Patents
Literature
Eureka-AI is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Eureka AI

748 results about "State-space representation" patented technology

In control engineering, a state-space representation is a mathematical model of a physical system as a set of input, output and state variables related by first-order differential equations or difference equations. State variables are variables whose values evolve through time in a way that depends on the values they have at any given time and also depends on the externally imposed values of input variables.

System and method of collision avoidance using intelligent navigation

A system and method of intelligent navigation with collision avoidance for a vehicle is provided. The system includes a global positioning system and a vehicle navigation means in communication with the global positioning system. The system also includes a centrally located processor in communication with the navigation means, and an information database associated with the controller, for identifying a location of a first vehicle and a second vehicle. The system further includes an alert means for transmitting an alert message to the vehicle operator regarding a collision with a second vehicle. The method includes the steps of determining a geographic location of a first vehicle and a second vehicle within an environment using the global positioning system on the first vehicle and the global positioning system on the second vehicle, and modeling a collision avoidance domain of the environment of the first vehicle as a discrete state space Markov Decision Process. The methodology scales down the model of the collision avoidance domain, and determines an optimal value function and control policy that solves the scaled down collision avoidance domain. The methodology extracts a basis function from the optimal value function, scales up the extracted basis function to represent the unscaled domain, and determines an approximate solution to the control policy by solving the rescaled domain using the scaled up basis function. The methodology further uses the solution to determine if the second vehicle may collide with the first vehicle and transmits a message to the user notification device.
Owner:TOYOTA MOTOR CO LTD

Heterogeneous multi-agent collaborative decision-making method based on depth deterministic policy gradient

InactiveCN108600379AAchieve collaborative decision-makingData switching networksState spaceComputer science
The invention relates to a heterogeneous multi-agent collaborative decision-making method based on a depth deterministic policy gradient, belonging to the collaborative decision-making field of a heterogeneous intelligent unmanned system, comprising the following steps of: firstly, defining heterogeneous multi-agent characteristic attributes and reward and punishment rules, defining multi-agent state space and action space, and constructing multi-agent motion environment for collaboratively making decision; then, establishing an actor module for decision-making action and a critic module for evaluating feedback based on the depth-deterministic strategy gradient algorithm, and training the parameters of the learning model; using the trained model to obtain the multi-agent state sequence; and evaluating the situation of the multi-agent motion state sequence according to the reward and punishment rules set in the environment. The invention may construct reasonable sports environment according to actual needs, achieve the purpose of intelligent sensing and strategy optimization through the synergy between multiple agents in the system, and has a positive effect on the development of the unmanned system field in China.
Owner:INST OF SOFTWARE - CHINESE ACAD OF SCI

Heterogeneous cloud wireless access network resource allocation method based on deep reinforcement learning

ActiveCN110493826AMaximize total throughputMeet stability requirementsNetwork traffic/resource managementNonlinear approximationSmall sample
The invention relates to a heterogeneous cloud wireless access network resource allocation method based on deep reinforcement learning, and belongs to the technical field of mobile communication. Themethod comprises the following steps: 1) taking queue stability as a constraint, combining congestion control, user association, subcarrier allocation and power allocation, and establishing a random optimization model for maximizing the total throughput of the network; 2) considering the complexity of the scheduling problem, the state space and the action space of the system are high-dimensional,and the DRL algorithm uses a neural network as a nonlinear approximation function to efficiently solve the problem of dimensionality disasters; and 3) aiming at the complexity and the dynamic variability of the wireless network environment, introducing a transfer learning algorithm, and utilizing the small sample learning characteristics of transfer learning to enable the DRL algorithm to obtain an optimal resource allocation strategy under the condition of a small number of samples. According to the method, the total throughput of the whole network can be maximized, and meanwhile, the requirement of service queue stability is met. And the method has a very high application value in a mobile communication system.
Owner:CHONGQING UNIV OF POSTS & TELECOMM

NEXT-GENERATION BANDWIDTH MANAGEMENT CONTROL SYSTEMS FOR MULTIPLE-SERVICE CALLS, SESSIONS, PACKET-LEVEL PROCESSES, AND QoS PARAMETERS - PART 1: STRUCTURAL AND FUNCTIONAL ARCHITECTURES

System and method for addressing immense, long-standing problem of bandwidth management, for example, in enterprise networks, VPNs, real-time and stored video services, mobile applications, wireless networks, and cloud computing applications. Described features include an automatic closed-loop control system infrastructure encompassing multiple time-scales and performing control actions optimized to the extent possible with respect to administrator-provided performance metrics. One aspect utilizes available or innovatively accessible means of session and QoS control (settings in configuration files, gateway APIs, QoS parameters, application bit-rate settings, etc.) within the context of practical multiple-vendor products in evolving multiple-service networks. Another aspect utilizes available or innovatively accessible means of session and QoS observations (values in reporting log files, gateway APIs, network monitoring, etc.) within the context of practical multiple-vendor products in evolving multiple-service networks. Traffic-measurement controlled adaptive reservations for distributed myopic single-service gatekeepers effectively shapes the permitted state-space boundary over a range of arbitrary curvatures.
Owner:AVISTAR COMM

Reinforcement learning based air combat maneuver decision making method of unmanned aerial vehicle (UAV)

The invention provides a reinforcement learning based air combat maneuver decision making method of a UAV. A motion model of an airplane platform is created; principle factors that influence the air combat situation are analyzed; on the basis of the motion model and analysis on the air combat situation factors, a dynamic fuzzy Q learning model of air combat maneuver decision making is designed, and essential factors and an algorithm flow of reinforcement learning are determined; a state space of air combat maneuver decision making is fuzzified and serves as state input of reinforcement learning; typical air combat motions are selected as basic motions of reinforcement learning, and the triggering intensities of fuzzy rules are summed in a weighted manner, and a continuous motion space is covered; and on the basis of an established air combat dominant function, a return value of reinforcement learning is set in a rewards and punishment values weighing-superposing method. Thus, the autonomous maneuver decision making capability of the UAV during air combat can be improved effectively, the robustness is higher, an autonomous searching optimization performance is higher, and decisionsmade by the UAV are improved continuously in continuous simulation and learning.
Owner:NORTHWESTERN POLYTECHNICAL UNIV

Method and device for estimating power battery charge state on line

The invention discloses a method and device for estimating a power battery charge state on line. The method comprises the steps that a capacity correction factor of a power battery is calculated according to the accumulated released energy of the power battery, the internal resistance of the power battery is calculated, a state space model of the power battery is established, and the state space model of the power battery is substituted into an unscented Kalman filter to obtain the estimated value of the battery charge state. According to the method and device for estimating the power battery charge state on line, the capacity correction factor of the power battery is calculated according to the accumulated released energy of the power battery, the accumulated released energy of the power battery is the sum of energy consumed by load acting and energy consumed by the internal resistance from the first time up to now, the accumulated released energy of the power battery can reflect the using history of the battery like the cycle index and is more beneficial to accurate measuring in numerical value, and therefore the method of correcting the related parameters of the battery through the accumulated released energy of the power battery is better in operability and is more accurate in estimated value.
Owner:SHENZHEN GRADUATE SCHOOL TSINGHUA UNIV +1

Method for predicting dynamic risk and vulnerability under fine dimension

The invention relates to a method for predicting dynamic risk and vulnerability at fine scale and belongs to the scientific field of global information. The method is mainly characterized in that an optimized Bayesian network is searched from multi-source heterogeneous spatiotemporal data on the basis of a grid format with certain resolution at fine scale; domain knowledge is combined to improve the network; therefore, the uncertain estimation of disaster risk and the vulnerability, namely probability estimation, is carried out. In the method, a nuclear density method is put forward to train a sample according to a sample derivative grid; an optimized discretization method is put forward to discretize continuous variables so as to provide discrete state space input for the network; a simulated annealing optimization algorithm is adopted to search an optimized network structure; and a method of accurate reasoning combined with approximate reasoning to predict the probabilities of risk and the vulnerability is adopted. The method provided by the invention can position the positions of the disaster risk and the vulnerability in real time at the fine spatial scale, estimate the spatial distribution of the risk probability and has important theoretical significance and practical value for improving the effects on the reduction and relief of disaster and building an intelligent public emergency pre-warning system by the state.
Owner:INST OF GEOGRAPHICAL SCI & NATURAL RESOURCE RES CAS

Dynamic spectrum access method based on policy planning constrain Q study

The invention provides a dynamic spectrum access method on the basis that the policy planning restricts Q learning, which comprises the following steps: cognitive users can divide the frequency spectrum state space, and select out the reasonable and legal state space; the state space can be ranked and modularized; each ranked module can finish the Q form initialization operation before finishing the Q learning; each module can individually execute the Q learning algorithm; the algorithm can be selected according to the learning rule and actions; the actions finally adopted by the cognitive users can be obtained by making the strategic decisions by comprehensively considering all the learning modules; whether the selected access frequency spectrum is in conflict with the authorized users is determined; if so, the collision probability is worked out; otherwise, the next step is executed; whether an environmental policy planning knowledge base is changed is determined; if so, the environmental policy planning knowledge base is updated, and the learning Q value is adjusted; the above part steps are repeatedly executed till the learning convergence. The method can improve the whole system performance, and overcome the learning blindness of the intelligent body, enhance the learning efficiency, and speed up the convergence speed.
Owner:COMM ENG COLLEGE SCI & ENGINEEIRNG UNIV PLA

Matrix converter switch open circuit fault diagnosis method based on prediction control

The invention discloses a matrix converter switch open circuit fault diagnosis method based on prediction control. The method comprises the following steps of under the condition that a matrix converter works normally, establishing relation models between an output voltage and an input voltage, and between an input current and a load current; determining numbers of all the switch combination states of the matrix converter; establishing a state space model and acquiring predicted values of the load current, the input current and the input voltage in a next sampling period; determining an evaluation function; using a finite set model prediction control strategy to select a switch state capable of making an evaluation function value be minimum in all the switch combination states in each sampling period and using the switch state as the switch state of the next sampling period; establishing the state space model and acquiring a change rule of a fault phase load current after a switch open circuit fault is generated; carrying out on-line monitoring on the load current, according to a finite set model, predicting and controlling the selected switch state so as to carry out fault diagnosis, and identifying a fault switch position. By using the method, an open circuit fault can be timely diagnosed.
Owner:CENT SOUTH UNIV

Multiwave beam-based depth-sounding joint inversion method for sound velocity profile and seafloor topography

The invention discloses a multiwave beam-based depth-sounding joint inversion method for a sound velocity profile and a submarine topography. The method comprises the following steps of: (1) transmitting multiwave beams to the seafloor of a sounded sea area through a transmitting transducer array of a multiwave-beam depth sounding system, and receiving echo signals through a receiving transducer array of the multiwave-beam depth sounding system; (2) obtaining the arrival angle and the arrival time of echoes according to the received echo signals by using the multiwave-beam depth sounding system; (3) establishing a state space model formed by using a state equation and a sounding equation; (4) obtaining inversion values of the sound velocity gradient and the seafloor depth of the sounded sea area by using a sequential filter method according to the established state space model and the arrival angle and the arrival time of received echoes, and obtaining an estimate of the sound velocity profile of the sounded sea area by using the inversion value of the sound velocity gradient; and further, calculating the water temperature profile of the sounded sea area by using the estimate of the sound velocity profile. According to the multiwave beam-based depth-sounding joint inversion method for the sound velocity profile and the submarine topography, disclosed by the invention, the estimates of the sound velocity profile and the depth of the seafloor can be obtained quickly and accurately.
Owner:北京南界电子技术有限公司 +1

Dynamic and static combined software security test method

InactiveCN102360334AIncreased security testing productivitySolve the path space explosion problemSoftware testing/debuggingCall graphBasic block
The invention relates to a computer software security test method, and in particular relates to a dynamic and static combined software security test method. The test method comprises the following steps: firstly carrying out disassembly and intermediate language transformation on an executable program so as to generate a function call graph (CG) and a control flow graph (CFG) of a file; finding out a vulnerable point of a system by means of static analysis of the function CG, and constructing a test case generation execution path by virtue of a dynamic analysis method; searching a called function based on the function CG, finding out a specific path for triggering the vulnerable point on a first-grade basic block according to the CFG if the function is located on the generated execution path, and then ending the loophole mining process corresponding to the sensitive point; and if the path can not be found, reconfiguring the test case generation execution path and then searching the called function in a cyclic manner. The dynamic and static combined software security test method has the advantages of better solving the problem of path state space blast caused by single Fuzz dynamic test, and greatly improving the path coverage hit rate and the software test analysis efficiency.
Owner:THE PLA INFORMATION ENG UNIV

SOC (state of charge) estimation method for controlling equivalent charging and discharging of lithium battery

InactiveCN108594135AHigh precisionRealize online statisticsElectrical testingObservational errorState parameter
The invention provides an SOC (state of charge) estimation method for controlling equivalent charging and discharging of a lithium battery. The SOC estimation method includes: establishing a first-order RC (resistor and capacitor) equivalent circuit model of a single lithium battery, and determining a state equation and an observation equation of a lithium battery system; determining a Kalman filtering discrete state space model, a state parameter variable and an observation parameter variable; updating an predicted value of the state parameter variable and a covariance matrix of measuring errors, and acquiring an innovation sequence according to an observed voltage value at the end of the lithium battery; introducing a self-adaptive fading factor to track and correct a predicted covariance matrix of the lithium battery system; calculating a Kalman filtering gain matrix of the discrete state space model, and updating the optimal estimated value and an error covariance matrix value at the present moment; acquiring a statistical property of process noise; acquiring an SOC estimation value at the present moment, and putting the parameters of the present moment in recursive iteration calculation of strong-tracking self-adaptive Kalman filtering for the next moment. By the aid of the SOC estimation method, real-time and precise estimation of SOC of the lithium battery is realized.
Owner:NANJING UNIV OF SCI & TECH

Storage battery surplus energy monitoring method and monitoring device

The invention discloses a storage battery surplus energy monitoring method and a monitoring device. The monitoring device comprises the following steps of: constructing a storage battery state space model; utilizing off-line sample data to carry out off-line identification on the state space model of a storage battery, and utilizing a model parameter off-line identification result and on-line sample data to carry out on-line identification on the state space model of the storage battery; constructing a state observer according to a model parameter on-line identification result and a storage battery state space model, and carrying out observation on a storage battery open-circuit voltage; and obtaining the surplus energy of the storage battery according to the corresponding relation of the storage battery open-circuit voltage and the surplus energy of the storage battery. The monitoring method and the monitoring device provided by the invention can carry out monitoring on the surplus energy of the storage battery through a method of the on-line identification model parameter, effectively reduce storage battery time-varying nonlinear characteristic, the influence of measuring errors and noises and the like on a surplus energy monitoring result, improve monitoring accuracy, reduce measuring errors, and strengthen the instantaneity of the monitoring.
Owner:QINGDAO TECHNOLOGICAL UNIVERSITY
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products