77 results about "Adaptive decision making" patented technology

An improved empirical model-based surveillance or control system for monitoring or controlling a process or machine provides adaptation of the empirical model in response to new operational states that are deemed normal or non-exceptional for the process or machine. An adaptation decision module differentiates process or sensor upset requiring alerts from new operational states not yet modeled. A retraining module updates the empirical model to incorporate the new states, and a pruning technique optionally maintains the empirical model by removing older states in favor of the added new states recognized by the model.

A bone distraction device applies guided incremental forces to opposing bone segments for the purpose of generating native bone in an osteotomy site (distraction osteogenesis). The bone distraction device automatically adjusts the rate of the distraction utilizing feedback received from the ultrasound transducer and other sensors, using an adaptive decision algorithm(s). A wireless transmitter allows for remote guidance, feedback and monitoring.

An adaptive decision process is disclosed for more effectively and efficiently determining and conducting information gathering and evaluation associated with decisions. The adaptive decision process integrates decision analysis, value of information analysis, design of experiment models, and the inferencing of gathered information, including experimental results. The process enables an automatic, adaptive, closed-loop process for attaining additional information and assimilating the attained information into the decision model.

A fault diagnosis method based on OLPP feature reduction in rotary machine fault diagnosis field includes that a vibration signal is performed with EMD to construct Shannon entropy to obtain high-dimensional feature vectors, and then OLPP is adopted reduce the high-dimensional vectors to low-dimensional feature vectors which are inputted to a Morlet MWSVM for fault recognition. The OLPP preserving local and overall structure retains the low-dimensional internal features of the nonlinear manifold structure, and the MWSVM has self-adaptive decisive power and can be used as a terminal classifier. The invention sufficiently excels the advantages of EMD, OLPP and MWSVM respectively in fault feature extraction, information compression and pattern recognition, not only realizes the full automatic process from fault feature extraction to fault diagnosis, but also has high fault diagnosis accuracy and self-adaptive diagnosis capacity.

The invention discloses an adaptive control method for driving and belongs to the technical field of intelligent driving and artificial intelligence. The adaptive control method includes the steps that a historical driving data set is obtained and divided into a training set, a testing set and a verifying set; a deep reinforcement learning algorithm based on a deep convolution neural network is adopted for constructing a network model used for driving control; training set data are used for training the network model, gradient iteration of a cost function is used for repeatedly training the network model, and the optimized network model is obtained; the testing set and the verifying set are used for verifying the performance of the optimized network model, and the network model meeting theperformance is adopted as an adaptive decision model; and the adaptive decision model is used for processing real-time environment data collected currently, and a driving decision is made. By means of the adaptive control method for driving, the driving decision made by people in a true complex environment and corresponding driving actions can be better simulated.

The communication device according to the present invention includes a communication condition detection unit that detects communication conditions with regard to communication with other communication devices, a link adaptation decision-making unit that makes a decision as to whether or not link adaptation needs to be executed in correspondence to the communication conditions, a link adaptation parameter selection unit that selects a link adaptation parameter to be used for the link adaptation when the link adaptation is judged to be necessary, a transfer rate decision-making unit that makes an advance decision as to whether or not the data transfer rate is to become, lower compared to the pre-link adaptation state if the link adaptation is executed based upon the selected link adaptation parameter and a link adaptation unit that executes the link adaptation based upon the selected link adaptation parameter if the data transfer rate is judged not to become lower.

An echo canceling circuit comprising a double talk detector, an upper band signal filter configured to pass only near-end upper band signals to the double talk detector and remove lower band signals, an adaptive filter circuit, a control circuit operatively coupled to the double talk detector and to the adaptive filter circuit, and a threshold estimator configured to iteratively calculate an upper adaptive decision threshold value and a lower adaptive decision threshold value. The double talk detector declares near-end speech to be present if an estimated power level of the upper band signals exceeds the upper adaptive decision threshold value, and declares the near-end speech to be absent if the estimated power level of the upper band signals falls below the lower adaptive decision threshold value for a predetermined number of iterative cycles.

The invention provides a smart city big data management system, and relates to the technical field of smart cities, and the system comprises: a big data center station which is used for collecting city big data provided by city infrastructures, carrying out the preprocessing of the city big data to obtain preprocessed big data, and storing the preprocessed big data to form a city brain database; acollaborative sharing platform which is used for carrying out capacity encapsulation and protocol adaptation on the capacity components, abstracting the capacity components into atomic energy, and providing services to the outside through an open communication interface so as to support business applications of different technical systems to call and interact with the preprocessed big data; and an auxiliary decision application which is used for carrying out artificial intelligence analysis according to a pre-generated adaptive decision analysis engine set and the preprocessed big data, and displaying an analysis result to an urban manager through an urban instrument panel so as to monitor an urban operation state in real time and provide technical support for decision making of the urbanmanager. According to the invention, the city can self-discover problems to the greatest extent and assist in solving the problems.

Disclosed are a method and apparatus for fact collection and analysis whereby relevant facts can be used in multiple implementations to accomplish due diligence or decision making processes, and used for automated adaptive quantification, assessment, analysis, rating and scoring of due diligence or decision options for a business, entity or organization in a proactive and reactive manner. Due diligence or decision topics such as, but not limited to, mergers, acquisitions, debt or equity financings, strategic planning, risk assessments, audits of the processes of the organization including human resources, regulated activities, intellectual property, contracting and other business practices may benefit from the use and application of the system.

A system for processing credit applications. An adaptive decision engine produces a score determined by the risk associated with a credit application. The credit application is automatically approved or rejected if the score is above or below certain thresholds, and is otherwise referred for manual processing. The system may include a rules system that automatically approves applications if all conditions are met, and generates exceptions if all conditions are not met and refers the application to the adaptive decision engine for further processing. A decision to approve or reject an application may be based on a combination of the risk score and number of exceptions.

An improved decoding technique useful for hard decision decoding, such as quadrature phase shift keying (PSK) and quadrature amplitude modulation (QAM), as well as soft-decision techniques, such as Viterbi decoding and trellis decoding. The system in accordance with the present invention provides adaptive decision regions for hard-decision decoding techniques and adaptive metrics for soft-decision detection techniques in which the decision boundaries and reference constellations, respectively are optimized in order to minimize the bit error rate (BER). In particular, the decision boundaries and metrics are optimized based on the locations of the received constellation points. By adaptively adjusting the decision boundaries and metrics, the BER can be greatly improved with the need for nonlinear predistortion at the transmitter thus reducing the hardware complexity and weight of the transmitter which provides additional benefits in applications, such as satellite communication systems, where the transmitter is located on the satellite.

A method and apparatus of recognizing a facial expression using a local feature-based adaptive decision tree are provided. A method of recognizing a facial expression by a facial expression recognition apparatus may include splitting a facial region included in an input image into local regions, extracting a facial expression feature from each of the local regions using a preset feature extracting algorithm, and recognizing a facial expression from the input image using the extracted facial expression feature based on a decision tree generated by repeatedly classifying facial expressions into two classes until one facial expression is contained in one class and determining a facial expression feature for a corresponding classification.

A system and method for providing adaptive decision support is disclosed in further detail herein. An embodiment of this system includes a local outcome report that documents a medical outcome, a difference engine that compares the local outcome report to a national outcome report and generates a localized rule, and a rule engine that receives patient data and applies the localized rule from the difference engine to the patient data to produce a prescribed treatment. An embodiment of the method disclosed herein begins with receiving local patient data and outcome data. The outcome data is compared to national outcome data and a local rule is generated from the comparison. Next, the local patient and the outcome data is analyzed and a standard rule is generated based upon this analysis. The standard rule and the local rule are then applied to the local patient data to generate a prescribed treatment.

The invention discloses a database query optimization method. The database query optimization method comprises a connection sequence selector and a self-adaptive decision network, wherein the connection sequence selector is used for selecting an optimal connection sequence in the query plan and comprises a new database query plan coding scheme, and codes are in one-to-one correspondence with the connection sequence; a value network which is used for predicting the execution time of the query plan, is trained by the query plan and the corresponding real execution time, and is used for reward feedback in Monte Carlo tree search; a Monte Carlo tree search method which is used for simulating and generating multiple different connection sequences, evaluating the quality of the connection sequences through a connection sequence value network, and returning a recommended connection sequence after preset exploration times are reached. And the adaptive decision network is used for distinguishing whether the query statement uses the connection sequence selector or not, so that the overall performance of the optimization system is improved. According to the method and the system, the limitation of a traditional query optimizer can be effectively avoided, and the database query efficiency is improved.

Methods, apparatus, systems, and software for architectures and mechanisms to accelerate tuple-space search with integrated GPUs (Graphic Processor Units). One of the architectures employs GPU-side lookup table sorting, under which local and global hit count histograms are maintained for work groups, and sub-tables containing rules for tuple matching are re-sorted based on the relative hit rates of the different sub-tables. Under a second architecture, two levels of parallelism are implemented: packet-level parallelism and lookup table-parallelism. Under a third architecture, dynamic two-level parallel processing with pre-screen is implemented. Adaptive decision making mechanisms are also disclosed to select which architecture is optimal in view of multiple considerations, including application preferences, offered throughput, and available GPU resources. The architectures leverage utilization of both processor cores and GPU processing elements to accelerate tuple-space searches, including searches using wildcard masks.

The invention discloses an HEVC intra-frame coding fast mode decision algorithm based on ResNet. According to the method, the CU depth prediction process and the PU mode number screening process are optimized, so that the computational complexity of CTU division and optimal mode decision in the intra-frame coding process is reduced; The proposed HEVC intra-frame coding optimization algorithm basedon ResNet mainly comprises: an HEVC intra-frame coding optimization algorithm, The method comprises the steps that (1) adopting a residual error network in the technology, using a CU original depthprediction result and a corresponding video sequence data set for training to obtain a ResNet model, then acquiring a CU depth prediction value through the model, and replacing an HEVC original CU division process; and (2) on the basis of the prediction depth, proposing a PU texture direction analysis and mode adaptive decision-making combined algorithm, grouping 35 intra-frame prediction modes, and performing adaptive mode selection, thereby simplifying the existing mode decision-making process of HEVC.

The invention provides a high-resolution remote sensing image multi-scale self-adaptive decision fusion segmentation method. Firstly, a series of increasing scale parameters are set by applying a fractal network evolution segmentation algorithm so that a multi-scale segmentation sequence is obtained; secondly, regional multi-scale Moran's I index and critical segmentation scale and under-segmentation Moran's I index thresholds are defined; and finally under-segmentation of regions is judged one by one with the maximum segmentation scale acting as an initial critical scale, if the judgment result is yes, down-scaling is performed through recursion in turn till the minimum segmentation scale layer or the current layer without under-segmentation region with the first time of minimum scale of the multi-scale Moran's I index acting as a new critical scale, and finally a segmentation result is obtained through combination of spatial inheritance relationship between multi-scale segmentation layers. Multi-scale segmentation information is fused, the contradiction between over-segmentation and under-segmentation and easy segmentation and accuracy can be effectively reduced, and the method can be widely applied to the field of object-oriented project target recognition.

Providing adaptive channel state feedback (CSF) reports in discontinuous reception (DRX) scenarios in a power-efficient manner. The described algorithm may be able to make adaptive decisions to carry over the CSF from previous DRX cycles based on channel conditions, DRX cycle length, and / or the requirements of CSF reporting for current DRX cycle. The proposed approach can allow for more efficient power consumption related to CSF reports in DRX scenarios where new CSF reports have little or no impact to throughput.

The invention relates to a digital signal adaptive code rate estimation method and device based on multi-wavelet basis combination. The method comprises two steps of preprocessing and adaptive code rate estimation. In the preprocessing step, original I / Q data of an observation window is used as input, and a bandwidth rough estimation result and baseband I / Q data in a specified oversampling range are output through fast Fourier transform, frequency estimation, bandwidth rough estimation, down-conversion and down-sampling processing; in the adaptive code rate estimation step, the wavelet transform scale is adaptively selected according to a bandwidth rough estimation result, wavelet transform is carried out on the baseband I / Q data, jump information after transform is extracted, and finally,the code rate of a digital signal and a wavelet basis for signal acquisition are output through adaptive decision. According to the method, the problems of single wavelet basis application range andlow estimation precision of the traditional wavelet transform-based digital signal code rate estimation are solved, the accuracy of the digital communication system code rate estimation under the non-cooperative condition is improved, and the method can be applied to the fields of spectrum monitoring, electromagnetic reconnaissance and the like.

An inter-frame fast prediction algorithm for multi-scene adaptive decision tree selection comprises the following steps: 1) a preliminary part: inputting a video test sequence, and performing probability statistics on each PU mode of CU inter-frame prediction in the video sequence to obtain the probability of each PU mode of the inter-frame prediction; 2) a training part: inputting video test sequences into several scene categories, collecting relevant features of CU blocks for video sequences of each scene category, further filtering and optimizing the collected features, and generating 8x8,16x16, 32x32, 64x64 decision trees based on the optimized features for each scene category; and 3) an execution part: inputting a segment of video sequences composed of various scenes, dividing the input video according to the transformation of the scenes in the video, collecting and optimizing correlation features of the CU blocks of the respective divided video sequences, determining whether thePU prediction mode after the traversal is continued, and ending the inter-frame prediction.

The invention discloses a panoramic surround view system for cars and an adaptive light supplementing method thereof, in the method, by performing grey level histogram brightness analysis on an original video image captured by surround view cameras mounted at two sides of a car body, intensity of external environment light of the surround view cameras mounted at left and right sides of the car body is judged, whether to perform light supplementing operation for the surround video cameras is decided adaptively, a surround video camera light supplementing lamp group is arranged, and thus, adaptive light supplementing for the panoramic surround view system is realized. According to the method and the system provided by the invention, a function of adaptively supplementing light for the surround view cameras mounted at two sides of the car body is realized, and a defect of failure of the panoramic surround view system in a scene where the external light is not enough or an obvious externallight source is not provided is overcome effectively.

Providing adaptive channel state feedback (CSF) reports in discontinuous reception (DRX) scenarios in a power-efficient manner. The described algorithm may be able to make adaptive decisions to carry over the CSF from previous DRX cycles based on a comparison between an offset at which CSF values are stable and an offset at which a CSF report is to be sent to a base station. If the CSF values are not stable by the time the CSF report is to be sent, a CSF report from a prior DRX cycle may be used. Alternatively, if the CSF value are stable by the time the CSF report is to be sent, a determination may be made to either generate a new CSF report or use a prior CSF report. The latter determination may be made based on various criteria, including channel conditions and DRX cycle length.

The invention provides a remote sensing image earthquake damage building identification method based on a decision tree and feature optimization, and aims to solve the problems and limitations existing in feature modeling and random forest classification of earthquake damage buildings only depending on post-earthquake remote sensing images under the condition of lack of pre-earthquake reference information. Firstly, a potential building object set is extracted in combination with image segmentation and non-building screening rules; on the basis, an adaptive decision tree quantity extraction strategy based on classification accuracy curve fluctuation discrimination is provided; meanwhile, under the guidance of the characteristic importance index, three types of characteristics of spectrum,texture and geometrical morphology are screened to obtain a representative earthquake damage characteristic set; and finally, the earthquake damage building is identified based on the constructed optimized random forest model. Experiments on four groups of different remote sensing images show that the method shows excellent performance in earthquake damage building identification in a complex scene after an earthquake, and the total precision can reach more than 85%.

Disclosed are a method and apparatus for fact collection and analysis whereby relevant facts can be used in multiple implementations to accomplish due diligence or decision making processes, and used for automated adaptive quantification, assessment, analysis, rating and scoring of due diligence or decision options for a business, entity or organization in a proactive and reactive manner. Due diligence or decision topics such as, but not limited to, mergers, acquisitions, debt or equity financings, strategic planning, risk assessments, audits of the processes of the organization including human resources, regulated activities, intellectual property, contracting and other business practices may benefit from the use and application of the system.

The invention provides a color number self-adaptive decision making method and an image compression method, and relates to the technical field of digital signal processing. The color number self-adaptive decision making method includes steps of researching given initial small color numbers; calculating corresponding image quantizing distortion and distortion change rate; and determining that the corresponding color number is the color number required for the final quantization if the image quantizing distortion and the distortion change rate is smaller than a threshold valve or the color number excesses the given maximum color number. The image compression method includes steps of acquiring reasonable color number of given images by the color number self-adaptive decision making method; then quantizing the images into given color number by a color self-adaptive quantizing algorithm; and compressing the images by means of PNG compression algorithm. The image compression method has the advantage of high compression efficiency and the like.

The invention relates to the field of unmanned aerial vehicles, in particular to an adaptive decision-making method for an unmanned aerial vehicle in response to the harsh environment, and the corresponding unmanned aerial vehicle. The adaptive decision-making method comprises the steps that the meteorological conditions within the preset range of the unmanned aerial vehicle are detected based ona sensor pre-arranged on the unmanned aerial vehicle, and the meteorological conditions are converted into corresponding meteorological data; the meteorological data are guided into a harsh meteorological condition mathematical model, and the threat degree of the meteorological conditions to the unmanned aerial vehicle is calculated by combining with the flight parameters of the unmanned aerial vehicle; and the flight mission of the unmanned aerial vehicle is evaluated by an artificial intelligence expert system according to the threat degree, corresponding execution strategies of the unmannedaerial vehicle are obtained, and the unmanned aerial vehicle is controlled to conduct mission re-planning according to the execution strategies. The unmanned aerial vehicle can make decisions autonomously during encountering the harsh environment, especially the harsh meteorological conditions, and thus the unmanned aerial vehicle can better complete flight operation.

The invention provides an intelligent agent adaptive decision generation method and system based on deep reinforcement learning, an agent adaptive decision problem is studied based on a deep reinforcement learning SoftActor-Cr-it ic (SAC) algorithm, the SAC algorithm is improved for problems occurring in a training process, and SAC + PER, SAC + ERE and SAC + PER + ERE algorithms are provided. The intelligent agent self-adaptive decision-making problem is solved by using the powerful perception ability of deep learning and the efficient decision-making ability of reinforcement learning, and an intelligent agent is trained through a deep reinforcement learning algorithm, so that the intelligent agent summarizes experience in the process of interacting with the environment, and thus the understanding of the intelligent agent on specific behavior application is formed; meanwhile, an anti-interception task of the unmanned aerial vehicle in a simulation environment is taken as a carrier, and the effectiveness of the algorithm is verified.

A method and apparatus is disclosed for adaptively determining threshold values in a decision device, such as a decision slicer. In one embodiment, a slicer receives one or more updated threshold values during operation to adaptively accommodate changes in the received signal or the channel that may occur over time. The updated threshold value is based on a MIN value and a MAX value. The MIN values represent a value related to a recently received input that was determined to qualify for a particular slicer output and was less than the particular slicer output value. The MAX values represent a value related to a recently received input that was determined to qualify for a particular slicer output and was greater than the particular slicer output value. In one embodiment, the MAX and MIN value computation is based on scaling factors that dampen the rate of change of the threshold value.

Providing adaptive channel state feedback (CSF) reports in discontinuous reception (DRX) scenarios in a power-efficient manner. The described algorithm may be able to make adaptive decisions to carry over the CSF from previous DRX cycles based on channel conditions, DRX cycle length, and / or the requirements of CSF reporting for current DRX cycle. The proposed approach can allow for more efficient power consumption related to CSF reports in DRX scenarios where new CSF reports have little or no impact to throughput.