812 results about "Multi agent technology" patented technology

The invention relates to an MAS-based (multi-agent system) multi-microgrid energy management system simulation method. The method is characterized in that a multi-microgrid energy management system is divided into a local management layer, a microgrid management layer and a microgrid coordinated management layer in different levels; various energy management functions of the different layers are used to establish a multi-microgrid energy management system model by different types of agents, and agent communication protocol is designed for coordination and convention of the agents. The agents in different control layers of the local management layer, the microgrid management layer and the coordinated management layer are integrated by a client-server architecture, and accordingly an MAT-based (multi-agent technology-based) multi-microgrid energy management system simulation platform is formed. The method is provided to verify different microgrid control strategies and scheduling plans.

A one number, multi-application, intelligent call processing system provides service benefits to a caller, a servicing location and/or a vanity number advertiser during a call, parallel to the call and/or post call in an integrated common architecture. The system utilizes VRU technology in conjunction with the national telecommunications network connected via Computer Telephone Integration (CTI) to a virtual telephone number database containing a nationwide master list of telephone numbers with attribute data items associated by Spatial Key linkage to each telephone number. The process of the invention is initiated by a caller dialing a selected telephone number to request information and/or services. Based on the number dialed, a caller or network provided ten-digit telephone number and VRU prompted for and received caller input, the system retrieves the application requested data from the virtual telephone number database and provides it to the network.

The invention relates to a multi-agent reinforcement learning scheduling method, a multi-agent reinforcement learning scheduling system and electronic equipment. The method comprises the following steps: step a, collecting server parameters of a network data center and virtual machine load information running on each server; b, establishing a virtual simulation environment by using the parametersof the server and the load information of the virtual machine, and establishing a deep reinforcement learning model of multiple agents; c, performing offline training and learning by utilizing the deep reinforcement learning model of the multiple agents and a simulation environment, and training an agent model for each server; and d, deploying the agent model to a real service node, and schedulingaccording to the load condition of each service node. The service running on the server is virtualized through the virtualization technology, load balancing is carried out in a virtual machine scheduling mode, resource distribution is more macroscopic, and a strategy that multiple agents generate cooperation in a complex dynamic environment can be achieved.

The present invention relates to multi-application, secure operating systems for small, secure devices, such as smart card microcontrollers. In particular, the present invention relates to mechanisms for secure runtime upload of applications onto small devices, authorisation mechanisms and the ability for authorised execution of multiple applications on the devices, where an application may be potentially larger than the microcontroller memory size. The mechanism simplifies life-cycle smart card management aspects related to post-issuance application (“applet”) upload and upgrade. Mechanisms to prepare applications (i.e. compiler techniques) using a common set of project files in one compiler toolset, for execution in a dual host & chip processor environment are described. These help automising the programming of the communication interfaces between the host and chip applications. An important motivation for the present invention is to provide a secure co-processor environment for general computer applications in order to counter software piracy, and to allow new models for secure electronic software distribution and software licensing.

Techniques for managing a system of heterogeneous applications are provided. A plurality of applications is classified into a plurality of application types. One or more of the plurality of applications in each of the plurality of application types are classified into one or more collections. A utility function of possible resource allocations is computed for each of the one or more collections. An application placement is computed that optimizes a global utility of the plurality of applications in accordance with the one or more utility functions. Placement and resource allocation of the plurality of applications are modified in the system in accordance with the application placement.

Techniques for improving the performance of a quantum processor are described. Some techniques employ reducing intrinsic/control errors by using quantum processor-wide problems specifically crafted to reveal errors so that corrections may be applied. Corrections may be applied to physical qubits, logical qubits, and couplers so that problems may be solved using quantum processors with greater accuracy.

The invention discloses a cloud architecture system of two-dimension code application, relates to the relative mobile internet technology, the two-dimension code technology, the geographic positioning technology and the telecommunication technology, and provides a comprehensive cloud architecture bottom platform support and a series of solutions to two-dimension universal application. The comprehensive cloud architecture bottom platform can effectively overcome difficulties existed in current two-dimension application, and particularly solve the problems that two-dimension information data integration which cannot be realized by the prior art but is needed to be realized is effectively built up between a buyer and a user to realize combination of simple and convenient businesses. Therefore, the two-dimension code application technology can be widely applied and popularized on-line and off-line among a buyer, a user and a third-party platform.

The invention discloses an ant colony optimization processing method of large-scale multi-target intelligent moving route selection; after data of NTSP target logistics delivery addresses, distances between every two addresses, and M price of cost for passing through each route are obtained, a route planning unit is solved by ant colony optimization technology so as to obtain a specific walking route for intelligent mobile-agent delivery, and the route is outputted to an executive mechanism for realization. When the method is used to solve the problem of large-scale multi-target intelligent moving route selection, the invention has good optimization performance, and has the advantages of parallelism, self-organization, strong robustness, and the like, and the obtained solutions are large in quantity, high in quality, and have strong approximation capability to the real Pareto solution set; the obtained solution set has uniform distribution; the calculation speed is high. The inventioncan be used in intelligent processing units of route planning systems in fields of logistics distribution, intelligent traffic, internet, robots, etc.

The invention discloses a multi-robot route planning method applied to a dangerous chemical environment, and belongs to the technical field of robot route planning. The method is applied to intelligent robots, can realize full map inspection of multi-robot under a complex and dangerous chemical environment with high dynamic performance, and can realize real-time formation reaction to special events. The core algorithm of the method is a deep reinforcement learning algorithm combined with a deep deterministic policy gradient and a curiosity driven algorithm. According to the core algorithm, thecomplex and tedious step of artificially designing environment rewarding in the same kind of algorithm are avoided, while the unknown complex environment is explored and adapted through the intrinsicrewarding and the sparse environment rewarding in the intelligent body itself; and meanwhile, a multi-Agent collaboration learning algorithm is adopted among the multi-robot to obtain the optimal cooperation strategy.

The invention discloses a hybrid electric vehicle energy management system and method based on a multi-agent technology. The energy management system comprises a signal collecting agent, a prediction agent, a storage battery agent, an energy management agent and an execution agent. The signal collecting agent collects vehicle condition information and environment information of a vehicle through a data collecting unit and transmits the information to the prediction agent, the prediction agent predicts the running conditions of the vehicle in the next small section of road and transmits the predication information to the energy management agent, the energy management agent calculates the vehicle energy optimal distribution according to the information transmitted by the prediction agent and the information transmitted by the storage battery agent and transmits distribution commands to the execution agent, and finally the execution agent executes a distribution scheme of energy between an engine and a motor. By means of the system, the running conditions of the vehicle in the next small section of road can be predicted in advance, energy management optimal control is conducted, the purpose of optimization is achieved, and the fact that a whole running condition is required in order to conduct overall optimization is avoided.

The invention relates to the technical field of networked control systems, in particular to a method for determining a consistent event trigger controller aiming at the condition that a multi-agent system with a leader has denial of service attack (DoS) by utilizing a modal dependent average residence time method, comprises the following steps of: step one, establishing a leader-follower multi-agent system model and a design event triggering condition; step two, establishing a DoS attack model and a switching system model; step 3, designing an event trigger controller and corresponding switching conditions. The average residence time method adopting modal dependence solves the design problem of the consistent event trigger controller in a leader-follower multi-agent system with DoS networkattack.

The invention discloses an unmanned cluster task collaboration method based on multi-agent reinforcement learning, and belongs to the technical field of unmanned cluster task planning. According to the method, the method comprises the steps of: establishing a reinforcement learning simulation environment for task planning of the multi-unmanned system based on Unity; using Gym to build the acquired information of the simulation environment into a reinforcement learning environment conforming to the specification; modeling an unmanned aerial vehicle cluster confrontation environment; using a Tensorflow deep learning library for building a multi-agent reinforcement learning environment; solving a multi-agent reinforcement learning problem by using a collaborative depth deterministic strategy gradient method; and outputting an unmanned cluster task planning result. According to the method, the prior art is greatly improved, and a better multi-unmanned-system cooperative task planning result can be obtained.

The invention provides and establishes a multi-agent frame structure based tour multimedia information personalized service system. The system adopts the Agent technology, finds and mines interests of users through constant independent study, and realizes automatic user interest modeling by adopting a vector included angle based K-nearest neighbor algorithm. As for each phase of a system working process, modules with relatively independent functions are divided, and each module is packaged into an Agent agency to form a multi-agent framework; and under the framework, the agencies are separated and cooperated so as to realize the characteristics of intelligence, personalization and the like of the multimedia information service. The system mainly comprises five functional modules, namely a user interface Agent, a filter sequencing Agent, an interest study Agent, a monitoring Agent and an information searching Agent. The method proposed by the invention has higher theoretical significance and application value for establishing high-quality intelligent tour information service.

The invention relates to a multi-agent anomaly detection method based on the Internet of things, which is an anomaly detection method for a wireless multimedia sensor network. In the method, the multi-agent technology is introduced mainly to solve the problem of the abnormal situation in the operating process of the wireless multimedia sensor network. The scheme of the multi-agent anomaly detection system is characterized in that: coordinating through a local agent, and monitoring a local sensor network autonomously, independently and promptly; when the terminal finds that the audio/video and other data are abnormal or the network needs regular maintenance, detecting whether the network is abnormal through issuing global agency; and if the abnormal situation occurs to the network, repairing the network dynamically and autonomously to ensure the correct operation of the network. The method of the invention can promptly detect whether the abnormal situation occurs to the network, and repair the network within a limited period to restore the normal operation of the network. Moreover, the invention can reduce the network congestion rate, balance the energy consumption of the network, efficiently utilize resources of the total network, improve the fault tolerance of the network and ultimately prolong the life cycle of the entire network.

The invention discloses a hybrid virtualization cloud computing resource optimal control method and a system catering to an X86/POWER hybrid virtualization cloud computing resource management platform in the technical field of cloud computing information processing. Through receiving applications of the cloud resource management platform, agent requests comprising virtual machine building, virtual machine destroying, virtual machine starting, virtual machine stopping, virtual machine backing-up and virtual machine state checking are sent; after the agent requests are analyzed, corresponding processing instructions are sent to virtual machine servers with different architectures; and finally, corresponding responses of the processing instructions are received and are returned to the applications of the cloud resource management platform to realize the optical resource control. The method and the system use the operation of the applications of the cloud resource management platform without considering the characteristic of different architectures of hybrid virtualization servers, so that the management control mode for the virtual machine resources is simplified.

The invention relates to a multi-modal fusion emotion recognition system and method based on multi-task learning and an attention mechanism and an experimental evaluation method, and aims to solve the problems that in the prior art, a multi-modal emotion recognition process without introducing a multi-modal fusion mechanism is low in efficiency and accuracy. The invention belongs to the field of human-computer interaction, and provides a multi-modal fusion emotion recognition method based on combination of multi-task learning and an attention mechanism, and compared with single-modal emotion recognition work, the multi-modal fusion emotion recognition method based on combination of multi-task learning and the attention mechanism is wider in application. Multi-task learning is utilized to introduce an auxiliary task, so that the emotion representation of each mode can be more efficiently learned, and an interactive attention mechanism can enable the emotion representations among the modes to mutually learn and complement each other, so that the recognition accuracy of the multi-mode emotion is improved; experiments are carried out on the multi-modal data sets CMU-MOSI and CMU-MOSEI, the accuracy and the F1 value are both improved, and meanwhile the accuracy and efficiency of emotion information recognition are improved.

A microgrid wide-area current protecting system and method based on the multi-Agent technology belongs to the technical field of microgrid protection. The system is composed of an intelligent electronic equipment layer, an area control and protection coordinating layer and a central processing layer, wherein the intelligent electronic equipment layer is composed of intelligent electronic devices (IED) configured on every breaker in a microgrid, and every IED comprises a state monitoring Agent, a measuring Agent, an network topology Agent and a protecting Agent; the area control and protection coordinating layer comprises an area control Agent and a protection coordinating Agent; the central processing layer comprises a central processing Agent. According to the microgrid wide-area current protecting system and method based on the multi-Agent technology, the microgrid current protecting range is divided through an improved taboo fusion breadth searching algorithm, gives full play to the advantage of a taboo algorithm capable of eliminating local optimization, makes full use of the advantages of wide search range and high efficiency of a breadth search algorithm and overcomes refusal operation and maloperation of the IEDs through a method of enlarging the current protecting range, thereby being capable of accurately and rapidly cutting off faults inside the microgrid.

This invention relates to a virtual interface structure user layer network communication system supported by hardware including network interface cards mounted on multiple computer hosts interacted with the host via PCI bus, cross switches connected with the network interface cards which are connected with the cross switches to form a communication network topological structure via physical links. The system also includes communication interface units composed of a user agent module and core agent module in each unit for realizing functions stipulated by VIA protocol and a drive module driving the cards to finish communication functions.