The invention discloses a broadband spread spectrum signal capturing device based on a rapid lock judgment feedback mechanism, and belongs to the technical field of spaceflight measurement and control communication. According to the method, a pseudo code parallel multi-channel capturing method and a carrier serial search capturing method are adopted, compromise processing is carried out between signal capturing time and consumed resources, and rapid carrier capturing can be completed through a rapid judgment locking feedback mechanism. According to the invention, rapid carrier capture of high-bandwidth spread spectrum signals can be completed under the condition of less resource consumption. According to the invention, the engineering realization occupies less chip resources, a low-performance processing chip can be adopted, the cost is saved, and the method is suitable for low-cost application scenarios. Through a rapid carrier wave lock judgment feedback mechanism, carrier wave serial search is combined with pseudo code parallel search, so that a corresponding carrier wave frequency is rapidly found during carrier wave frequency search, and a corresponding chip phase is found through pseudo code parallel search. The method is simple in design, easy in engineering realization and low in difficulty.

The invention discloses a commodity type selection recommendation method and system based on big data, and relates to the technical field of big data intelligent recommendation, and the method comprises the steps: carrying out the feature screening through CIIF and Shannon entropy, carrying out the optimization through an edge contribution degree and an adaptive pruning threshold value, carrying out the adjustment through an RL-QN strategy, carrying out the aggregation through HGCN, and generating a behavior enhancement rule set through NVR logic regularization. The dynamic weight of the rule is calculated through incremental reasoning, PoNSGA-III is used for optimization, a Combb-K objective function and a particle optimization algorithm are combined for global search, and a virtual center is adjusted to optimize a particle update path. According to the method, the personalized precision of commodity recommendation is improved through a collaborative linkage mechanism among feature screening, heterogeneous graph optimization and HGCN aggregation, and the diversity and stability of commodity recommendation are improved through fusion of a rule logical reasoning engine and a neural network regularization technology in combination with a parallel search mechanism of a multi-objective optimization algorithm.

The invention discloses a bee colony navigation method and system based on multi-source grouping-evidence fusion, and belongs to the technical field of multi-robot collaborative search and environmental perception. According to the method, parallel search can be carried out in a multi-source environment through a grouping-sharing mechanism, and the multi-source recall rate is remarkably improved. And through IP evidence fusion and multi-model Bayesian fitness, the robustness of the algorithm to odor intermittency, turbulence disturbance and model mismatch is enhanced, and the false drop rate is reduced. Finally, through virtual potential field navigation, reachability and safety of a search path are guaranteed, and search failure caused by obstacles or cooperation conflicts is avoided. The method is suitable for multiple fields of industrial leakage detection, environmental pollution monitoring, dangerous goods search and rescue, public safety monitoring and the like, and has a wide application prospect.

The invention discloses a high-dimensional space-oriented distributed graph index nearest neighbor search method, which comprises the following steps of: firstly, designing a boundary sensing balanced partition strategy, and measuring boundary characteristics of quantized nodes through a connection ratio; secondly, constructing a hierarchical mixed index architecture, vertically integrating a sparse global navigation layer and a fine-grained local precision layer, managing high-importance boundary nodes by adopting an M-Tree structure, and remarkably improving the cross-partition retrieval efficiency by utilizing the relative insensitivity of a tree structure to a curse in a boundary region; finally, a distributed multi-starting-point parallel search framework is provided, complementary entry points are dynamically generated based on query features and graph topology, vector space multi-region parallel exploration is achieved, irrelevant partitions are filtered through a navigation layer, and then accurate similarity calculation is executed in selected partitions. According to the method, on the premise that the high recall rate is guaranteed, the throughput of high-dimensional vector distributed search is remarkably improved, and an efficient and reliable solution is provided for large-scale high-dimensional vector retrieval.

The invention discloses a fine-grained flow scheduling priority marking system for a hardware unloading network, which comprises a packet analysis module for extracting quintuple flow identification information from a data packet and supporting multi-layer protocol field extraction, and a flow table module for maintaining the real-time state of the current active data flow by adopting a double-hash table structure and supporting parallel search and update, the priority judgment module is used for allocating scheduling priorities to current messages through configurable boundary matching rules on the basis of state information returned by a flow table, the message recombination and marking module is used for encoding priority labels into DSCP fields of an IPv4 message header, and the whole system adopts a full-assembly-line hardware architecture and can be ensured to continuously run at a linear speed under a 100Gbps network load. Dynamic packet-by-packet scheduling control is supported, a scheduling priority based on a real-time flow state can be given to each data packet, and fine scheduling of network resources is realized.

The invention discloses a hot word-fragment mapping management method, which comprises the following steps that: a hot word-fragment mapping list is constructed, a plurality of records are recorded in the hot word-fragment mapping list, and each record comprises a hot word and a fragment list corresponding to the hot word; the hot word is a certain word / statement of which the search frequency reaches a search threshold value in a set time window; according to a search request, preferentially matching the hot word-fragment mapping list, and if the hot word-fragment mapping list is hit, distributing the search request to a corresponding fragment for processing; and if not, broadcasting all the fragments to execute the search request. The invention also discloses an intelligent fragment scheduling method for parallel search, which comprises the following steps of: when a search request arrives, preferentially matching the hot word-fragment mapping list according to the search request, if the hot word-fragment mapping list is hit, acquiring the comprehensive score of each fragment in the corresponding fragment list, processing the search request by the fragment with the highest comprehensive score, and constructing a fragment request; and merging the fragmentation results to complete the search request.

This invention discloses a GNSS integer ambiguity fixing method integrating intelligent optimization algorithms and VIB strategies, belonging to the field of satellite navigation and positioning technology. This method combines the advantages of VIB strategies and swarm intelligence fusion optimization algorithms. By directionally grouping the ambiguity floating-point solution covariance structure, the ambiguities are guided into groups. The PSOAF and AWDE intelligent optimization algorithms are used to perform parallel searches on the grouped ambiguities, improving search efficiency and positioning accuracy. Simultaneously, the probability assessment and fixing judgment of the integer solutions of each sub-block ambiguity and the overall ambiguity ensure the reliability of the ambiguity solutions, making it suitable for complex and dynamic GNSS positioning scenarios.

The invention discloses a discrete space global optimization method based on neighborhood mapping and multi-path parallel search. The method comprises the following steps: acquiring a discrete state variable to be optimized as an initial search starting point; generating an initial candidate set by applying a neighborhood mapping rule, and executing structure pre-screening; constructing a multi-path parallel computing architecture, and distributing the effective candidate states to a plurality of independent computing threads by taking the effective candidate states as local initial starting points of the independent search paths; executing deep iterative search on each search path, wherein each round of iterative search is screened according to an evaluation index system comprising a first evaluation index for macroscopic evaluation and a second evaluation index for microscopic difference distinguishing; and periodically synchronizing local optimal solutions output by all parallel search paths, and screening to obtain a global optimal solution. According to the method, the problem of search stagnation during construction of the high-performance password component is solved, and the high-security password component with both high nonlinearity and low difference uniformity can be efficiently and stably generated.

A system and method for performing parallel local and remote searches, in particular to searching a local cache and remote database in parallel while accepting a search input, to obtain additional data for the local cache. The method includes, responsive to detecting a first portion of a search input, searching a local cache using the first portion to populate a search results list and initiating a remote search using the first portion; and responsive to receiving remote search results from the remote search, updating the search results list based on the remote search results and updating the local cache with data associated with items in the remote search results.

The invention provides an intelligent query system and method for cam configuration of a circular knitting machine, and is applied to the technical field of data processing. According to the method, multi-source data such as triangle models and process requirements and parameters such as configuration priorities are collected and are preprocessed and standardized into a unified query data set; core attributes are analyzed and extracted and are sorted according to parameter importance, and an optimized data set is generated; determining a retrieval strategy, a retrieval range and the number of iterations in combination with the demand complexity and the computing power to form a query batch; efficiency is improved through parallel retrieval, a query algorithm is optimized based on a multi-dimensional architecture, and a strategy is verified and dynamically adjusted through double indexes; and finally, according to a model rule optimization result, a redundant scheme is eliminated, and accurate, efficient and unique triangular configuration intelligent query is realized.

The invention discloses an e-commerce advertisement putting strategy generation method and system based on quantum computing optimization, and particularly relates to the technical field of quantum computing, comprising parameter coding, quantum optimization, line evolution, strategy resolving, dynamic optimization and strategy verification and output. According to the method, a high-dimensional parallel search mechanism based on a parameterized quantum circuit is constructed by introducing a quantum calculation normal form, the problem of calculation complexity index increase of a classical computer in e-commerce advertisement multi-parameter combination optimization is effectively solved, and the multi-parameter combination optimization of e-commerce advertisements is realized by designing a classical and quantum mixed architecture and a dynamic feedback optimization mechanism. The real-time adaptive adjustment of the advertisement putting strategy is realized, the practicability and robustness of the system are remarkably improved, the precise alignment of the technical means and the commercial target is realized by deeply coupling the quantum optimization process with the specific e-commerce advertisement business index, and a quantifiable business improvement effect is provided.

The invention discloses a concurrent implementation method of motion vector search, which relates to the technical field of H264 video compression implementation, and comprises the following steps of: judging whether a current macro block meets a condition of becoming a PSKIP macro block according to PSKIP macro block type information of a block adjacent to the current macro block, if the current macro block is judged to be in a PSKIP macro block type, ending search, and otherwise, entering the next step; under the assumption that the partitioning mode is a P16 * 16 partitioning mode, taking a pskipmv variable value obtained by predicting an adjacent block of the current macro block as a prediction vector value of the current macro block, and calculating and judging whether the type of the current macro block is a PSKIP macro block type or not; and when the judgment result does not meet the condition of the PSKIP macro block, carrying out parallel search processing, and carrying out concurrent motion vector search in different blocking modes. Under the condition that a small number of code rates are increased, the compression speed is greatly increased, and the method is more suitable for the use scene of real-time video compression.

The invention discloses a fluid pulsation accelerator architecture for reinforcement learning of a large language model, and belongs to the technical field of artificial intelligence hardware acceleration. The system aims at solving the problems that when a current universal processor executes an RLHF working load, an instruction driving normal form is not matched with data flow calculation, fixed parallel granularity cannot adapt to a dynamic load, and the efficiency is low due to the fact that a framework does not sense data statistical characteristics. The core of the system is a liquid systolic array calculation fabric capable of being dynamically reconstructed, execution is triggered through data flow, and kernel boundaries and instruction overhead are eliminated. According to the system, a global lookup table subsystem is fused, and nonlinear calculation is optimized through merging query and parallel lookup by using data distribution prior; and a dynamic scheduling unit is configured, and elastic parallelism is realized by adopting a resource allocation algorithm supporting work stealing. According to the method, the throughput and the energy efficiency of RLHF training and a large language model reasoning stage can be remarkably improved, and a new design normal form is provided for a next-generation AI special computing architecture.

The technical scheme of the invention discloses a compact type 5T0C eDRAM (Dynamic Random Access Memory) gain unit. According to the other technical scheme, the invention discloses a high-density content addressing circuit adopting the compact 5T0C eDRAM gain unit. Aiming at the problems of large storage unit area and limited density in the traditional CAM design based on SRAM (Static Random Access Memory), the invention provides a novel architecture based on eDRAM (Enhanced Dynamic Random Access Memory), and by optimizing the storage unit structure and a reading circuit, the storage density is obviously improved, and CAM search and in-memory logic operation are supported at the same time. Compared with an existing 6T SRAM, the area of the CAM array based on the eDRAM is reduced by 15%. According to the invention, the highest search and logic operation frequency of the binary CAM (BCAM) of 637 / 658 MHz is realized, and the search power consumption and the logic operation power consumption under the voltage of 1.1 V are respectively 0.91 fJ / search / bit and 27.47 fJ / bit. The method is suitable for data intensive applications such as artificial intelligence accelerators, network routers, genome analysis and the like which need efficient parallel search and near-memory calculation.

A photonic content addressable memory (CAM) (100) is provided. The photonic CAM (100) comprises: N search lines (101) and a plurality of match lines (102), each match line comprising N photonic CAM cells (110). Each search line (101) is configured to receive a search optical signal and each search optical signal comprises a different wavelength encoding one of N parallel search symbols. Each match line (102) is configured to receive: a wavelength division multiplexed content optical signal comprising a plurality of wavelengths, each wavelength encoding one of N parallel content symbols; and the N search optical signals from the N search lines (101). Each photonic CAM cell is configured to compare a respective search symbol with a respective content symbol and provide an output optical signal that is responsive to the comparison. Each match line (102) also comprises a content optical demultiplexer (120), configured to demultiplex the wavelength division multiplexed content optical signal and provide each CAM cell (110) with the respective content symbol.

The invention discloses a GNSS integer ambiguity fixing method fusing an intelligent optimization algorithm and a VIB strategy, and relates to the technical field of satellite navigation positioning, the method fuses the advantages of the VIB strategy and a swarm intelligent fusion optimization algorithm, directivity grouping is carried out on an ambiguity floating point solution covariance structure, ambiguity is subjected to guiding grouping, and the ambiguity is subjected to guiding grouping. The PSOAF and AWDE intelligent optimization algorithms are adopted to perform parallel search on the grouped ambiguity, so that the search efficiency and the positioning precision are improved. Meanwhile, probability evaluation and fixed judgment of real solutions are carried out on integer solutions of the ambiguity of each sub-block and the overall ambiguity, the reliability of the ambiguity solutions is ensured, and the method is suitable for GNSS positioning scenes in complex and dynamic environments.

The application discloses a kind of wideband spread spectrum signal acquisition device based on fast lock judging feedback mechanism, belong to spaceflight measurement and control communication technical field.The method uses the method of parallel acquisition of pseudo code multiple ways, carrier serial search acquisition, compromise between signal acquisition time and resource consumption, and can complete the fast acquisition of carrier by fast lock judging feedback mechanism.Can complete the fast acquisition of carrier of high bandwidth spread spectrum signal under the condition of less resource consumption.The engineering implementation of the application occupies less chip resources, can use low-performance processing chip, save cost, and can be applied to low-cost application scenarios.Can find corresponding carrier frequency quickly when searching carrier frequency by fast carrier lock judging feedback mechanism, and find corresponding chip phase by pseudo code parallel search.The application is simple in design, easy to implement in engineering, and has low difficulty.

The application discloses a kind of energy-saving optimization method and system of manufacturing factory water-cooled air conditioner, it is related to intelligent energy-saving optimization technical field, including collection multimodal data, construct initial gene bank, form cluster gene bank by spectrum decomposition and K-means, using genetic algorithm to update cluster gene bank, output configuration parameter table, set target function and heat balance constraint condition, convert into quadratic binary optimization problem, using quantum annealing algorithm parallel search, by discretization mapping and screening, output initial operation optimization parameter, carry out use and calculate deviation rate, update GP model parameter, output operation optimization parameter.The application forms configuration parameter table by exergy analysis combined with advanced algorithm, improves the accuracy and adaptability of system initial configuration, introduces quantum annealing and GP model dynamic optimization operating parameter, improves the dynamic adaptability and optimization efficiency of operating parameter.

This invention relates to the field of unmanned aerial vehicle (UAV) inspection technology, specifically to a dynamic flight path planning method and system for UAVs in substations. The method converts inspection requirements into a detailed set of point data. Using coordinate transformation, safety radius calculation, and shooting parameter settings, a unified set of safe inspection points is generated. Based on an improved ant colony algorithm, safety constraints are introduced through a path selection probability formula. Combined with multi-ant colony parallel search and an adaptive pheromone update mechanism, an optimized shortest flight path is generated. Subsequently, the optimized path point set is marked for review and verification. Path smoothing is then performed on the path point set. The starting point, ending point, and intermediate inspection points are fitted using a third-order Bézier curve to smoothly connect the flight path, satisfying safety turning angle, path smoothness, and safety radius constraints. This ensures a stable and smooth UAV flight path and ultimately outputs a temporary optimal flight path. This invention meets the safety, efficiency, and accuracy requirements of substation inspection tasks.

The application discloses a GPU-based multi-modal index integration method. The method comprises the following steps: acquiring multi-modal original data and extracting a feature vector; mapping the feature vector to a unified common semantic space; constructing a double-layer hybrid index structure on a GPU, wherein a first layer adopts IVF-PQ index for coarse screening, and a second layer adopts graph index for fine screening; after receiving a query request, performing two-stage parallel search on the GPU, first screening a candidate set through IVF-PQ index, and then fine screening a final result through graph index. The application solves the cross-modal semantic alignment problem through the unified semantic space, realizes high-precision and low-delay retrieval of massive multi-modal data by using GPU parallel computing and a double-layer index structure, and effectively overcomes the technical bottleneck that precision and efficiency are difficult to be considered in the prior art.

The invention discloses a new energy automobile charging power grid load prediction method based on a crowdsourcing feature selection genetic algorithm, and the method comprises the following steps: initializing crowdsourcing worker nodes, and distributing a crowdsourcing data collection budget for each worker node; performing data cleaning and feature construction on the collected new energy charging load historical data; executing a feature selection optimization task by each worker node, constructing an objective function of the feature selection optimization task, and performing parallel search and fitness evaluation on a feature subset based on a multi-objective genetic algorithm; outputting a local optimal feature solution set after iterative optimization of a multi-target genetic algorithm; aggregating the local optimal feature solution set to obtain a solution set of the crowdsourcing task for load prediction of the new energy charging station; and performing repeated detection on the solution set of the crowdsourcing task, removing repeated solutions, and performing non-dominated sorting on all reserved solution sets to obtain an optimal feature subset in the new energy charging station load prediction task. According to the invention, the precision and generalization ability of the new energy load prediction model can be improved.

This invention provides a method, apparatus, and vehicle for constructing a KD-tree from a vehicle-mounted LiDAR point cloud. The method includes: determining the first median of the point cloud data; performing KD-tree construction steps based on the left and right interval data until the unprocessed data variable is 0, and outputting the KD-tree. Thus, by obtaining the median using a quicksort algorithm, a full sort of the point cloud data is not required. Instead, the data is partitioned into two parts: less than a benchmark value and greater than a benchmark value. The median is then recursively searched, improving the efficiency of finding the first median. Since the first median divides the point cloud data into independent left and right interval data, multi-threaded parallel searching of left and right subtree nodes can be used during KD-tree construction, thereby improving the efficiency of KD-tree construction for the vehicle-mounted LiDAR point cloud and ensuring the response speed and operational sensitivity of the autonomous driving system.

The invention relates to the technical field of sensor networks, in particular to a forestry disease and pest dynamic monitoring system and method based on multi-source perception.According to the forestry disease and pest dynamic monitoring system and method, backoff time slot associated pheromones are extracted by calling an ant colony algorithm, the index product is calculated and divided by the total value, and the transition probability is output; a self-adaptive positive feedback optimization mechanism is used for accurately bypassing channel fading and a low-electric-quantity bottleneck area, the end-to-end message delivery ratio is greatly improved, multi-hop path node message fragments are extracted by calling a genetic algorithm, and redundancy dimensions are superposed and verified; forward erasure is executed based on global parallel search and coding feature recombination advantages to eliminate handshake retransmission communication overhead and tolerate a link random packet loss state, a probe timestamp is extracted and multiplied by a hop count to deduct a delay calculation error lower limit to solve an intercept slope, the intercept slope is substituted into a fitting calculation offset parameter, a count value is overwritten, and a hierarchical absolute timestamp axis is established. And the hardware temperature drift of the crystal oscillator is compensated, and microsecond-level alignment of multi-source sensing sampling is ensured.

The application provides a full-size subsonic wind tunnel loop aerodynamic profile multi-objective optimization method based on steady CFD, and belongs to the technical field of wind tunnels.The application reduces the initial geometric parameters to a sensitive parameter subspace through principal component analysis, generates initial sample points by using Latin hypercube sampling, extracts response values by executing steady Reynolds average Navier-Stokes equation solving, constructs a Kriging surrogate model, performs multi-start parallel search based on an expected improvement function, extracts a separation zone length and a backflow zone height to input an unsteady correction factor model to calculate a correction coefficient for correcting a diffusion section pressure recovery coefficient, searches a Pareto front solution set by using a non-dominated sorting genetic algorithm, selects an optimal parameter combination through a comprehensive performance index, and reconstructs the wind tunnel loop aerodynamic profile through a hierarchical parameterization method and performs verification calculation, so that the problem that the steady calculation framework cannot accurately predict the separation flow and the optimization result deviates from the true performance is solved.

The application discloses a fusion system of automatic activation value checkpoint search and automatic tensor parallel search, relates to the technical field of deep learning, and comprises network linearization, meta information extraction, an automatic activation value checkpoint solver, an automatic tensor parallel solver, and a two-stage solver. The meta information extraction is used to provide required memory overhead and operation overhead information for the automatic tensor parallel solver, the automatic tensor parallel solver solves according to the given memory budget and the meta information, the two-stage solver linearizes a series of tensor parallel strategies to a network, and the linearized network is transmitted to the automatic activation value checkpoint solver. The automatic activation value checkpoint solver controls the memory overhead below the real memory budget. The strategies of the automatic tensor parallel and the automatic activation value checkpoint are combined, the memory wall problem encountered in deep learning training is further slowed down, and meanwhile, the two-stage solver can guarantee that the memory problem is solved while the performance of model training is considered.

The embodiment of the invention provides a strategy searching method and device, and the method comprises the steps: receiving a data message from another device, and obtaining an attribute value set of a first attribute of an endpoint of the data message; determining a search key set based on the attribute value set; searching a strategy matched with the first search key from the first storage unit based on the first search key, searching a strategy matched with the second search key from the second storage unit based on the second search key to obtain a strategy set corresponding to the attribute value set, and executing the search step based on the first search key and the search step based on the second search key in parallel; and processing the data message based on the strategy set. And a parallel searching function is utilized, so that a plurality of strategies corresponding to a plurality of attribute values can be searched as short as possible.