CN110793693A - Force sensor based sliding prediction method and device, electronic equipment and storage medium - Google Patents
Force sensor based sliding prediction method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN110793693A CN110793693A CN201911067291.5A CN201911067291A CN110793693A CN 110793693 A CN110793693 A CN 110793693A CN 201911067291 A CN201911067291 A CN 201911067291A CN 110793693 A CN110793693 A CN 110793693A
- Authority
- CN
- China
- Prior art keywords
- force sensor
- output result
- data signal
- lstm
- result
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01L—MEASURING FORCE, STRESS, TORQUE, WORK, MECHANICAL POWER, MECHANICAL EFFICIENCY, OR FLUID PRESSURE
- G01L5/00—Apparatus for, or methods of, measuring force, work, mechanical power, or torque, specially adapted for specific purposes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Force Measurement Appropriate To Specific Purposes (AREA)
Abstract
The invention discloses a force sensor-based sliding prediction method and device, electronic equipment and a storage medium, wherein the force sensor-based sliding prediction method comprises the following steps: collecting a force sensor data signal; taking the force sensor data signal as the input of a long-short term memory network (LSTM), and obtaining an output result after testing the force sensor data signal through an LSTM network training model; and predicting the sliding of the force sensor according to the output result. By the embodiment of the invention, the sliding prediction is realized by combining a data-driven learning mode of training model test judgment on the force sensor data signal by the LSTM network, the threshold value modification and adjustment for specific articles are not needed, the adaptability is better, and the accuracy of the result is improved; and the determination is made based on the force change of the force sensor before the relative motion occurs.
Description
Technical Field
The present invention relates to the field of sensors, and in particular, to a method and an apparatus for predicting a slip based on a force sensor, an electronic device, and a storage medium.
Background
At present, the sliding prediction is a key technology for realizing stable grabbing of the robot, and the current grabbing state is predicted by feedback of a sensor, so that the robot can be controlled to realize more stable and reliable grabbing actions.
Currently, the prediction of sliding can be classified into force, tactile perception, and visual perception according to the way of perception. Force and touch perception are realized by force and touch feedback, and because force change and relative displacement can occur between an object and a sensor when relative sliding occurs, the judgment can be carried out on data time domain and frequency domain information, and the judgment can be carried out through the characteristics of force, displacement, vibration and the like. And the relative position and deformation change of the object and the sensing unit of the end effector are judged by visual perception to judge the occurrence of sliding.
For force sensors, the current technical solutions mainly include the following categories:
1) setting a threshold value for judgment according to the time domain signal, and when the force change exceeds a certain range, considering that the sliding occurs, wherein the situation needs to be adjusted according to different objects and the grabbing force;
2) and extracting a specific range signal for judgment according to the frequency domain signal, wherein corresponding adjustment is still needed when the situation aims at grabbing objects made of different materials.
The above technical scheme needs to modify and adjust the threshold value for a specific article, and has poor adaptability and low accuracy of the prediction result.
Disclosure of Invention
In view of this, the force sensor-based slip prediction method, device, electronic device and storage medium provided in the embodiments of the present invention implement slip prediction by combining a data-driven learning manner in which an LSTM network performs training model test determination on a force sensor data signal, and can have better adaptability and improve accuracy of a result; and the determination is made based on the force change of the force sensor before the relative motion occurs.
The technical scheme adopted by the invention for solving the technical problems is as follows:
according to an aspect of the embodiments of the present invention, there is provided a force sensor-based slip prediction method, including:
collecting a force sensor data signal;
taking the force sensor data signal as the input of a long-short term memory network (LSTM), and obtaining an output result after testing the force sensor data signal through an LSTM network training model;
and predicting the sliding of the force sensor according to the output result.
In one possible design, the force sensor data signal includes a time domain data signal, and the obtaining of the output result after the force sensor data signal is used as the input of the long-term and short-term memory network LSTM and tested by the LSTM network training model includes:
and taking the force sensor time domain data signal as the input of a long-short term memory network (LSTM), and testing the force sensor time domain data signal through an LSTM network training model to obtain a first output result.
In one possible design, the force sensor data signal includes a frequency domain data signal, and the obtaining of the output result after the force sensor data signal is used as the input of the long-term and short-term memory network LSTM and tested by the LSTM network training model includes:
processing the force sensor frequency domain data signals through Fourier FFT to obtain frequency domain change composition of signals at each moment;
and taking the frequency domain variation composition as the input of a long-short term memory network (LSTM), and testing the frequency domain variation composition through an LSTM network training model to obtain a second output result.
In one possible design, the force sensor data signal includes a time domain data signal and a frequency domain data signal, and the obtaining of the output result after the force sensor data signal is used as the input of the long-term and short-term memory network LSTM and tested by the LSTM network training model includes:
the force sensor time domain data signal is used as the input of a long-short term memory network (LSTM), and a first output result is obtained after the force sensor time domain data signal is tested by an LSTM network training model;
processing the force sensor frequency domain data signals through Fourier FFT to obtain frequency domain change composition of signals at each moment; the frequency domain variation composition is used as the input of a long-short term memory network (LSTM), and a second output result is obtained after the frequency domain variation composition is tested by an LSTM network training model;
and combining the first output result and the second output result by adopting a full connection layer and obtaining a third output result by softmax.
In one possible design, the fully-connected layer includes a training model formula, and combining the first output result and the second output result with the fully-connected layer and softmax to obtain a third output result includes:
substituting the first output result and the second output result into the training model formula Z ═ (w)1x1+b1)+(w2x2+b2) Calculating to obtain model output value, wherein x1、x2Representing a first output result and a second output result obtained by passing the time domain data signal and the frequency domain data signal through the LSMT as inputs; w is a1、w2Representing that the two output results respectively account for the weight in the model; b1、b2Represents a correction coefficient; and obtaining the third output result by the model output value through softmax.
In one possible design, the predicting the sliding of the force sensor based on the output includes: and comparing the output result with a preset value according to the output result, and predicting whether the force sensor slides.
In a possible design, the comparing the output result with a preset value result according to the output result to predict whether the force sensor slides includes:
if the output result is larger than or equal to the preset value result, predicting that the force sensor slides;
and if the output result is smaller than the preset value result, predicting that the force sensor does not slide.
According to another aspect of the embodiments of the present invention, there is provided a force sensor-based slip prediction apparatus applied to a force sensor-based slip prediction method, the apparatus including: collection module, test module, prediction module, wherein:
the acquisition module is used for acquiring a force sensor data signal;
the testing module is used for taking the force sensor data signal as the input of a long-short term memory network (LSTM), and obtaining an output result after testing the force sensor data signal through an LSTM network training model;
and the prediction module is used for predicting the sliding of the force sensor according to the output result.
According to another aspect of the present invention, there is provided an electronic device including: the force sensor-based slip prediction method comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the computer program realizes the steps of the force sensor-based slip prediction method provided by the embodiment of the invention when being executed by the processor.
According to another aspect of the present invention, there is provided a computer readable storage medium having a program of the force sensor-based slip prediction method stored thereon, where the program of the force sensor-based slip prediction method is executed by a processor to implement the steps of the force sensor-based slip prediction method provided by the embodiments of the present invention.
Compared with the related art, the sliding prediction method and device based on the force sensor, the electronic equipment and the storage medium provided by the embodiment of the invention comprise the following steps: collecting a force sensor data signal; taking the force sensor data signal as the input of a long-short term memory network (LSTM), and obtaining an output result after testing the force sensor data signal through an LSTM network training model; and predicting the sliding of the force sensor according to the output result. By the embodiment of the invention, the sliding prediction is realized by combining a data-driven learning mode of training model test judgment on the force sensor data signal by the LSTM network, the threshold value modification and adjustment for specific articles are not needed, the adaptability is better, and the accuracy of the result is improved; and the determination is made based on the force change of the force sensor before the relative motion occurs.
Drawings
Fig. 1 is a schematic flow chart of a sliding prediction method based on a force sensor according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a force sensor based slip prediction method according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a force sensor based slip prediction method according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a force sensor based slip prediction method according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a sliding prediction apparatus based on a force sensor according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
In order to make the technical problems, technical solutions and advantageous effects to be solved by the present invention clearer and clearer, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
In the following description, suffixes such as "module", "component", or "unit" used to denote elements are used only for facilitating the explanation of the present invention, and have no specific meaning in itself. Thus, "module", "component" or "unit" may be used mixedly.
It should be noted that the terms first, second and the like in the description and in the claims, and in the drawings are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order.
In one embodiment, as shown in fig. 1, the present invention provides a force sensor based slip prediction method, the method comprising:
step S1, collecting force sensor data signals;
step S2, the force sensor data signal is used as the input of a Long Short-term memory network LSTM (Long Short-term memory), and the output result is obtained after the force sensor data signal is tested by an LSTM network training model;
step S3 predicts the slippage of the force sensor based on the output result.
In the present embodiment, the force sensor data signal is processed to predict the slip state. The sliding prediction is realized by combining a data-driven learning mode of training model test judgment on the force sensor data signal by the LSTM network, so that better adaptability can be realized, and the accuracy of the result is improved; and the determination is made based on the force change of the force sensor before the relative motion occurs.
In one embodiment, as shown in fig. 2, the force sensor data signal includes a time domain data signal, and in step S2, the obtaining an output result after testing the force sensor data signal through an LSTM network training model by using the force sensor data signal as an input of a long-short term memory network LSTM includes:
the time domain data signal of the force sensor is used as the input of a long-short term memory network (LSTM), and a first output result is obtained after the time domain data signal of the force sensor is tested by an LSTM network training model.
In step S3, the predicting of the slippage of the force sensor based on the output result includes: comparing the first output result with a preset value result according to the first output result, predicting whether the force sensor slides, and predicting that the force sensor slides if the first output result is greater than or equal to the preset value result; if the first output result is smaller than the preset value result, it is predicted that the force sensor does not slide, in this embodiment, the first output result may be a value between (0,1), the preset value result may be a set threshold, for example, 0.5, the output result is compared with the threshold, if the output result is greater than 0.5, the output value is 1, if the output result is less than 0.5, the output value is 0, it is predicted whether the sensor slides, if the comparison output value is 0, it is predicted that no sliding occurs, and if the output value is 1, it is predicted that sliding occurs.
In the present embodiment, the force sensor data signal is subjected to time-domain processing to predict the slip state. The sliding prediction is realized by combining a data-driven learning mode of training model test judgment on the time domain data signal of the force sensor by the LSTM network, and threshold modification and adjustment are not needed for specific articles, so that the training result has better prediction accuracy and generalization capability, a more accurate prediction result can be obtained, and better adaptability is realized; and the determination is made based on the force change of the force sensor before the relative motion occurs.
In one embodiment, as shown in fig. 3, the force sensor data signal includes a frequency domain data signal, and in step S2, the obtaining an output result after testing the force sensor data signal through an LSTM network training model by using the force sensor data signal as an input of a long-short term memory network LSTM includes:
processing the force sensor frequency domain data signals through Fourier transform (FFT) to obtain frequency domain variation components of signals at all times;
and taking the frequency domain variation composition as the input of a long-short term memory network (LSTM), and testing the frequency domain variation composition through an LSTM network training model to obtain a second output result.
In step S3, the predicting of the slippage of the force sensor based on the output result includes: according to the second output result, comparing the second output result with a preset value result, predicting whether the force sensor slides, and if the second output result is larger than or equal to the preset value result, predicting that the force sensor slides; if the second output result is smaller than the preset value result, it is predicted that the force sensor does not slide, in this embodiment, the second output result may be a value between (0,1), the preset value result may be a set threshold, for example, 0.5, the output result is compared with the threshold, if the output result is greater than 0.5, the output value is 1, if the output result is less than 0.5, the output value is 0, it is predicted whether the sensor slides, if the comparison output value is 0, it is predicted that no sliding occurs, and if the output value is 1, it is predicted that sliding occurs.
In the present embodiment, the force sensor data signal is subjected to processing in the frequency domain to predict the slip state. The FFT can well extract frequency domain information in the frequency domain data signal of the force sensor, and the frequency domain change can be obtained by combining the LSTM, so that the output result is more stable; the sliding prediction is realized by combining a data-driven learning mode of training model test judgment on the force sensor frequency domain data signal by an LSTM network, and threshold modification and adjustment are not needed for specific articles, so that the output result is more stable, and the accuracy of the result is improved; and the determination is made based on the force change of the force sensor before the relative motion occurs.
In one embodiment, as shown in fig. 4, the force sensor data signal includes a time domain data signal and a frequency domain data signal, and the step S2 of taking the force sensor data signal as an input of the long-short term memory network LSTM and obtaining an output result after testing the force sensor data signal through the LSTM network training model includes:
the force sensor time domain data signal is used as the input of a long-short term memory network (LSTM), and a first output result is obtained after the force sensor time domain data signal is tested by an LSTM network training model;
processing the force sensor frequency domain data signals through Fourier FFT to obtain frequency domain change composition of signals at each moment; the frequency domain variation composition is used as the input of a long-short term memory network (LSTM), and a second output result is obtained after the frequency domain variation composition is tested by an LSTM network training model;
and combining the first output result and the second output result by adopting a full connection layer and obtaining a third output result by softmax.
In step S2, the full connection layer includes a training model formula, and the first output result and the second output result are substituted into (w) the training model formula Z1x1+b1)+(w2x2+b2) Calculating to obtain model output value, wherein x1、x2Representing a first output result and a second output result obtained by passing the time domain data signal and the frequency domain data signal through the LSMT as inputs; w is a1、w2Representing that the two output results respectively account for the weight in the model; b1、b2Represents a correction coefficient; training the parameters w in the model1、w2And b1、b2The values are determined after parameter correction according to a plurality of groups of sample data sets trained by a training model, namely, a plurality of groups of sample data sets (including input data and corresponding output marks) are trained by the training model, the result naturally output by the training model is compared with the output marks, and correction adjustment is carried out to determine model parameters w1、w2And b1、b2。
And obtaining the third output result by the model output value through softmax.
In step S3, the predicting of the slippage of the force sensor based on the output result includes: comparing the third output result with a preset value result according to the third output result to predict whether the force sensor slides, and predicting that the force sensor slides if the third output result is greater than or equal to the preset value result; if the third output result is smaller than the preset value result, it is predicted that the force sensor does not slide, in this embodiment, the third output result may be a value between (0,1), the preset value result may be a set threshold, for example, 0.5, the output result is compared with the threshold, if the output result is greater than 0.5, the output value is 1, if the output result is less than 0.5, the output value is 0, it is predicted whether the sensor slides, if the comparison output value is 0, it is predicted that no sliding occurs, and if the output value is 1, it is predicted that sliding occurs.
In the present embodiment, the force sensor data signal is subjected to processing in the time domain and the frequency domain to predict the slip state. The FFT can well extract frequency domain information in the frequency domain data signal of the force sensor, and the frequency domain change can be obtained by combining the LSTM, so that the output result is more stable; the time domain signal directly used as the training result of the LSTM input has better prediction accuracy and generalization capability, and a more accurate prediction result can be obtained. By combining the advantages of the FFT and the LSTM network, more stable sliding prediction with good generalization capability is obtained, threshold modification and adjustment are not needed for specific articles, better adaptability can be realized, and the accuracy of the result is improved; and the determination is made based on the force change of the force sensor before the relative motion occurs.
In one embodiment, as shown in fig. 5, the present invention is a force sensor-based slip prediction apparatus applied to a force sensor-based slip prediction method according to any one of the following embodiments, the apparatus comprising: collection module, test module, prediction module, wherein:
the acquisition module is used for acquiring a force sensor data signal;
the testing module is used for taking the force sensor data signal as the input of a long-short term memory network (LSTM), and obtaining an output result after testing the force sensor data signal through an LSTM network training model;
and the prediction module is used for predicting the sliding of the force sensor according to the output result.
In the embodiment, the sliding prediction is realized by combining a data-driven learning mode of training model test judgment on the force sensor data signal by the LSTM network, so that better adaptability can be realized, and the accuracy of the result is improved; and the determination is made based on the force change of the force sensor before the relative motion occurs.
In one embodiment, the force sensor data signal comprises a time domain data signal, and the testing module is specifically configured to:
the time domain data signal of the force sensor is used as the input of a long-short term memory network (LSTM), and a first output result is obtained after the time domain data signal of the force sensor is tested by an LSTM network training model.
The prediction module is specifically configured to: comparing the first output result with a preset value result according to the first output result, predicting whether the force sensor slides, and predicting that the force sensor slides if the first output result is greater than or equal to the preset value result; if the first output result is smaller than the preset value result, it is predicted that the force sensor does not slide, in this embodiment, the first output result may be a value between (0,1), the preset value result may be a set threshold, for example, 0.5, the output result is compared with the threshold, if the output result is greater than 0.5, the output value is 1, if the output result is less than 0.5, the output value is 0, it is predicted whether the sensor slides, if the comparison output value is 0, it is predicted that no sliding occurs, and if the output value is 1, it is predicted that sliding occurs.
In the embodiment, the sliding prediction is realized by combining a data-driven learning mode of training model test judgment on the time domain data signal of the force sensor by the LSTM network, so that the training result has better prediction accuracy and generalization capability, a more accurate prediction result can be obtained, and better adaptability is realized; and the determination is made based on the force change of the force sensor before the relative motion occurs.
In one embodiment, the force sensor data signal comprises a frequency domain data signal, and the testing module is specifically configured to:
processing the force sensor frequency domain data signals through Fourier FFT to obtain frequency domain change composition of signals at each moment;
and taking the frequency domain variation composition as the input of a long-short term memory network (LSTM), and testing the frequency domain variation composition through an LSTM network training model to obtain a second output result.
The prediction module is specifically configured to: according to the second output result, comparing the second output result with a preset value result, predicting whether the force sensor slides, and if the second output result is larger than or equal to the preset value result, predicting that the force sensor slides; if the second output result is smaller than the preset value result, it is predicted that the force sensor does not slide, in this embodiment, the second output result may be a value between (0,1), the preset value result may be a set threshold, for example, 0.5, the output result is compared with the threshold, if the output result is greater than 0.5, the output value is 1, if the output result is less than 0.5, the output value is 0, it is predicted whether the sensor slides, if the comparison output value is 0, it is predicted that no sliding occurs, and if the output value is 1, it is predicted that sliding occurs.
In the embodiment, the FFT can well extract frequency domain information in the frequency domain data signal of the force sensor, and the frequency domain change can be obtained by combining the LSTM, so that the output result is more stable; the sliding prediction is realized by combining a data-driven learning mode of training model test judgment on the force sensor frequency domain data signal by an LSTM network, so that the output result is more stable, and the accuracy of the result is improved; and the determination is made based on the force change of the force sensor before the relative motion occurs.
In one embodiment, the force sensor data signal includes a time domain data signal and a frequency domain data signal, and the testing module is specifically configured to:
the force sensor time domain data signal is used as the input of a long-short term memory network (LSTM), and a first output result is obtained after the force sensor time domain data signal is trained by the LSTM network;
processing the force sensor frequency domain data signals through Fourier FFT to obtain frequency domain change composition of signals at each moment; the frequency domain variation composition is used as the input of a long-short term memory network (LSTM), and a second output result is obtained after the frequency domain variation composition is tested by an LSTM network training model;
and combining the first output result and the second output result by adopting a full connection layer and obtaining a third output result by softmax. In this step, the process is carried out,
the full-link layer includes a training model formula, and the first output result and the second output result are substituted into the training model formula Z ═ w1x1+b1)+(w2x2+b2) Calculating to obtain model output value, wherein x1、x2Representing a first output result and a second output result obtained by passing the time domain data signal and the frequency domain data signal through the LSMT as inputs; w is a1、w2Representing that the two output results respectively account for the weight in the model; b1、b2Represents a correction coefficient; parameter in training modelNumber w1、w2And b1、b2The values are determined after parameter correction according to a plurality of groups of sample data sets trained by a training model, namely, a plurality of groups of sample data sets (including input data and corresponding output marks) are trained by the training model, the result naturally output by the training model is compared with the output marks, and correction adjustment is carried out to determine model parameters w1、w2And b1、b2。
And obtaining the third output result by the model output value through softmax.
In step S3, the predicting of the slippage of the force sensor based on the output result includes: comparing the third output result with a preset value result according to the third output result to predict whether the force sensor slides, and predicting that the force sensor slides if the third output result is greater than or equal to the preset value result; if the third output result is smaller than the preset value result, it is predicted that the force sensor does not slide, in this embodiment, the third output result may be a value between (0,1), the preset value result may be a set threshold, for example, 0.5, the output result is compared with the threshold, if the output result is greater than 0.5, the output value is 1, if the output result is less than 0.5, the output value is 0, it is predicted whether the sensor slides, if the comparison output value is 0, it is predicted that no sliding occurs, and if the output value is 1, it is predicted that sliding occurs.
The prediction module is specifically configured to: and predicting the slip of the force sensor based on the third output result, predicting no slip if the output result is 0, and predicting slip if the output result is 1.
In the embodiment, the FFT can well extract frequency domain information in the frequency domain data signal of the force sensor, and the frequency domain change can be obtained by combining the LSTM, so that the output result is more stable; the time domain signal directly used as the training result of the LSTM input has better prediction accuracy and generalization capability, and a more accurate prediction result can be obtained. By combining the advantages of the FFT and the LSTM network, the sliding prediction which is more stable and has good generalization capability is obtained, better adaptability can be realized, and the accuracy of the result is improved; and the determination is made based on the force change of the force sensor before the relative motion occurs.
It should be noted that the device embodiment and the method embodiment belong to the same concept, and specific implementation processes thereof are described in the method embodiment in detail, and technical features in the method embodiment are correspondingly applicable in the device embodiment, which is not described herein again.
In addition, an embodiment of the present invention further provides an electronic device, as shown in fig. 6, including: a memory, a processor, and one or more computer programs stored in the memory and executable on the processor, the one or more computer programs when executed by the processor implementing the following steps of a force sensor based slip prediction method provided by an embodiment of the invention:
step S1, collecting force sensor data signals;
step S2, the force sensor data signal is used as the input of the long-short term memory network LSTM, and the output result is obtained after the force sensor data signal is tested by the LSTM network training model;
step S3 predicts the slippage of the force sensor based on the output result.
Preferably, the step S2 of using the force sensor data signal as an input of a long-short term memory network LSTM, and after testing the force sensor data signal by an LSTM network training model, obtaining an output result includes:
the time domain data signal of the force sensor is used as the input of a long-short term memory network (LSTM), and a first output result is obtained after the time domain data signal of the force sensor is tested by an LSTM network training model.
In step S3, the predicting of the slippage of the force sensor based on the output result includes: comparing the first output result with a preset value result according to the first output result, predicting whether the force sensor slides, and predicting that the force sensor slides if the first output result is greater than or equal to the preset value result; if the first output result is smaller than the preset value result, it is predicted that the force sensor does not slide, in this embodiment, the first output result may be a value between (0,1), the preset value result may be a set threshold, for example, 0.5, the output result is compared with the threshold, if the output result is greater than 0.5, the output value is 1, if the output result is less than 0.5, the output value is 0, it is predicted whether the sensor slides, if the comparison output value is 0, it is predicted that no sliding occurs, and if the output value is 1, it is predicted that sliding occurs.
Preferably, the step S2 of using the force sensor data signal as an input of the long-short term memory network LSTM, and testing the force sensor data signal through the LSTM network training model to obtain an output result includes:
processing the force sensor frequency domain data signals through Fourier FFT to obtain frequency domain change composition of signals at each moment;
and taking the frequency domain variation composition as the input of a long-short term memory network (LSTM), and testing the frequency domain variation composition through an LSTM network training model to obtain a second output result.
In step S3, the predicting of the slippage of the force sensor based on the output result includes: according to the second output result, comparing the second output result with a preset value result, predicting whether the force sensor slides, and if the second output result is larger than or equal to the preset value result, predicting that the force sensor slides; if the second output result is smaller than the preset value result, it is predicted that the force sensor does not slide, in this embodiment, the second output result may be a value between (0,1), the preset value result may be a set threshold, for example, 0.5, the output result is compared with the threshold, if the output result is greater than 0.5, the output value is 1, if the output result is less than 0.5, the output value is 0, it is predicted whether the sensor slides, if the comparison output value is 0, it is predicted that no sliding occurs, and if the output value is 1, it is predicted that sliding occurs.
Preferably, the force sensor data signal includes a time domain data signal and a frequency domain data signal, and in step S2, the obtaining an output result after the force sensor data signal is used as an input of a long-short term memory network LSTM and is tested by an LSTM network training model includes:
the force sensor time domain data signal is used as the input of a long-short term memory network (LSTM), and a first output result is obtained after the force sensor time domain data signal is tested by an LSTM network training model;
processing the force sensor frequency domain data signals through Fourier FFT to obtain frequency domain change composition of signals at each moment; the frequency domain variation composition is used as the input of a long-short term memory network (LSTM), and a second output result is obtained after the frequency domain variation composition is tested by an LSTM network training model;
and combining the first output result and the second output result by adopting a full connection layer and obtaining a third output result by softmax.
In step S2, the full connection layer includes a training model formula, and the first output result and the second output result are substituted into (w) the training model formula Z1x1+b1)+(w2x2+b2) Calculating to obtain model output value, wherein x1、x2Representing a first output result and a second output result obtained by passing the time domain data signal and the frequency domain data signal through the LSMT as inputs; w is a1、w2Representing that the two output results respectively account for the weight in the model; b1、b2Represents a correction coefficient; training the parameters w in the model1、w2And b1、b2The values are determined after parameter correction according to a plurality of groups of sample data sets trained by a training model, namely, a plurality of groups of sample data sets (including input data and corresponding output marks) are trained by the training model, the result naturally output by the training model is compared with the output marks, and correction adjustment is carried out to determine model parameters w1、w2And b1、b2。
And obtaining the third output result by the model output value through softmax.
In step S3, the predicting of the slippage of the force sensor based on the output result includes: comparing the third output result with a preset value result according to the third output result to predict whether the force sensor slides, and predicting that the force sensor slides if the third output result is greater than or equal to the preset value result; if the third output result is smaller than the preset value result, it is predicted that the force sensor does not slide, in this embodiment, the third output result may be a value between (0,1), the preset value result may be a set threshold, for example, 0.5, the output result is compared with the threshold, if the output result is greater than 0.5, the output value is 1, if the output result is less than 0.5, the output value is 0, it is predicted whether the sensor slides, if the comparison output value is 0, it is predicted that no sliding occurs, and if the output value is 1, it is predicted that sliding occurs.
The method disclosed in the above embodiments of the present invention may be applied to the processor 901, or implemented by the processor 901. The processor 901 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the above method may be implemented by an integrated logic circuit of hardware or an instruction in the form of software in the processor 901. The processor 901 may be a general purpose processor, a DSP, or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, or the like. The processor 901 may implement or perform the methods, steps and logic blocks disclosed in the embodiments of the present invention. A general purpose processor may be a microprocessor or any conventional processor or the like. The steps of the method disclosed by the embodiment of the invention can be directly implemented by a hardware decoding processor, or can be implemented by combining hardware and software modules in the decoding processor. The software modules may be located in a storage medium located in the memory 902, and the processor 901 reads the information in the memory 902 and performs the steps of the foregoing method in combination with the hardware thereof.
It is to be understood that the memory 902 of embodiments of the present invention may be either volatile memory or nonvolatile memory, and may include both volatile and nonvolatile memory. Among them, the nonvolatile Memory may be a Read-Only Memory (ROM), a Programmable Read-Only Memory (PROM), an Erasable Programmable Read-Only Memory (EPROM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), a magnetic Random Access Memory (FRAM), a magnetic Random Access Memory (Flash Memory) or other Memory technologies, a Compact disc Read-Only Memory (CD-ROM), a Digital Versatile Disc (DVD), or other optical disc storage, magnetic cartridge, magnetic tape, magnetic Disk storage, or other magnetic storage devices; volatile Memory can be Random Access Memory (RAM), and by way of exemplary and not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Double Data Rate Synchronous Random Access Memory (ESDRAM), Synchronous Link Dynamic Random Access Memory (SLDRAM), Direct Memory bus Random Access Memory (DRRAM). The described memory for embodiments of the present invention is intended to comprise, without being limited to, these and any other suitable types of memory.
It should be noted that the embodiments of the electronic device and the embodiments of the method belong to the same concept, and specific implementation processes thereof are described in the embodiments of the method, and technical features in the embodiments of the method are correspondingly applicable in the embodiments of the electronic device, which is not described herein again.
In addition, an embodiment of the present invention further provides a computer-readable storage medium, on which a force sensor-based slip prediction method program is stored, and when the force sensor-based slip prediction method is executed by a processor, the following steps of a force sensor-based slip prediction method provided by an embodiment of the present invention are implemented.
Step S1, collecting force sensor data signals;
step S2, the force sensor data signal is used as the input of the long-short term memory network LSTM, and the output result is obtained after the force sensor data signal is tested by the LSTM network training model;
step S3 predicts the slippage of the force sensor based on the output result.
Preferably, the step S2 of using the force sensor data signal as an input of a long-short term memory network LSTM, and after testing the force sensor data signal by an LSTM network training model, obtaining an output result includes:
the time domain data signal of the force sensor is used as the input of a long-short term memory network (LSTM), and a first output result is obtained after the time domain data signal of the force sensor is tested by an LSTM network training model.
In step S3, the predicting of the slippage of the force sensor based on the output result includes: comparing the first output result with a preset value result according to the first output result, predicting whether the force sensor slides, and predicting that the force sensor slides if the first output result is greater than or equal to the preset value result; if the first output result is smaller than the preset value result, it is predicted that the force sensor does not slide, in this embodiment, the first output result may be a value between (0,1), the preset value result may be a set threshold, for example, 0.5, the output result is compared with the threshold, if the output result is greater than 0.5, the output value is 1, if the output result is less than 0.5, the output value is 0, it is predicted whether the sensor slides, if the comparison output value is 0, it is predicted that no sliding occurs, and if the output value is 1, it is predicted that sliding occurs.
Preferably, the step S2 of using the force sensor data signal as an input of the long-short term memory network LSTM, and testing the force sensor data signal through the LSTM network training model to obtain an output result includes:
processing the force sensor frequency domain data signals through Fourier FFT to obtain frequency domain change composition of signals at each moment;
and taking the frequency domain variation composition as the input of a long-short term memory network (LSTM), and testing the frequency domain variation composition through an LSTM network training model to obtain a second output result.
In step S3, the predicting of the slippage of the force sensor based on the output result includes: according to the second output result, comparing the second output result with a preset value result, predicting whether the force sensor slides, and if the second output result is larger than or equal to the preset value result, predicting that the force sensor slides; if the second output result is smaller than the preset value result, it is predicted that the force sensor does not slip, in this embodiment, the second output result may be a value between (0,1), the preset value result may be a set threshold, for example, 0.5, the output result is compared with the threshold, if the output result is greater than 0.5, the output value is 1, if the output result is less than 0.5, the output value is 0, it is predicted whether the sensor slips, if the comparison output value is 0, it is predicted that no slip occurs, and if the output value is 1, it is predicted that slip occurs.
Preferably, the force sensor data signal includes a time domain data signal and a frequency domain data signal, and in step S2, the obtaining an output result after the force sensor data signal is used as an input of a long-short term memory network LSTM and is tested by an LSTM network training model includes:
the force sensor time domain data signal is used as the input of a long-short term memory network (LSTM), and a first output result is obtained after the force sensor time domain data signal is tested by an LSTM network training model;
processing the force sensor frequency domain data signals through Fourier FFT to obtain frequency domain change composition of signals at each moment; the frequency domain variation composition is used as the input of a long-short term memory network (LSTM), and a second output result is obtained after the frequency domain variation composition is tested by an LSTM network training model;
and combining the first output result and the second output result by adopting a full connection layer and obtaining a third output result by softmax.
In step S2, the full connection layer includes a training model formula, and the first output result and the second output result are substituted into (w) the training model formula Z1x1+b1)+(w2x2+b2) Calculating to obtain model output value, wherein x1、x2Representing a first output result and a second output result obtained by passing the time domain data signal and the frequency domain data signal through the LSMT as inputs; w is a1、w2Representing that the two output results respectively account for the weight in the model; b1、b2Represents a correction coefficient; training the parameters w in the model1、w2And b1、b2The values are determined after parameter correction according to a plurality of groups of sample data sets trained by a training model, namely, a plurality of groups of sample data sets (including input data and corresponding output marks) are trained by the training model, the result naturally output by the training model is compared with the output marks, and correction adjustment is carried out to determine model parameters w1、w2And b1、b2。
And obtaining the third output result by the model output value through softmax.
In step S3, the predicting of the slippage of the force sensor based on the output result includes: comparing the third output result with a preset value result according to the third output result to predict whether the force sensor slides, and predicting that the force sensor slides if the third output result is greater than or equal to the preset value result; if the third output result is smaller than the preset value result, it is predicted that the force sensor does not slide, in this embodiment, the third output result may be a value between (0,1), the preset value result may be a set threshold, for example, 0.5, the output result is compared with the threshold, if the output result is greater than 0.5, the output value is 1, if the output result is less than 0.5, the output value is 0, it is predicted whether the sensor slides, if the comparison output value is 0, it is predicted that no sliding occurs, and if the output value is 1, it is predicted that sliding occurs.
It should be noted that, the embodiment of the force sensor-based sliding prediction method program on the computer-readable storage medium and the embodiment of the method belong to the same concept, and the specific implementation process thereof is described in detail in the embodiment of the method, and the technical features in the embodiment of the method are correspondingly applicable to the embodiment of the computer-readable storage medium, and are not described herein again.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The above-mentioned serial numbers of the embodiments of the present invention are merely for description and do not represent the merits of the embodiments.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present invention may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present invention.
While the present invention has been described with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, which are illustrative and not restrictive, and it will be apparent to those skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (10)
1. A force sensor based slip prediction method, the method comprising:
collecting a force sensor data signal;
taking the force sensor data signal as the input of a long-short term memory network (LSTM), and obtaining an output result after testing the force sensor data signal through an LSTM network training model;
and predicting the sliding of the force sensor according to the output result.
2. The method of claim 1, wherein the force sensor data signal comprises a time domain data signal, and wherein the using the force sensor data signal as an input to a long-short term memory network (LSTM) results in an output after testing the force sensor data signal with an LSTM network training model comprises:
and taking the force sensor time domain data signal as the input of a long-short term memory network (LSTM), and testing the force sensor time domain data signal through an LSTM network training model to obtain a first output result.
3. The method of claim 1, wherein the force sensor data signal comprises a frequency domain data signal, and wherein the using the force sensor data signal as an input to a long-short term memory network (LSTM) results in an output after testing the force sensor data signal via an LSTM network training model comprises:
processing the force sensor frequency domain data signals through Fourier FFT to obtain frequency domain change composition of signals at each moment;
and taking the frequency domain variation composition as the input of a long-short term memory network (LSTM), and testing the frequency domain variation composition through an LSTM network training model to obtain a second output result.
4. The method of claim 1, wherein the force sensor data signals comprise time domain data signals and frequency domain data signals, and wherein the obtaining an output result using the force sensor data signals as an input to a long-short term memory network (LSTM) after testing the force sensor data signals using an LSTM network training model comprises:
the force sensor time domain data signal is used as the input of a long-short term memory network (LSTM), and a first output result is obtained after the force sensor time domain data signal is tested by an LSTM network training model;
processing the force sensor frequency domain data signals through Fourier FFT to obtain frequency domain change composition of signals at each moment; the frequency domain variation composition is used as the input of a long-short term memory network (LSTM), and a second output result is obtained after the frequency domain variation composition is tested by an LSTM network training model;
and combining the first output result and the second output result by adopting a full connection layer and obtaining a third output result by softmax.
5. The method of claim 4, wherein the fully-connected layer comprises a training model formula, and wherein combining the first output result and the second output result with the fully-connected layer and softmax to obtain a third output result comprises:
substituting the first output result and the second output result into the training model formula Z ═ (w)1x1+b1)+(w2x2+b2) Calculating to obtain model output value, wherein x1、x2Representing a first output result and a second output result obtained by passing the time domain data signal and the frequency domain data signal through the LSMT as inputs; w is a1、w2Representing that the two output results respectively account for the weight in the model; b1、b2Represents a correction coefficient; obtaining the third model output value through softmaxAnd outputting the result.
6. The method of any of claims 1 to 5, wherein predicting the force sensor slippage based on the output comprises: and comparing the output result with a preset value result according to the output result, and predicting whether the force sensor slides.
7. The method of claim 6, wherein comparing the output result with a preset value result according to the output result to predict whether the force sensor is slipping comprises:
if the output result is larger than or equal to the preset value result, predicting that the force sensor slides;
and if the output result is smaller than the preset value result, predicting that the force sensor does not slide.
8. A force sensor-based slip prediction apparatus applied to the force sensor-based slip prediction method according to any one of claims 1 to 7, the apparatus comprising: collection module, test module, prediction module, wherein:
the acquisition module is used for acquiring a force sensor data signal;
the testing module is used for taking the force sensor data signal as the input of a long-short term memory network (LSTM), and obtaining an output result after testing the force sensor data signal through an LSTM network training model;
and the prediction module is used for predicting the sliding of the force sensor according to the output result.
9. An electronic device, comprising: memory, a processor and a computer program stored on the memory and executable on the processor, the computer program when executed by the processor implementing the steps of a force sensor based slip prediction method according to any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a program of a force-sensor-based slip prediction method, which when executed by a processor implements the steps of a force-sensor-based slip prediction method as claimed in any one of claims 1 to 7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911067291.5A CN110793693A (en) | 2019-11-04 | 2019-11-04 | Force sensor based sliding prediction method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911067291.5A CN110793693A (en) | 2019-11-04 | 2019-11-04 | Force sensor based sliding prediction method and device, electronic equipment and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110793693A true CN110793693A (en) | 2020-02-14 |
Family
ID=69442587
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911067291.5A Pending CN110793693A (en) | 2019-11-04 | 2019-11-04 | Force sensor based sliding prediction method and device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110793693A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112449362A (en) * | 2020-11-18 | 2021-03-05 | 中国科学院上海微***与信息技术研究所 | Intelligent sensor self-adaptive configuration method, device and system based on letter model |
CN113253063A (en) * | 2020-10-29 | 2021-08-13 | 诸暨市迪朗物联科技有限公司 | Fault arc detection system and method based on long-time memory network deep learning |
CN113510700A (en) * | 2021-05-19 | 2021-10-19 | 哈尔滨理工大学 | Touch perception method for robot grabbing task |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1806940A (en) * | 2006-01-23 | 2006-07-26 | 湖南大学 | Defective goods automatic sorting method and equipment for high-speed automated production line |
CN106204620A (en) * | 2016-07-21 | 2016-12-07 | 清华大学 | A kind of tactile three-dimensional power detection method based on micro-vision |
CN109343046A (en) * | 2018-09-19 | 2019-02-15 | 成都理工大学 | Radar gait recognition method based on multifrequency multiple domain deep learning |
EP3486769A1 (en) * | 2017-11-20 | 2019-05-22 | Guangdong OPPO Mobile Telecommunications Corp., Ltd. | Method for preloading application, terminal device, and medium |
CN109917200A (en) * | 2017-12-12 | 2019-06-21 | 株洲中车时代电气股份有限公司 | Train traction current transformer method for diagnosing faults, system, equipment and storage medium |
CN110109109A (en) * | 2019-04-26 | 2019-08-09 | 西安电子科技大学 | HRRP target identification method based on multiresolution attention convolutional network |
-
2019
- 2019-11-04 CN CN201911067291.5A patent/CN110793693A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1806940A (en) * | 2006-01-23 | 2006-07-26 | 湖南大学 | Defective goods automatic sorting method and equipment for high-speed automated production line |
CN106204620A (en) * | 2016-07-21 | 2016-12-07 | 清华大学 | A kind of tactile three-dimensional power detection method based on micro-vision |
EP3486769A1 (en) * | 2017-11-20 | 2019-05-22 | Guangdong OPPO Mobile Telecommunications Corp., Ltd. | Method for preloading application, terminal device, and medium |
CN109917200A (en) * | 2017-12-12 | 2019-06-21 | 株洲中车时代电气股份有限公司 | Train traction current transformer method for diagnosing faults, system, equipment and storage medium |
CN109343046A (en) * | 2018-09-19 | 2019-02-15 | 成都理工大学 | Radar gait recognition method based on multifrequency multiple domain deep learning |
CN110109109A (en) * | 2019-04-26 | 2019-08-09 | 西安电子科技大学 | HRRP target identification method based on multiresolution attention convolutional network |
Non-Patent Citations (5)
Title |
---|
KARL VAN WYK等: "Calibration and Analysis of Tactile Sensors as Slip Detectors", 《 2018 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA)》 * |
MATTHIAS KERZEL等: "Neuro-Robotic Haptic Object Classification by Active Exploration on a Novel Dataset", 《2019 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN)》 * |
吕应宇: "基于深度学习的非特定目标跟踪算法研究", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 * |
常晓龙: "基于深度学习的人体运动目标实时检测定位与动作识别", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 * |
毛燕: "基于卷积神经网络的网上银行验证码识别", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113253063A (en) * | 2020-10-29 | 2021-08-13 | 诸暨市迪朗物联科技有限公司 | Fault arc detection system and method based on long-time memory network deep learning |
CN112449362A (en) * | 2020-11-18 | 2021-03-05 | 中国科学院上海微***与信息技术研究所 | Intelligent sensor self-adaptive configuration method, device and system based on letter model |
CN112449362B (en) * | 2020-11-18 | 2021-08-20 | 中国科学院上海微***与信息技术研究所 | Intelligent sensor self-adaptive configuration method, device and system based on letter model |
CN113510700A (en) * | 2021-05-19 | 2021-10-19 | 哈尔滨理工大学 | Touch perception method for robot grabbing task |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110793693A (en) | Force sensor based sliding prediction method and device, electronic equipment and storage medium | |
US9341499B2 (en) | Sensor system | |
CN110109441B (en) | Laser inertial measurement unit fault prediction method and system | |
CN111324764B (en) | Image detection method and device, electronic equipment and storage medium | |
US20150177076A1 (en) | Method for determining the ambient temperature of a mobile device | |
CN112100884B (en) | Vibration amount determining method of touch device of linear motor and related equipment | |
KR20020008209A (en) | Method Of Electronically Monitoring And Controlling A Process For The Adjustment Of Mobile Parts | |
CN111881764A (en) | Target detection method and device, electronic equipment and storage medium | |
CN112986693B (en) | Capacitor baseline updating method, chip and capacitor detection device | |
Dong et al. | On-line identification algorithm and convergence analysis for sandwich systems with backlash | |
CN115219883A (en) | Automatic detection system and detection method for electric automation equipment | |
Zhou et al. | Interaction-based dynamic measurement of haptic characteristics of control elements | |
CN110869859A (en) | Device and method for determining the state of a spindle of a machine tool | |
KR20090112892A (en) | Fault diagnosis apparatus for gas sensor and method at the same | |
Tan et al. | Fault detection of mechanical systems with inherent backlash | |
EP3992930A3 (en) | Method and apparatus for controlling data collection, storage medium, and computer program product | |
Christensen et al. | Quantitative study on the modal parameters estimated using the PLSCF and the MITD methods and an automated modal analysis algorithm | |
CN110334244B (en) | Data processing method and device and electronic equipment | |
AU2008274592B2 (en) | A system and a method for processing a signal from an electric device | |
Dymkou et al. | Optimal control of non-stationary differential linear repetitive processes | |
CN111752149A (en) | Design method of digital repetitive controller for designated time attraction of servo motor | |
Voros | Recursive identification of Wiener systems with two-segment polynomial nonlinearities | |
Mrad | Non-linear systems representation using ARMAX models with time-dependent coefficients | |
Voros | Recursive identification of Hammerstein systems with polynomial nonlinearities | |
CN113514148B (en) | Equipment vibration early warning method and system based on low-delay data smoothing |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200214 |
|
RJ01 | Rejection of invention patent application after publication |