20 results about "Deployment algorithm" patented technology

The invention discloses an intelligent video monitoring system and method supporting AI strategy cloud side end cooperative execution. According to the system, unified management of computing power resources and dynamic deployment of an AI algorithm are realized based on an extended standardized protocol through a three-level architecture of a cloud side center platform, an edge computing node and an end side camera. The method comprises the following steps: a cloud platform acquires computing power information of edge equipment; an AI algorithm execution strategy is generated and issued according to service requirements; if the required algorithm is not deployed in the equipment, the equipment is controlled to download and automatically deploy an algorithm module; and finally, the driving equipment executes AI analysis and recovers the structured data. According to the invention, AI computing tasks are reasonably distributed through a cloud side-end cooperative computing architecture, so that the computing power construction cost of a central platform is effectively reduced; through algorithm and hardware decoupling and a standardized protocol interface, ecological binding is broken, on-demand loading and flexible scheduling of the algorithm are realized, and the economical efficiency, the flexibility and the intelligent level of the system in large-scale scenes such as smart cities and the like are remarkably improved.

The invention discloses an unmanned aerial vehicle service simulation method and system based on an Internet of Things platform. The method comprises the following steps: constructing an unmanned aerial vehicle information model; according to input from a real scene and / or a virtual scene, performing service representation on the unmanned aerial vehicle information model; independently deploying an algorithm process and a decision process for the service, and adjusting the sensing data supply of the Internet of Things platform to the unmanned aerial vehicle information model according to the processing conditions of the algorithm process and the decision process; changing the operation environment of the unmanned aerial vehicle information model in the virtual reality scene according to the sensing data supplied by the Internet of Things platform; and obtaining unmanned aerial vehicle behavior data in the operation environment, and according to the unmanned aerial vehicle behavior data, predicting an abnormal business execution condition of the unmanned aerial vehicle and feeding back the abnormal business execution condition to the Internet of Things platform. The business is represented for the simulation model in the real scene and the virtual scene, the appropriate process is deployed to respond to the simulation condition in time, the sensing data of the Internet of Things platform is utilized to generate a diversified operation environment, the business execution is effectively predicted, and the simulation credibility is improved.

The application discloses an information processing method and device, a storage medium and an electronic equipment. The method comprises the following steps: receiving search information; processing the search information by a target recommendation algorithm to obtain a search result, the target recommendation algorithm being obtained by iteratively training an initial recommendation algorithm by a training sample set constructed by a plurality of target data information, the plurality of target data information being obtained by screening a plurality of historical data information, the historical data information at least comprising a plurality of search result sets and behavior feedback data for search results in the plurality of search result sets, wherein the behavior feedback data is used to represent whether the search result is selected; and displaying the search result. The application solves the technical problem that the accuracy of the search result output by the iteratively updated algorithm model is relatively low after the algorithm model is deployed and the algorithm model is directly iteratively updated according to the feedback data of the user.

The application discloses a multi-access edge computing server deployment method and device and a storage medium, and relates to the technical field of edge computing. The method comprises the following steps: acquiring regional tasks and environmental information; establishing an associated optimization model based on a double-layer packing paradigm; using non-probabilistic interval parameters to represent interval constraints; determining the interval constraints and interval objective functions based on the center and radius weighted sum and the satisfaction threshold; obtaining edge server deployment strategies and resource allocation strategies based on a double-layer deployment algorithm of an improved grouping genetic optimizer; enabling and configuring edge servers based on the edge server deployment strategies, allocating computing, storage and multi-concurrent network bandwidth resources based on the resource allocation strategies, recording the achievement during the operation to generate operation indexes and returning the operation indexes. Through the associated optimization model and the improved solving algorithm, the application takes into account the profit and robustness, ensures the feasibility of the strategy in an uncertain scene, and improves the service response speed.

The application provides a computing power network task scheduling method and system, which comprises the following steps: obtaining a to-be-deployed micro service, a deployment algorithm, to-be-deployed micro service requirements and cluster information, screening a first cluster set matched with the to-be-deployed micro service based on the to-be-deployed micro service requirements and the cluster information; determining a deployment strategy based on the first cluster set and the deployment algorithm, wherein the deployment strategy comprises the number of replicas and the name of a deployment cluster, the deployment cluster belongs to the first cluster set, and the to-be-deployed micro service is deployed based on the deployment strategy; obtaining a to-be-scheduled task, a scheduling algorithm and to-be-scheduled task requirements, screening a second cluster set from the deployment cluster set based on the to-be-scheduled task requirements; determining a scheduling strategy based on the second cluster set and the scheduling algorithm, wherein the scheduling strategy comprises the name of a scheduling cluster and the number of task requests corresponding to each scheduling cluster, the scheduling cluster belongs to the second cluster set, and the to-be-scheduled task request is scheduled based on the scheduling strategy.

The application provides a kind of aerospace engineering data algorithm model full life cycle management method and system, it is related to aerospace engineering algorithm model management and service technical field.The method provided by the application stores multiple algorithm model files of different categories in the model library, generates the inference flowchart corresponding to the target data analysis process in response to the editing operation and the configuration operation input by the user, generates the algorithm model service file corresponding to the target data analysis process according to the inference flowchart and sends it to the server in response to the release operation input by the user, so that the server deploys the algorithm model service according to the algorithm model service file.The method provided by the application can quickly complete the deployment of the corresponding algorithm model service based on the analysis configuration operation and the algorithm model file set by the user, and then complete the analysis and processing of data through the algorithm model service, reduce the model service construction cost, improve the efficiency of model application, and enhance the data analysis and processing capability.

This invention discloses a multi-task-aware microservice deployment method. Using microservices as vertices and the basic rules governing their call relationships, timing relationships, and constraints as directed edges, a graph model is established. The resource indicators of each microservice represented by a vertex are mapped to the weights of its outgoing edges, forming an Area of ​​Effect (AOE) network. Microservices undertaking critical tasks are identified according to their importance, and these, combined with the resource indicators of the remaining microservices, constitute a multi-task-aware deployment model based on the AOE network. A multi-task-aware deployment algorithm calculates priority paths composed of microservices from this model. Using a greedy approach, paths are preferentially allocated to cluster physical nodes with more remaining resources according to their priority. Finally, an optimized scheme for splitting and deploying the original microservice groups is provided based on the allocation results. This invention helps users complete deployment work flexibly and efficiently, thus solving the problem of difficult deployment.

The invention provides a collaborative optimization method and system for micro-service deployment and request routing, and belongs to the technical field of deployment optimization and user request routing under a cloud computing micro-service architecture. According to a to-be-deployed micro-service instance, a deployment algorithm takes a micro-service calling relation, micro-service resource consumption characteristics and a real-time resource state of a data center server as a core basis, and an intelligent deployment strategy giving consideration to service delay control and server load balance is generated through cooperation of a communication perception grouping algorithm and reinforcement learning; after a user initiates an application request, accurate routing is realized by means of an A-star queue perception scheduler in combination with the multi-chain structure characteristics of the request and the real-time queue state of each micro-service instance, end-to-end delay is effectively reduced, and finally, low-delay and high-stability service use experience is provided for the user.

The application discloses a kind of based on target k-coverage multi-unmanned plane auxiliary charging and data acquisition method, comprising: obtaining and defining wireless chargeable sensor network area and related parameter set;Formalization based on target k-coverage unmanned plane minimization deployment problem;Using the clustering division algorithm based on coalition formation game, obtain disjoint sensor set family;Using restricted Prim algorithm, obtain candidate charging sensor set;Using the unmanned plane minimization deployment algorithm without neighborhood based on edge weight threshold, obtain the flight trajectory of all unmanned planes, supply power for the sensor of insufficient power, and collect the perception data of data concentrator.Greatly reduce the deployment cost of unmanned plane, ensure the long-term operation of wireless chargeable sensor network, while collecting important perception data for professional data analysis.

The application belongs to the technical field of edge computing, and discloses a kind of edge service autonomous deployment method for service-intensive Internet of Things scene, comprising: abstracting Internet of Things service as process component, planning data flow and service scheduling in network using various process modeling methods, and mathematically analyzing process model;Using service network model and distributed microservice data communication model to automatically monitor network data;According to the mobility of entity service node, set up appropriate service deployment algorithm to get the optimal service deployment scheme;Automatically deploy various microservices in code repository to Internet of Things service network.The application starts from process graphical modeling, autonomously completes process programming, process feasibility analysis, service network data and topology monitoring, deployment scheme calculation, and automatically deploys the whole life cycle content of running, combines the theoretical innovation of deployment algorithm with the realization of practical engineering scene application.

The invention provides a heterogeneous hardware platform-oriented automatic AI algorithm model deployment method. The method comprises the following steps of: obtaining a to-be-deployed AI algorithm model and a heterogeneous hardware platform; creating an AI algorithm model deployment task, and setting model adaptation task parameters; constructing a model lightweight algorithm library and a rule constraint engine, and establishing an exception handling mechanism; performing model lightweight and unified representation based on a rule constraint engine and an exception handling mechanism to form a recommended AI algorithm model; matching the recommended AI algorithm model with a heterogeneous hardware platform; and performing performance evaluation based on a heterogeneous hardware platform test environment. According to the automatic AI algorithm model deployment method oriented to the heterogeneous hardware platform, an AI algorithm automatic deployment process is designed and a rule constraint engine and an exception handling mechanism are constructed based on quantization, pruning, distillation and other model optimization algorithms, automatic model efficient deployment of the model is realized, and meanwhile, the performance of the deployed model is prevented from being greatly reduced.

The application discloses an enhanced practical Byzantine fault tolerance method for service function chain deployment. Firstly, a three-layer trusted network system model integrating blockchain and deep reinforcement learning is constructed, and network parameters and SFC deployment constraints are defined. Then, a VRPBFT enhanced consensus mechanism integrating a verifiable random function (VRF) and a node reputation grading model is designed to quantify node credibility and realize dynamic hierarchical division. Next, a master node fair selection method based on VRF is proposed to reduce consensus delay and improve Byzantine node detection capability. Finally, an SDRL deep reinforcement learning deployment algorithm is designed to dynamically adjust VNF and link deployment strategies in combination with node credibility, thereby optimizing resource utilization and service quality. Compared with the traditional PBFT, the consensus delay is reduced by about 30%, and the proportion of Byzantine nodes is reduced by 40% after 100 rounds of consensus. Compared with existing algorithms, the long-term average income of the SDRL algorithm is increased by 17%, the SFC request acceptance rate is increased by 14.49%, the income-cost ratio is increased by 20.35%, the CPU resource utilization rate is 42% and is increased by 27.96%, the safety of SFC deployment in a heterogeneous network is ensured, and the collaborative optimization of resource utilization and service quality is realized, so that the application is suitable for SFC trusted deployment in a heterogeneous network environment such as the Internet of Things and 5G.

The invention discloses a cost and energy consumption optimization-oriented application dynamic deployment method and device in a cloud edge environment and a medium. The method comprises the steps of S01, obtaining an initial deployment scheme through an application program deployment algorithm based on multilevel graph division; s02, for a dynamic change event of the application program, performing dynamic preprocessing on the initial deployment scheme to update the deployment scheme of the application program; s03, performing iterative optimization on the updated deployment scheme in combination with cost and energy consumption gain calculation on the basis of a label propagation algorithm on the premise of meeting the capacity limitation of the data center and the time delay requirement of an application program, so as to enable the model to converge or reach a set constraint condition; according to the scheme, in the face of dynamic scenes such as application scale expansion and task cooperation relation change, the overall deployment cost and energy consumption can be remarkably reduced, and good expandability and scheduling efficiency are achieved; according to the method, the cost and energy consumption optimization of application deployment in the cloud edge environment can be realized on the premise of meeting the capacity and time delay limitation.

The application discloses a dynamic scene binocular vision SLAM method based on target tracking and belongs to the field of simultaneous localization and mapping. In the remote end, an embedded device is used to control a follow-up holder camera installed on a moving body and collect images. In the service end, an embedded device is used to deploy an algorithm. The algorithm adopts a SLAM algorithm combined with target tracking to exclude the interference of the shielding mechanism of the moving body itself and dynamic objects in the field of view, so as to reduce the error of pose calculation and map construction. The application can not only detect and segment dynamic objects, but also segment the objects shielding the field of view. Based on a light target detection network and a distributed structure, resource consumption is reduced, so that the algorithm can be deployed on an embedded device.

The application provides a network resource pre-allocation method, a terminal, an electronic device and a medium, wherein the method is applied to the terminal and includes the following steps: S1, determining first parameters, second parameters and third parameters; S2, running a service deployment algorithm based on the second parameters to obtain a target function value; S3, running a derivative-free optimization algorithm based on the target function value, the first parameters and the third parameters to determine a network resource pre-allocation scheme of the next round of iteration; and S4, iteratively executing S2 and S3 until the derivative-free optimization algorithm converges to obtain a target resource pre-allocation scheme, wherein the service deployment algorithm is run based on the network resource pre-allocation scheme of the current round of iteration in each iteration. In the application, the relationship between various heterogeneous network resources can be established, and then the derivative-free optimization is performed through a network resource pre-allocation strategy, so that the pre-allocation of the network resources is realized.

The invention discloses a multi-access edge computing server deployment method and device and a storage medium, and relates to the technical field of edge computing. The method comprises the following steps: acquiring a regional task and environment information; establishing a correlation optimization model based on a double-layer boxing normal form; representing interval constraints by adopting non-probabilistic interval parameters; on the basis of the center and radius weighted sum satisfaction degree threshold value, interval constraint and an interval objective function are determined respectively; solving based on a double-layer deployment algorithm of an improved grouping genetic optimizer to obtain an edge server deployment strategy and a resource allocation strategy; and starting and configuring an edge server based on an edge server deployment strategy, allocating calculation, storage and multi-concurrent network bandwidth resources based on a resource allocation strategy, recording an achievement condition during operation, generating an operation index, and returning the operation index. According to the invention, through the association optimization model and the improved solution algorithm, the profit and robustness are considered, the feasibility of the strategy in an uncertain scene is ensured, and the service response speed is improved.

The application provides an engineering structure and a cluster wisdom monitoring operation system thereof, and relates to the technical field of structural wisdom health monitoring.The system comprises a node unit of a monitoring system function algorithm, a data format matching plug-in system in a monitoring information flow processing and control process, an algorithm workflow arrangement system for assembling a complex calculation scheme, an application market system for regulating the behaviors such as market entry, transaction and delisting of various functional algorithm nodes and other kinds of monitoring data processing services, and a technical guarantee mechanism for guaranteeing the normal operation of the monitoring algorithm workflow system.The application adopts the above engineering structure and the cluster wisdom monitoring operation system thereof, so that algorithm engineers in the field of engineering structure health monitoring can quickly deploy algorithm modules, and application engineers in the field of engineering structure health monitoring can conveniently build, arrange and run the structural wisdom monitoring system based on actual engineering requirements.

The present disclosure relates to an automatic driving model deployment method, device, equipment and storage medium. A configuration file of a hardware platform to be deployed is obtained, a search space is constructed, and the similarity of a calculation node of an algorithm model to be deployed to a corresponding associated operator in the search space is determined. A candidate operator is determined according to the similarity. The candidate operator corresponding to the calculation node is selected, combined according to a calculation node relationship graph, and a candidate deployment algorithm model is obtained. For a hardware platform that needs to deploy an automatic driving algorithm model, an automatic driving algorithm model adapted to the hardware platform can be automatically output according to the hardware information of the hardware platform. The problem that the current model deployment work requires high professional ability and experience of algorithm deployment engineers, and the model deployment efficiency is low and the cycle is long is solved. The workload of automatic driving algorithm model deployment in the vehicle development process of the host factory is greatly reduced, and the model deployment efficiency is improved.

The invention relates to a cross-language vehicle cloud collaborative algorithm cloud integrated deployment method, which comprises the following steps of: packaging an SDK (Software Development Kit) by algorithms, respectively packaging the algorithms written by Python, C or C + + and Java programming languages into the SDK, and packaging a third-party library or a configuration file on which the algorithms depend into the SDK; algorithm SDK integration, algorithm SDK deployment, algorithm SDK calling and algorithm result issuing: the cloud control platform further processes the algorithm result, converts the result into a corresponding instruction or information according to a communication protocol between vehicles and cloud, and issues the instruction or information to the vehicle-mounted terminal through a communication network; algorithm reusability and integration efficiency are improved: by packaging algorithms written by different programming languages into a unified SDK form, differences between languages and environments are shielded, so that the algorithms can be conveniently reused in different projects, and the workload of repeated development is greatly reduced.

The invention discloses an unmanned aerial vehicle temperature and humidity early warning method based on embedded optimization, and relates to the technical field of unmanned aerial vehicle safety, a temperature and humidity prediction model is constructed through a Transform three-level lightweight optimization algorithm, and the model is deployed on an unmanned aerial vehicle through a cloud training and embedded fine tuning dual-strategy deployment algorithm. Generalization and prediction precision of the model under different working conditions are improved, and the requirements of the unmanned aerial vehicle for model volume, reasoning speed and resource occupation are met; when the change degree of the sensing data exceeds the change threshold value of the sensor, temperature and humidity data are collected, and the occupancy rate of a central processing unit is reduced; then the temperature and humidity data are preprocessed, and the data quality is improved; and finally, a temperature and humidity prediction model is scheduled to perform multi-time-step prediction on the preprocessed temperature and humidity data, and a three-level early warning mechanism is adopted to perform unmanned aerial vehicle temperature and humidity early warning according to the predicted temperature and humidity and a temperature and humidity early warning threshold value, so that quick response is realized, and time is won for abnormal intervention.