52 results about "Decision making capacity" patented technology

The invention relates to the technical field of traffic control and artificial intelligence and provides a traffic signal self-adaptive control method based on deep reinforcement learning. The method includes the following steps that 1, a traffic signal control agent, a state space S, a motion space A and a return function r are defined; 2, a deep neutral network is pre-trained; 3, the neutral network is trained through a deep reinforcement learning method; 4, traffic signal control is carried out according to the trained deep neutral network. By preprocessing traffic data acquired by magnetic induction, video, an RFID, vehicle internet and the like, low-layer expression of the traffic state containing vehicle position information is obtained; then the traffic state is perceived through a multilayer perceptron of deep learning, and high-layer abstract features of the current traffic state are obtained; on the basis, a proper timing plan is selected according to the high-layer abstract features of the current traffic state through the decision making capacity of reinforcement learning, self-adaptive control of traffic signals is achieved, the vehicle travel time is shortened accordingly, and safe, smooth, orderly and efficient operation of traffic is guaranteed.

The invention provides a question-answering system based on man-machine integration. The question-answering system based on man-machine integration comprises an input module, a denoising module, a semantic parsing module, an information processing module, an intelligent decision-making module, an output module, a manual customer service processing module and a training module. The invention further provides a question-answering method based on man-machine integration. The method comprises the steps of inputting questioning information; converting the questioning information into standard text formatting information; carrying out denoising and module structuralization on the text formatting information; carrying out semantic parsing on the questioning information; judging an information processing result and making a decision; generating an answer and inputting the answer if the answer is correct; presenting the question to manual customer service to be processed if no answer is available or the answer is wrong, judging again according to a processing result and making a decision, training judging capacity and decision making capacity, and outputting a correct answer. According to the question-answering system and the question-answering method based on man-machine integration, self-learning capacity and self-improving capacity are improved, humanization and intelligentization are improved, and the accuracy and the timeliness of the system are guaranteed.

The invention relates to an indoor emergency evacuation simulation method for a complex building. According to the method disclosed by the invention, a three-dimensional extended hierarchical node relation model (EHI-NRM) of an internal space unit of the complex building is automatically generated, and the update of a dynamic path of a large-scale complex building is supported; an improved cellular automaton model (ICA) can better simulate jam, avoidance, mass-following and other phenomena of a large-scale crowd during the evacuation process; and an optimal evacuation path is searched in real time in the macro-level through an EHI-NRM dynamic path-finding method, and the ICA is applied in the local level for simulating the behavior characteristics of the crowd. According to the method disclosed by the invention, the evacuation simulation of the large-scale crowd in the large-scale complex building is supported, and the method has important application values for ensuring the safety of staff in the building and improving the decision-making capacity of an emergency rescue department.

Techniques are described herein to converge and conditionally filter data and to present such data in a manner that is beneficial to situational data analysis, in association with emergency management, law enforcement, or other such situations. Data may be combined to create enhanced situational awareness for responders and improve their decision making ability.

The invention discloses a meteorological service performance evaluation method based on a geographical information system (GIS), which is used for carrying out meteorological service performance analysis and evaluation by comprehensively utilizing a neural network method, an analytic hierarchy process, a structural equation method, a gray system method and a statistical analysis technique through using corresponding social statistics and survey data. The method is used for carrying out uniform control and management on hydrologic information, humanity information and meteorological information in a researched region by using a cartographic database and remote sensing data as basis geospatial data through using population data, casualty loss data and geographical hydrologic thematic data based on a GIS technology and a database technology by adopting a design mode of centralized management and maintenance. The invention realizes graphic and feature integrated maintenance management, comprehensive query, space analysis, thematic map, simulation analysis and meteorological service evaluation functions, improves the informatization level and the aid decision making capability of the meteorological service evaluation, and provides aid decision basis for the geographical hydrologic management and the meteorological service performance evaluation.

The principles of the present invention provide for optimizing healthcare/dental-care revenue cycle management through dynamic information analysis. The principles of the present invention incorporate interactive, exploratory data analyses of the full revenue cycle management processes and interrelated systems. Such analyses may include functions of appointments, scheduling, billing, posting, denial management, collections, whilst integrating hospital, physician, practice management, sales, marketing, finance and other interrelated systems and/or functions. As a result, complete, consistent, derivative, quality data analysis allowing users to identify initial and sub-process performances and their effects on the main business process enriching decision-making capabilities for medical service providers.

The invention discloses a satellite cluster configuration adjustment planning method for space target collaborative observation. According to the method, each satellite is regarded as an intelligent Agent with autonomous planning and decision-making capability, a coordination graph based on a cluster communication structure is established, and tasks optimized by a global optimal configuration adjustment strategy are decomposed to local learning of each satellite based on the communication coordination graph. According to the method, the top-layer planning of configuration adjustment and bottom-layer control are comprehensively considered; solving is carried out from the perspective of cluster cooperation planning; the global planning learning task is divided into the paired Agents according to the communication relationship, so that the autonomy and intelligence of the satellite cluster are enhanced, the on-orbit viability and the task execution level of the satellite cluster are improved, and a new idea and method are provided for on-orbit autonomous operation and management of the satellite cluster.

The invention discloses an excavating method based on air activity target data, relating to the association and sequence algorithm excavating technology in the data excavating application field. The invention takes the conventional type data excavating method as basis, the spatial data excavating technology is infused, the whole excavating flow and policy from air activity target data preprocessing, statistic analysis, excavating data extracting to result knowledge excavating is completed, and air activity target data excavating is completed. The invention also has the advantages that the method can be applicable to various activity target data excavation, comparison with real time data trend can be carried out by excavation and predicting is carried out on the air target activity. The invention is especially applicable to searching unknown air target activity information in air activity target data accumulated for a long time by computer assist and improves engineering application decision making ability.

The invention discloses a teleconference integration method and a unified communication system in a unified communication terminal. The method comprises the following steps: adding a conference user interface by calling the WEB service provided by a conference server; adding teleconference accounts with different authorization classes, wherein the account numbers and passwords of the teleconference accounts are the same as the account numbers and passwords of the unified communication users; and in the unified communication terminal, configuring a WEB service client terminal, calling different WEB service control interfaces on the conference server through the WEB service client terminal, and realizing the functions of authenticating user logon, holding a conference, stopping the conference, adding members, eliminating members, enabling a conference member to exit from the conference, and saving and inquiring logs. By using the invention, the teleconference system and the unified communication system are integrated together, the teleconference is held by the unified communication terminal at any time in any place, saves cost, and great convenience is provided, thereby greatly improving the service operation capacity and communication and decision-making capacity of an enterprise.

The invention provides a method for achieving data fusion storage in a distributed heterogeneous system. The method for achieving the data fusion storage in the distributed heterogeneous system comprises the steps of building a distributed heterogeneous storage model, a distributed heterogeneous data model and a heterogeneous data safe model; utilizing distribution and pieced rules and mechanisms to perform data processing; confirming a data access mode according to data relevance relation and information application requirement. By means of a fusion technology, the reaction speed of enterprises to latest data dynamic can be improved, decision-making capacity can be enhanced, the requirements of users in places can be obtained, and service quality can be improved.

The invention discloses a data blood relationship analyzing method for power grid regulation and control running. The data blood relationship analyzing method comprises the following steps of (S1) establishing a data blood relationship structure for power grid regulation and control running and forming a hash table; (S2) positioning nodes of target data in the data blood relationship structure when data are abnormal; and (S3) traversing a search data queue on the basis of the data blood relationship structure and the target data, inquiring data blood relationship information of queue head elements of the search data queue, adding gene information and heredity operational characters in the data blood relationship information to the tail of a source data queue, and using the final source data queue as a tracking result of a source of the target data. By the data blood relationship analyzing method for power grid regulation and control running, the correlation between relevant abnormal data and data can be quickly found out, the abnormal data can be disposed effectively, the scheduling, producing, assisting and decision-making ability is improved, and safety and stability of running of a power grid are improved.

The invention discloses an improved cellular automaton traffic flow simulation analysis method based on a geographic information system. According to the method, multi-criteria judgment, logistic regression, an improved principal component analysis method and an improved decision-making tree and kernel-based learning machine method are comprehensively used for mining and extracting the transformational rule of a cellular automaton, then the improved cellular automaton is used for conducting simulation analysis of traffic flow, and corresponding traffic flow data, geographic information data and the like are used for conducting traffic flow modeling and analysis. The method is based on a geographic information system technology and a database technology to conduct unified control and management on traffic flow information, geographic information and social information in a studied area. The method achieves integrated simulation analysis and maintenance management with diagrams and texts, comprehensive inquiry, thematic map and spatial analysis and traffic flow simulation analysis, the intelligence level and the aid decision making capacity of traffic management are improved, and the method is of certain reference significance to work study in the related fields.

The invention discloses an unmanned plane independent robust decision-making method based on scenario building. The unmanned plane independent robust decision-making method based on scenario building is characterized by comprising the following steps that first, the decision variable of a Red unmanned plane and the decision variable of a Blue position in a battle field are defined; second, a scenario is built; third, an influence diagram is established by using an influence diagram modeling method; fourth, the expected utility value of each scenario centralized in scenario of each decision candidate action plan in decision candidate action plans of the Red unmanned plane is calculated; fifth, the Red unmanned plane makes robust decisions according to the expected utility value. The unmanned plane independent robust decision-making method based on scenario building can effectively solve the problem of weak independent decision making capacity in a complex battle field environment, the unmanned plane is made to be capable of making independent decision in various scenarios, accordingly, the independent decision making efficiency and the decision making effect of the unmanned plane are improved, optimal action scheme choosing is achieved, and the feasibility, the effectiveness and the safety of the decision-making scheme are guaranteed.

The invention discloses a design method for an unmanned surface vehicle tracking controller, and relates to the technical field of an unmanned surface vehicle tracking control technology. The method specifically comprises the following steps that: S1: according to unmanned surface vehicle tracking task requirements, firstly, designing a decision network, training to enable the decision network toobtain a decision capability, and designing an exploration function to explore an unmanned surface vehicle path tracking state; S2: designing a reward function, and obtaining the motion state of the unmanned surface vehicle through a state observation method. By use of the design method for the unmanned surface vehicle tracking controller, a deep reinforcement learning algorithm is adopted to train a deep convolutional neural network to serve as the unmanned surface vehicle tracking controller, the design process of the unmanned surface vehicle tracking controller is simplified, a neural network approaching form is favorably used for replacing mathematical derivation, in addition, the controller is automatically trained through a program, manual intervention is not required, calculation issimple, and transportability is high.

The invention provides a data processing method for train set failure prediction. The method includes the steps of sending a request for failure prediction required data to a data preprocessing module, and receiving the requested data from the data preprocessing module for train set failure prediction, wherein the data obtained from service systems of a train set is classified to be stored in thedata preprocessing module according to a preset data model. The invention further provides a data processing device for train set failure prediction. Multi-source heterogeneous data related to the train set failure prediction are unified and classified to be stored, network resources are saved, data response speed is increased, the failure prediction speed and accuracy are improved, and the decision making capacity of a rail system for operation and maintenance activities is improved.

The present invention discloses a warehouse location method by taking consideration of online dealers in a crossing warehouse network. The method comprises the following steps of: the step 1, giving acustomer distribution set, related features of client logistics satisfaction and related features of customer purchase ability, and based on the given features and the given set, employing an artificial neural network method to predict a product sales demand in a year in future; and the step 2, solving an optimization warehouse location problem through adoption of an E&M clustering algorithm witha distance weight, and performing reduction of the problem through a hierarchical clustering mode. The warehouse location method can rapidly, accurately and effectively reduce the storage transportation cost, enhance the decision-making ability of enterprises leaders, optimize the enterprise supply chain management, reduce the management cost and improve the industrial competitiveness of enterprises.

The invention discloses a routing inspection robot control method based on a human-computer coordination system. The control method comprises task-level control and instruction-level control, and themethod comprises the following specific steps that a user performs task description to the robot system at the task level, and the target state of the robot system is obtained based on the task description amount through an inference mechanism; the current characteristic parameters of the robot are obtained from the target state, and the current characteristic parameters serve as target instructions to be stored in the background; and at the instruction level, the user performs automatic planning on a task executing process according to a feedback result in the task level planning, and the system judges user planning rationality on the background and gives out a prompt and recommended task path. According to the method, human planning and decision-making capability is combined with autonomous planning of a humanoid robot, so that the robot has the capability of adapting to environment change; and by virtue of combined use of the task-level parameters and instruction-level parameters, the movement track of the robot is more accurate, and the completion of the intelligent task is more accurate and reliable.

The invention discloses a dynamic layered early warning modeling method for food safety risks. Based on the tree structure and hierarchical analysis, the association relationship, prediction conditions, and prediction indicators of key food safety factors are used as the attributes of the factor nodes, and the tree-shaped association is built. Create a hierarchical early warning model for food safety risks; including: determine the index system for food safety risk early warning; establish a tree structure for the early warning model; Model, establish a layered early warning model of food safety risks; get the early warning and prediction status value of the node. The technical scheme of the invention is used for dynamic layered early warning of food safety risks, can effectively utilize resources, reduce work costs, save expenses, and improve food safety risk prediction capabilities, decision-making capabilities and defense capabilities for major public safety events.

A method (200) is described for creating decision trees for processing a sampled signal indicative of speech. The method includes providing model sub vectors from partitioned statistical speech models of phones, the models comprising vectors of mean values and associated variance values. The method (200) then provides for statistically analyzing (230) the model sub vectors of mean values to provide projection vectors indicating directions of relative maximum variance between the sub vectors and thereafter calculating projection values (240) of the projection vectors. The potential threshold values is determined from analysis of a range of the projection values. Finally a step of creating the decision trees (270) divides the model sub vectors into groups, the groups being leaves of the tree. The decisions are based upon selected threshold values selected from the potential threshold values, the selected threshold values being selected by change in variance between said model sub vectors the variance being determined from said mean values and associated variance values. There is also described a method for speech recognition (300) that uses the decisions trees created by the method.

The invention belongs to the technical fields of computer application and artificial intelligence, and relates to a deep reinforcement learning-based behavior control strategy of a yellow-peach stonedigging robot. For the problem that in traditional mechanical control methods, behavior control on yellow-peach stone digging robots is difficult to effectively carry out, the invention provides a deep reinforcement learning-based method carrying out behavior control on the yellow-peach stone digging robot with a visual function, and aims to improve working performance thereof. According to the method, perception ability of deep learning and decision-making ability of reinforcement learning are exerted, the robot is enabled to utilize deep learning to identify peach stone status, and then a single chip microcomputer is guided through a method of reinforcement learning to control motors to dig out a peach stone to finally complete a stone digging task. The method has advantages for utilizing machinery to replace manpower labor of stone digging tasks.