25 results about "Total error" patented technology

The invention discloses a hand-eye calibration method suitable for a five-axis series-parallel mechanical arm, which comprises the following steps of: analyzing coordinate conversion error distribution characteristics according to an initial hand-eye calibration matrix calculation result, and separating a total error into two components, namely a translation error and a rotation error through an error vector decomposition algorithm, obtaining distribution rules and variation trends of the errors in different spatial directions; establishing a calibration precision evaluation system according to the verification result, calculating the repeated positioning precision and the absolute positioning precision of the system by counting and analyzing the position deviation data of multiple tests, and obtaining the system performance index in the current calibration state; a self-adaptive updating mechanism is established for system performance indexes, local updating of calibration parameters is automatically triggered when environment temperature change or small deformation of a mechanical structure is detected, and a compensation model is adjusted in an incremental learning mode to avoid comprehensive recalibration.

The invention relates to the technical field of computers, in particular to a precision configuration method and system for multi-level calculation, a terminal and a storage medium, and the method comprises the steps: configuring initial precision for each level in the multi-level calculation; in the multi-level calculation process, when each level is executed, recording the absolute deviation between the theoretical value and the alternative value of the current level calculation as the error of the current level, meanwhile, accumulating to obtain the total error when the multi-level calculation is executed to the current level, and judging whether the total error meets a preset threshold condition or not; and if not, reconfiguring the precision of the corresponding level according to a preset precision adjustment strategy until the corrected total error meets a preset threshold condition. The method is used for solving the problem that the calculation result of the final calculation layer of the multi-level calculation with more levels is easy to increase and is not beneficial to the total error control target in the multi-level calculation scene.

The present invention provides a computer program, a method for generating a trained model, an information processing method, and an information processing device that can be expected to perform searching of time-series data related to substrate processing.　This computer program causes a computer to execute a process for generating a trained model that receives time-series data related to substrate processing as an input and outputs feature quantities of the time-series data, the computer program causing the computer to execute a process in which: training data in which time-series data related to substrate processing and a plurality of correct answer values related to a process performed on the basis of the feature quantities of the time-series data are associated with each other is acquired; time-series data of the acquired training data is inputted to the trained model, and feature quantities outputted by the trained model are acquired; a plurality of errors are calculated on the basis of the acquired feature quantities, the correct answer values for the training data, and a plurality of loss functions; the total error is calculated on the basis of the calculated plurality of errors; and the parameters of the trained model are updated on the basis of the calculated total error.

A method, apparatus and product comprising: obtaining alternative quantum circuits implementing a quantum program, each having a respective design error; identifying, for each quantum circuit, respective quantum resources that are used by the quantum circuit and are associated with respective hardware errors of a quantum computer; determining a total hardware error bound for the quantum circuit; determining a total error bound for the quantum circuit; applying a search algorithm over a plurality of total error bounds of the plurality of alternative quantum circuits, to identify an optimal total error bound that is associated with an optimal quantum circuit; and providing the optimal quantum circuit for execution on the quantum computer.

To provide a resolver decoding method, a system, a device, and a storage medium regarding the resolver technical field.SOLUTION: This resolver decoding method comprises: a step of generating a first sine reference signal and a first cosine reference signal; a step of calculating a difference between the first sine reference signal and a real time sine signal of a resolver to acquire a first difference signal and calculating a difference between the first cosine reference signal and a real time cosine signal of the resolver to acquire a second difference signal; a step of inputting the first and second difference signals to a preset error model to acquire an overall error; a step of repeating the step of generating first sine and cosine reference signals when it is detected that the overall error is larger than an error threshold value; and a step of setting an angle output from the resolver as a decoding result when it is detected that the overall error is smaller than or equal to the preset error threshold value. This method improves the decoding accuracy of the resolver.SELECTED DRAWING: Figure 1

A system and related method includes a memory and processing circuitry to receive data and corresponding expected error detection code values. The processing circuitry processes data in the at least two portions by calculating and storing error detection code values for the respective portions in the memory. The processing circuitry is then to calculate a total error detection code value based on the respective error detection code values of the at least two portions. When the total error detection code value does not match the expected error detection code value, the processing circuitry is to correct the at least one portion and to correct the at least one portion by calculating an updated error detection code value for a respective one of the corrected portions and calculating an updated total error detection code value based on the updated error detection code value for each corrected portion and the stored error detection code value. The corrected portion is processed.

The invention discloses a cross-domain high-precision impact-free leap second processing method which is suitable for all safety-critical systems (including cross-domain scenes of finance, aerospace, medical treatment, industry and the like), wherein the time synchronization error of the safety-critical systems is less than or equal to 1ns and the time continuity is not interrupted. The core scheme is as follows: pre-calibration before leap second and enabling an initial synchronization error to meet the subsequent calibration requirement, uniformly dividing the total step number corresponding to a transition period into M (a positive integer greater than or equal to 2) sections, calculating a uniform step length, executing adjustment section by section, dynamically sharing an error (an absolute value of a delta E section is less than or equal to 0.000001 [mu] s * the step number of each section), and combining two-way closed-loop calibration (adjusting a time sequence of related steps). The total error in the whole process is smaller than or equal to 0.2 ns (preferably smaller than or equal to 0.1 ns), and the step length is not changed suddenly (the difference value of adjacent steps is smaller than or equal to 0.000001 microseconds). The method supports bidirectional compensation and cross-field seamless adaptation, can meet compliance requirements of different fields without modifying a core scheme, is high in anti-interference capability, low in landing cost and comprehensive in protection range, can effectively prevent various evasion behaviors, and is suitable for various high-precision scenes such as high-frequency transactions, aerospace navigation, medical surgical robots and the like.

Techniques are described for quickly finding a compression mode for BC7 encoding by modelling three sources of error in compression, namely, projection error, endpoint quantization error, and interpolation index quantization error. The mode with the lowest total error is selected for a block to be compressed.

The present application discloses a resolver decoding method, system, device and storage medium, and relates to the technical field of resolvers. [Solution] A resolver decoding method is disclosed, which includes the steps of generating a first sine reference signal and a first cosine reference signal, calculating a difference between the first sine reference signal and a real-time sine signal of the resolver to obtain a first difference signal and calculating a difference between the first cosine reference signal and the real-time cosine signal of the resolver to obtain a second difference signal, inputting the first difference signal and the second difference signal into a predetermined error model to obtain an overall error, repeating the step of generating the first sine reference signal and the first cosine reference signal when it is detected that the overall error is greater than an error threshold, and taking the angle output by the resolver as the decoding result when it is detected that the overall error is equal to or less than the predetermined error threshold. The technical solution of this application aims to solve the technical problem of how to improve the accuracy of resolver decoding.

Embodiments of the present application provide a neural network input adjustment method, device, medium and equipment based on an RRAM array. The method comprises: obtaining input data comprising to-be-input values corresponding to each row in the RRAM array; determining ideal unit current values corresponding to each row in the RRAM array and actual unit current values after weight writing according to a plurality of weights of the neural network; determining a total error of the RRAM array between actual output and ideal output according to the to-be-input values, the ideal unit current values and the actual unit current values corresponding to each row in the RRAM array; if the absolute value of the total error is greater than or equal to a first threshold, calculating a cumulative error corresponding to each row in the RRAM array row by row, and adjusting the to-be-input value corresponding to each row according to the cumulative error corresponding to each row to obtain target input data. The technical solution of the embodiments of the present application improves the accuracy of the calculation result based on the RRAM device, and further ensures the accuracy of the recognition result of the neural network.

Devices, systems, and methods for improving performance of an iterative decoder in a non-volatile memory are described. An example method includes obtaining a hard error percentage, a spare byte percentage, and a total error count associated with a first page in the memory device, determining that the total error count is greater than a threshold value that is determined based on the hard error percentage and the spare byte percentage, and performing, based on the determining, a data relocation operation that relocates at least some data from the first page to a second page different from the first page.

The application provides a camera and IMU joint calibration method, which comprises the following steps: step 1, initializing a target camera and a target IMU to obtain initialization parameters; step 2, constructing a first error L1 and performing one-time iterative optimization by using an LM algorithm; step 3, updating the initialization parameters and the bias of the target IMU acceleration, and determining whether the total error determined by the first error is less than a set threshold value; if yes, entering step 7, otherwise, entering the next step; step 4, obtaining gyro noise and acceleration noise according to the bias of the gyroscope and the acceleration of the target IMU at each time after updating; step 5, constructing a second error according to the acceleration noise and the first error L1; step 6, performing iterative optimization on the second total error L2 by using the LM algorithm, and determining whether the error is less than a preset threshold value; if yes, entering the next step, otherwise, entering step 3; and step 7, obtaining a calibration quantity. The application realizes joint calibration of the camera and the IMU, and has high efficiency and high accuracy.

The invention discloses a leveling example generation method and a teaching auxiliary system, and belongs to the technical field of surveying and mapping teaching. The core of the method comprises the following steps: deducing a theoretical observation value through reverse calculation based on a known benchmark elevation and a measurement section distance; dynamically generating an observation record with a random error in combination with an error parameter within a standard tolerance; for an attachment route, a closed loop and a leveling network, a closing error is automatically distributed and a full-mean error / accidental-mean error is calculated. The system module comprises a route design module which supports diversified format data import and dynamically adjusts route levels and structures; the observation simulation module is used for realizing height difference and distance reverse derivation and error simulation; the handbook output module is used for generating a training handbook containing key field spaces according to a standard format; and the intelligent scoring module automatically verifies the submitted result of the user and scores the submitted result. The method solves the problem of lack of examples in teaching and competition, realizes automatic generation and intelligent evaluation of diversified and standardized leveling training examples, provides an efficient tool for surveying and mapping major teaching and skill competition, and has outstanding practical value.

The invention provides a machine tool expected running track error compensation method and device and a medium, and relates to the technical field of machine tool expected running track error compensation. The method comprises the steps that the complexity of an original running track is determined; if the complexity of the original moving trajectory is higher than a first preset complexity threshold value, performing total error prediction on an expected moving trajectory corresponding to the original moving trajectory subsequently; if the complexity is between a second preset complexity threshold value and a first preset complexity threshold value, dividing the original running track into a plurality of original running sub-tracks; performing error prediction on a track segment corresponding to each target original operation sub-track in the expected operation track; if the complexity of the original moving trajectory is smaller than or equal to a second preset complexity threshold value, error prediction and compensation are not carried out on the expected moving trajectory; on the premise of ensuring the compensation precision of the key part, the error prediction calculation amount of the non-key sub-track can be greatly reduced, and the computing power consumption of a machine tool is reduced.

Techniques are described for quickly finding a compression mode for BC7 encoding by modelling three sources of error in compression, namely, projection error, endpoint quantization error, and interpolation index quantization error. The mode with the lowest total error is selected for a block to be compressed.

The invention discloses a precise error quantification and traceability analysis method and system, and relates to the technical field of precise measurement and error analysis, and the method comprises the following steps: carrying out the layering preprocessing of original data, so as to remove abnormal values, and constructing a comprehensive multi-dimensional error quantification model; based on the multi-dimensional error quantification model, positioning a key error source with the highest contribution to a total error in a mode of combining probabilistic simulation and sensitivity analysis; and analyzing the coupling relationship between different error sources and the influence of the coupling relationship on the total error. According to the accurate error quantification and traceability analysis method provided by the invention, a layered preprocessing strategy is adopted, abnormal values in original data are effectively eliminated, a solid foundation is laid for follow-up analysis, a built multi-dimensional error quantification model comprehensively covers random errors, system errors and truncation errors, and the accuracy of the analysis is improved. The integrity and accuracy of an error quantification result are ensured, and a clear direction is provided for error control.

The invention discloses an error calibration compensation method and system for a three-coordinate measuring instrument, and relates to the technical field of precision measurement. Comprising the following steps: S1, fusing a geometric error compensation sub-model, a thermal deformation error sub-model and a dynamic load error sub-model to determine a total error compensation amount; s2, correcting the original target instruction according to the total error compensation amount, obtaining the actual position coordinates of each axis, and obtaining the final corrected position coordinates through a feedback compensation algorithm; and S3, determining a confidence interval corresponding to each final corrected position coordinate through the digital twin model of the three-coordinate measuring instrument and the uncertainty variable, and determining a corresponding monitoring result through a comparison result between an interval width corresponding to the confidence interval and an interval threshold range. According to the invention, systematic compensation of various error sources of the three-coordinate measuring instrument under different working conditions is realized, a motion instruction can be corrected in advance, and errors are reduced from the source.

To provide functional system and method for data filtering for reducing outlier bias of a trend line.SOLUTION: The present invention is directed to an objective, statistic method of eliminating an outlier from a data set. The method has the steps of determining a bias based on an absolute error, a relative error or both of them, calculating an error value from calculation of data, model coefficients or trend lines, and eliminating an outlier data record if the error value exceeds a reference set by a user. For the purpose of a repeated calculation such as an optimization method, the eliminated data is re-applied to the model for use in calculation of a new result upon each repeated calculation. With use of a model value for a completed data set, a new error value is calculated, and an outlier bias reduction process is re-applied to minimize the total error for the model coefficient and the outlier eliminated data repeated until the value reaches a user definition error improvement limit. Filtered data is used for the purpose of authentication, the outlier bias reduction, and data quality operation.SELECTED DRAWING: Figure 1

The present disclosure relates to a cloud service control method and device, a storage medium and an electronic device, to specifically handle memory errors and improve the processing efficiency of memory errors. The method comprises: when a memory error is detected, determining a memory address at which the memory error occurs, and determining an error type of the memory error according to the memory address, the error type being used to identify whether the memory error will cause a virtual machine to crash; determining a total error number of the detected memory error; and according to the error type of the memory error and the total error number, controlling a cloud scheduler to schedule a corresponding scheduling service from a scheduling service library, the scheduling service library storing a first scheduling service and a second scheduling service, the first scheduling service being used to control restarting a virtual machine on an original host to which the virtual machine belongs, and the second scheduling service being used to control migrating the virtual machine from the original host to which the virtual machine belongs to another host.

The invention discloses a laboratory data fault-tolerant control method considering total errors, and belongs to the field of medical data statistics, and the method comprises the steps: building a fault-tolerant mechanism which is used for defining a fault-tolerant object and a fault type; inputting the fault-tolerant mechanism and related original data of the fault-tolerant mechanism into a first prediction model to establish a data set, and dividing the data set into a training data set and a test data set in proportion; calculating a total error of the data samples in each training data set, wherein the total error comprises a proportional error, a constant error and an index error; inputting the proportional error, the constant error and the index error into a data set of the first prediction model to obtain a second prediction model; the fault-tolerant object or fault type is redefined when the predicted value error of both the first prediction model and the second prediction model is above a threshold. According to the method, the reliability of the algorithm can be ensured, a fault-tolerant mechanism can be adaptively adjusted, and the fault-tolerant capability is further expanded.

A low-complexity AGENT error precision control method solves the problems of difficult control, complex compensation, and poor real-time performance of AGENT full-process error + core Jiang's initial error decomposition formula Δ=Δ₁+Δ₂+Δ₃, combined with phase compensation and closed-loop verification, total error ≤1.1%, calculation period ≤1s + for industrial / creative / public service AGENT error control.

The application discloses a kind of test method and system of optical module, it is related to optical communication device test technical field, for solving the problem that existing PAM4 optical module error code test is only judged qualified according to total error rate, it is difficult to identify the defect that error code is concentrated under specific symbol jump type, is covered by average;The method obtains the jump description table and error code bit position set corresponding to the preset PAM4 test sequence, converts error code bit position into error code symbol position, and determines the symbol jump type to be measured corresponding to error code according to jump description table;The effective occurrence number and error code symbol quantity of each symbol jump type to be measured are counted, the jump error code density is calculated, and then the jump sensitive error code test result is output;Through the scheme, the implicit error code risk under specific PAM4 jump type can be identified in the case where total error level is qualified, and the accuracy of optical module communication performance test and the defect screening capability are improved.

The invention provides a thermoelectric load optimal distribution method based on a dynamic fitness function, belongs to the technical field of combined heat and power generation, and can at least partially solve the problems that an existing fitness function is fixed in weight, single in error processing and difficult to give consideration to economical efficiency and boundary conditions. The method comprises the steps of obtaining a thermoelectric load error; according to an error mapping rule, small errors are mapped to be zero, medium errors are amplified moderately, and large errors are amplified exponentially; according to an error accumulation rule, only accumulating mapping errors exceeding an absolute error threshold to calculate a total error; according to a penalty term weight calculation rule, dynamically adjusting a penalty weight by utilizing a smoothing function in combination with the total error and the tolerance range; and constructing a dynamic fitness function based on the adjusted weight, and guiding an optimization algorithm to iteratively solve. The method has the beneficial effects that the global optimization capability and the convergence precision are improved while the boundary conditions are met, the ineffective calculation interference is reduced, and the accuracy, the stability and the economical efficiency of thermoelectric load distribution are improved.

This invention discloses a tolerance allocation and structural optimization design method for an optical tracking and measurement system based on Abbe error. It establishes a system error propagation model for the optical tracking and measurement system based on Lie group and Lie algebra theory; analytically expresses the macroscopic and microscopic Abbe errors caused by measurement distance and scanning depth; and establishes a system tolerance optimization allocation model by combining sensitivity analysis of each error source with the goal of minimizing the total system cost. For the positioning target, a design criterion for the positioning target radius based on Abbe error constraints is derived. Compared with existing technologies, this invention represents the total error as a vector synthesis of three error vectors in space. The analytical model can not only predict the average level of error but also accurately describe the statistical boundary of error fluctuations, achieving a balance between tolerance allocation and accuracy-cost. It automatically adjusts the accuracy of each component based on the error method effect, achieving minimum cost tolerance allocation, and derives the formula for the minimum target radius under the target accuracy.

A method, apparatus and product comprising: obtaining alternative quantum circuits implementing a quantum program, each having a respective design error; identifying, for each quantum circuit, respective quantum resources that are used by the quantum circuit and are associated with respective hardware errors of a quantum computer; determining a total hardware error bound for the quantum circuit; determining a total error bound for the quantum circuit; applying a search algorithm over a plurality of total error bounds of the plurality of alternative quantum circuits, to identify an optimal total error bound that is associated with an optimal quantum circuit; and providing the optimal quantum circuit for execution on the quantum computer.