WO2023116043A1 - 模糊度确认模型的训练方法、模糊度确认方法及装置 - Google Patents

模糊度确认模型的训练方法、模糊度确认方法及装置 Download PDF

Info

Publication number
WO2023116043A1
WO2023116043A1 PCT/CN2022/116375 CN2022116375W WO2023116043A1 WO 2023116043 A1 WO2023116043 A1 WO 2023116043A1 CN 2022116375 W CN2022116375 W CN 2022116375W WO 2023116043 A1 WO2023116043 A1 WO 2023116043A1
Authority
WO
WIPO (PCT)
Prior art keywords
ambiguity
confirmation
difference
model
feature information
Prior art date
Application number
PCT/CN2022/116375
Other languages
English (en)
French (fr)
Inventor
王兴
何思彤
戴鹏
Original Assignee
千寻位置网络有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 千寻位置网络有限公司 filed Critical 千寻位置网络有限公司
Publication of WO2023116043A1 publication Critical patent/WO2023116043A1/zh

Links

Images

Classifications

    • 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
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • 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
    • 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
    • G06N3/0475—Generative networks
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems

Definitions

  • the application belongs to the technical field of positioning, and in particular relates to a training method of a ambiguity confirmation model, a ambiguity confirmation method and a device.
  • Ambiguity confirmation results affect the fixation rate and accuracy of positioning results. For example, confirming the actual and effective ambiguity as non-true and effective ambiguity will affect the positioning fixation rate, and confirming the actual non-real and effective ambiguity as real and effective The ambiguity will affect the positioning accuracy. Therefore, it is of great significance to accurately confirm the ambiguity.
  • observation thresholds are mainly set through an empirical model, and the ambiguity is confirmed by comparing the real-time observations with the set observation thresholds.
  • this method relies heavily on the empirical model and the set observation threshold, and the accuracy of ambiguity confirmation is low.
  • the purpose of the embodiments of the present application is to provide a training method for a ambiguity confirmation model, a method and a device for ambiguity confirmation, which can solve the problem of low accuracy of ambiguity confirmation.
  • the embodiment of the present application provides a method for training an ambiguity confirmation model, including:
  • Extract feature information of the sample positioning data wherein the feature information includes: the difference between the wide-lane ambiguity and the ambiguity difference, the position difference between the first position and the second position, and the position difference between the first position and the third position.
  • the ambiguity difference is the difference between the ambiguities of two frequency points
  • the first position is the position calculated according to the wide-lane ambiguity
  • the second position is based on any one of the two frequency points
  • the third position is the position obtained by positioning and calculation according to the ambiguity of the ultra-wide lane;
  • the neural network model is trained to obtain the ambiguity confirmation model.
  • the embodiment of the present application provides a method for confirming ambiguity, including:
  • the characteristic information is input into the ambiguity confirmation model trained by using the ambiguity confirmation model training method provided in the first aspect of the embodiment of the present application, and the confirmation result of the ambiguity corresponding to the target positioning data is obtained.
  • the embodiment of the present application provides a training device for an ambiguity confirmation model, including:
  • a first acquisition module configured to acquire sample positioning data
  • the first extraction module is used to extract feature information of the sample positioning data, wherein the feature information includes: the difference between the wide-lane ambiguity and the ambiguity difference, the position difference between the first position and the second position, the first position and the second position
  • the position difference of the three positions is at least one of them.
  • the ambiguity difference is the difference of the ambiguities of the two frequency points.
  • the first position is the position calculated according to the wide-lane ambiguity.
  • the position obtained by positioning and calculating the ambiguity of any frequency point in the point, and the third position is the position obtained by positioning and calculating according to the ambiguity of the ultra-wide lane;
  • the training module is used to train the neural network model according to the feature information to obtain the ambiguity confirmation model.
  • the embodiment of the present application provides an apparatus for confirming ambiguity, including:
  • the second acquisition module is used to acquire target positioning data
  • the second extraction module is used to extract feature information of the target positioning data
  • the confirmation module is used to input the feature information into the ambiguity confirmation model trained by the ambiguity confirmation model training method provided by the first aspect of the embodiment of the present application, and obtain the confirmation result of the ambiguity corresponding to the target positioning data.
  • the embodiment of the present application provides an electronic device, including a processor and a memory, and the memory stores programs or instructions that can run on the processor.
  • the programs or instructions are executed by the processor, the first aspect or the second The steps of the method described in the aspect.
  • an embodiment of the present application provides a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the method described in the first aspect or the second aspect are implemented .
  • the embodiment of the present application provides a chip, including a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the method described in the first aspect or the second aspect step.
  • an embodiment of the present application provides a computer program product, the program product is stored in a storage medium, and the program product is executed by at least one processor to implement the method described in the first aspect or the second aspect.
  • the neural network model is trained to obtain Ambiguity Confirmation Model.
  • the ambiguity confirmation model to confirm the ambiguity corresponding to the acquired target positioning data, a confirmation result of the ambiguity corresponding to the target positioning data can be obtained.
  • the confirmation of ambiguity does not depend on empirical models and the set observation thresholds, which can improve the accuracy of ambiguity confirmation.
  • the ambiguity confirmation model is based on at least one of the above-mentioned difference between the wide-lane ambiguity and the ambiguity difference, the position difference between the first position and the second position, and the position difference between the first position and the third position.
  • the difference between the wide-lane ambiguity and the ambiguity difference, the position difference between the first position and the second position, and the position difference between the first position and the third position are important features related to ambiguity confirmation. Therefore, the ambiguity confirmation model trained by the above three features can accurately confirm the ambiguity and improve the accuracy of ambiguity confirmation.
  • Fig. 1 is a schematic flow chart of the training method of the ambiguity confirmation model provided by the embodiment of the present application;
  • Fig. 2 is a schematic flow chart of the ambiguity confirmation method provided by the embodiment of the present application.
  • Fig. 3 is a schematic diagram of the process of ambiguity confirmation provided by the embodiment of the present application.
  • Fig. 4 is a schematic structural diagram of a training device for an ambiguity confirmation model provided in an embodiment of the present application
  • Fig. 5 is a schematic structural diagram of an ambiguity confirmation device provided in an embodiment of the present application.
  • FIG. 6 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.
  • FIG. 7 is a schematic diagram of a hardware structure of an electronic device implementing an embodiment of the present application.
  • FIG. 1 is a schematic flowchart of a training method for an ambiguity confirmation model provided by an embodiment of the present application.
  • the training method of the ambiguity confirmation model may include:
  • the ambiguity difference is the difference between the ambiguity of two frequency points
  • the first position is the position calculated according to the wide-lane ambiguity
  • the second position is based on any one of the two frequency points
  • the third position is the position calculated according to the ambiguity of the ultra-wide lane
  • the neural network model is trained to obtain Ambiguity Confirmation Model.
  • the ambiguity confirmation model to confirm the ambiguity corresponding to the acquired target positioning data, a confirmation result of the ambiguity corresponding to the target positioning data can be obtained.
  • the confirmation of ambiguity does not depend on empirical models and the set observation thresholds, which can improve the accuracy of ambiguity confirmation.
  • the ambiguity confirmation model is based on at least one of the above-mentioned difference between the wide-lane ambiguity and the ambiguity difference, the position difference between the first position and the second position, and the position difference between the first position and the third position.
  • the difference between the wide-lane ambiguity and the ambiguity difference, the position difference between the first position and the second position, and the position difference between the first position and the third position are important features related to ambiguity confirmation. Therefore, the ambiguity confirmation model trained by the above three features can accurately confirm the ambiguity and improve the accuracy of ambiguity confirmation.
  • the two frequency points in the embodiment of the present application may be any two frequency points satisfying the condition of forming a widelane.
  • the wide-lane ambiguity in the embodiments of the present application may be to combine the carrier phase observations corresponding to the above two frequency points to obtain the carrier phase combination observations. According to the carrier phase combination Observation fixed ambiguity.
  • the ultra-wide-lane ambiguity in the embodiments of the present application may be to combine the carrier-phase observations of at least two frequency points that meet the conditions for forming the ultra-wide-lane to obtain the combined carrier-phase observations value, according to which the carrier phase combines observations with a fixed ambiguity.
  • the ambiguity of the sample positioning data may be marked in advance.
  • the feature information and its corresponding ambiguity actual mark can be input into the neural network model, and the prediction result of the ambiguity of the sample positioning data corresponding to the feature information is obtained, and the neural network is trained according to the ambiguity actual mark and the prediction result.
  • the network model is used to obtain the ambiguity confirmation model, wherein, when the network model is trained, the actual mark of ambiguity is used as the expected output of the characteristic information of the corresponding sample positioning data.
  • the threshold value of the number of cyclic iteration training times can be set, and when the number of training times of the neural network model reaches the set threshold value of the number of cyclic iteration training times, the training is stopped to obtain Ambiguity Confirmation Model.
  • the loss value of the neural network model when training the neural network model, may also be calculated, and when the loss value of the neural network model reaches a minimum, the training is stopped to obtain a ambiguity confirmation model.
  • the accuracy of the neural network model prediction when training the neural network model, can also be calculated according to the actual ambiguity mark and the above prediction results.
  • the accuracy of the neural network model prediction is greater than the accuracy
  • the training is stopped and the ambiguity confirmation model is obtained.
  • part of the sample positioning data may be selected from the sample positioning data as test samples.
  • the characteristic information of the test sample is input into the neural network model at this time, and the prediction result of the ambiguity of the test sample is obtained.
  • the prediction accuracy of the neural network model at this time calculates the prediction accuracy of the neural network model at this time, when the prediction accuracy of the neural network model is not greater than the accuracy threshold, use the characteristic information of the sample positioning data to continue training the neural network model, and then test the The characteristic information of the sample is input into the neural network model at this time, and the accuracy of the neural network model prediction at this time is calculated until the accuracy of the neural network model prediction is greater than the accuracy threshold, and the neural network model at this time is used as the ambiguity confirmation model.
  • the embodiment of the present application does not limit the specific process of training the neural network model and obtaining the ambiguity confirmation model according to the characteristic information.
  • the training of the neural network model can refer to the process of training the neural network model in the related art.
  • the embodiment of the present application It will not be described in detail here.
  • the feature information in the embodiments of the present application may also include at least three of the following items:
  • the number of fixed ambiguities, the post-test unit weight error, the precision factor value, the number of reference satellite changes, the good or bad sign of the single-point solution positioning result, the number of effective pseudo-ranges, pseudo-range residuals, the number of satellites, according to the alternative The ratio of the smallest unit weight variance obtained from the ambiguity to the next smallest unit weight variance, the mean value of the fixed carrier double difference residual, and the average carrier-to-noise ratio.
  • the neural network model in the embodiment of the present application is a binary classification model.
  • the predicted result is one of the above two situations, either one or the other.
  • the binary classification model in the embodiment of the present application can be: a binary classification model based on logistic regression, a binary classification model based on the K nearest neighbor algorithm, a binary classification model based on a binary decision tree , Two-category model based on support vector machine, two-category model based on Naive Bayes.
  • the binary classification model in the embodiment of the present application is preferably: a binary classification model based on logistic regression.
  • the classification function of the logistic regression-based binary classification model in the embodiments of the present application is a Sigmoid function
  • the Sigmoid function is shown in the following formula (1):
  • x is the vector corresponding to the feature information
  • ⁇ is the weight vector to be fitted.
  • ⁇ may be solved by maximum likelihood estimation.
  • solving ⁇ by maximum likelihood estimation is to find ⁇ when the logarithmic likelihood function of the binary classification model based on logistic regression takes the maximum value.
  • the obtained ⁇ is the optimal value of the binary classification model based on logistic regression parameter.
  • the gradient ascending method may be used to solve ⁇ .
  • Fig. 2 is a schematic flowchart of a method for confirming ambiguity provided by an embodiment of the present application.
  • the ambiguity confirmation method may include:
  • S203 Input the feature information into the ambiguity confirmation model trained by using the ambiguity confirmation model training method provided by the embodiment of the present application, and obtain the ambiguity confirmation result corresponding to the target positioning data.
  • the ambiguity corresponding to the acquired target positioning data can be confirmed through the ambiguity confirmation model, and the confirmation result of the ambiguity corresponding to the target positioning data can be obtained.
  • the confirmation of ambiguity does not depend on empirical models and the set observation thresholds, which can improve the accuracy of ambiguity confirmation.
  • the ambiguity confirmation model is trained according to at least one of the difference between wide-lane ambiguity and ambiguity difference, the position difference between the first position and the second position, and the position difference between the first position and the third position Obtained, wherein, the ambiguity difference is the difference between the ambiguities of the two frequency points, the first position is the position calculated according to the wide-lane ambiguity, and the second position is based on any one of the two frequency points
  • the position calculated based on the ambiguity of the frequency point, the third position is the position calculated according to the ultra-wide-lane ambiguity, and the difference between the wide-lane ambiguity and the ambiguity difference, the first position and the second position
  • the position difference of , the position difference between the first position and the third position are important features among the features related to ambiguity confirmation. Therefore, the ambiguity confirmation model trained by the above three features can accurately confirm the ambiguity , which can improve the accuracy of ambiguity confirmation.
  • the extracted feature information is the same feature information as that used when training the neural network model.
  • the verification result of the ambiguity includes: the ambiguity is true and valid (ie, the ambiguity is correct, and the confirmation is passed), and the ambiguity is not real and valid (ie, the ambiguity is wrong, and the confirmation is not passed).
  • the ambiguity confirmed in the embodiment of the present application may be an ambiguity of whole cycles (ambiguity of whole cycles).
  • the integer ambiguity also known as the integer unknown, is the integer unknown corresponding to the first observed value of the phase difference between the carrier phase and the reference phase during the carrier phase measurement of the GPS technology.
  • Fig. 3 is a schematic diagram of the process of ambiguity confirmation provided by the embodiment of the present application.
  • the process of ambiguity confirmation includes two stages, which are offline training stage and online confirmation stage.
  • the feature information of the sample location data and the corresponding ambiguity actual mark are input into the binary classification model based on machine learning (that is, the neural network model) to train the binary classification model and obtain the ambiguity confirmation
  • machine learning that is, the neural network model
  • the feature information of the actual positioning data is extracted, and the feature information is input into the ambiguity confirmation model trained in the offline training stage, and the confirmation result of the ambiguity corresponding to the actual positioning data can be obtained.
  • the training method for the ambiguity confirmation model provided in the embodiment of the present application may be executed by a training device for the ambiguity confirmation model.
  • the training device for the ambiguity confirmation model provided by the embodiment of the present application is described by taking the training method of the ambiguity confirmation model executed by the training device for the ambiguity confirmation model as an example.
  • Fig. 4 is a schematic structural diagram of a training device for an ambiguity confirmation model provided by an embodiment of the present application.
  • the training device 400 of the ambiguity confirmation model may include:
  • a first acquisition module 401 configured to acquire sample location data
  • the first extraction module 402 is used to extract feature information of the sample positioning data, wherein the feature information includes: the difference between the wide-lane ambiguity and the ambiguity difference, the position difference between the first position and the second position, the first position and the second position
  • the position difference of the third position is at least one of them.
  • the ambiguity difference is the difference between the ambiguities of two frequency points.
  • the first position is the position calculated according to the wide-lane ambiguity.
  • the position obtained by positioning and calculating the ambiguity of any frequency point in the frequency points, and the third position is the position obtained by positioning and calculating according to the ambiguity of the ultra-wide lane;
  • the training module 403 is used to train the neural network model according to the feature information to obtain the ambiguity confirmation model.
  • the neural network model is trained to obtain Ambiguity Confirmation Model.
  • the ambiguity confirmation model to confirm the ambiguity corresponding to the acquired target positioning data, a confirmation result of the ambiguity corresponding to the target positioning data can be obtained.
  • the confirmation of ambiguity does not depend on empirical models and the set observation thresholds, which can improve the accuracy of ambiguity confirmation.
  • the ambiguity confirmation model is based on at least one of the above-mentioned difference between the wide-lane ambiguity and the ambiguity difference, the position difference between the first position and the second position, and the position difference between the first position and the third position.
  • the difference between the wide-lane ambiguity and the ambiguity difference, the position difference between the first position and the second position, and the position difference between the first position and the third position are important features related to ambiguity confirmation. Therefore, the ambiguity confirmation model trained by the above three features can accurately confirm the ambiguity and improve the accuracy of ambiguity confirmation.
  • the feature information further includes at least three of the following items:
  • the number of fixed ambiguities, the post-test unit weight error, the precision factor value, the number of reference satellite changes, the good or bad sign of the single-point solution positioning result, the number of effective pseudo-ranges, pseudo-range residuals, the number of satellites, according to the alternative The ratio of the smallest unit weight variance obtained from the ambiguity to the next smallest unit weight variance, the mean value of the fixed carrier double difference residual, and the average carrier-to-noise ratio.
  • the neural network model includes:
  • the binary classification model includes:
  • a binary classification model based on logistic regression A binary classification model based on logistic regression.
  • the ambiguity confirmation method provided in the embodiment of the present application may be executed by a ambiguity confirmation device.
  • the ambiguity confirmation device provided in the embodiment of the present application is described by taking the ambiguity confirmation method performed by the ambiguity confirmation device as an example.
  • Fig. 5 is a schematic structural diagram of an apparatus for confirming ambiguity provided by an embodiment of the present application.
  • the ambiguity confirmation device 500 may include:
  • the second acquiring module 501 is configured to acquire target positioning data
  • the second extraction module 502 is used to extract feature information of the target positioning data
  • the confirmation module 503 is configured to input the feature information into the ambiguity confirmation model trained by using the ambiguity confirmation model training method provided in the embodiment of the present application, and obtain the confirmation result of the ambiguity corresponding to the target positioning data.
  • the ambiguity corresponding to the acquired target positioning data can be confirmed through the ambiguity confirmation model, and the confirmation result of the ambiguity corresponding to the target positioning data can be obtained.
  • the confirmation of ambiguity does not depend on empirical models and the set observation thresholds, which can improve the accuracy of ambiguity confirmation.
  • the ambiguity confirmation model is trained according to at least one of the difference between wide-lane ambiguity and ambiguity difference, the position difference between the first position and the second position, and the position difference between the first position and the third position Obtained, wherein, the ambiguity difference is the difference between the ambiguities of the two frequency points, the first position is the position calculated according to the wide-lane ambiguity, and the second position is based on any one of the two frequency points
  • the position calculated based on the ambiguity of the frequency point, the third position is the position calculated according to the ultra-wide-lane ambiguity, and the difference between the wide-lane ambiguity and the ambiguity difference, the first position and the second position
  • the position difference of , the position difference between the first position and the third position are important features among the features related to ambiguity confirmation. Therefore, the ambiguity confirmation model trained by the above three features can accurately confirm the ambiguity , which can improve the accuracy of ambiguity confirmation.
  • the training device of the ambiguity confirmation model or the ambiguity confirmation device in the embodiment of the present application may be an electronic device, or a component in the electronic device, such as an integrated circuit or a chip.
  • the electronic device may be a terminal, or other devices other than the terminal.
  • the electronic device can be a mobile phone, a tablet computer, a notebook computer, a handheld computer, a vehicle electronic device, a mobile Internet device (Mobile Internet Device, MID), an augmented reality (augmented reality, AR)/virtual reality (virtual reality, VR) ) equipment, robots, wearable devices, ultra-mobile personal computer (ultra-mobile personal computer, UMPC), netbook or personal digital assistant (personal digital assistant, PDA), etc., can also serve as server, network attached storage (Network Attached Storage, NAS), personal computer (personal computer, PC), television (television, TV), teller machine, or self-service machine, etc., which are not specifically limited in this embodiment of the present application.
  • Network Attached Storage Network Attached Storage
  • the training device for the ambiguity confirmation model or the ambiguity confirmation device in the embodiment of the present application may be a device with an operating system.
  • the operating system may be an Android operating system, an iOS operating system, or other possible operating systems, which are not specifically limited in this embodiment of the present application.
  • the training device for the ambiguity confirmation model provided by the embodiment of the present application can realize each process in the embodiment of the training method for the ambiguity confirmation model provided by the embodiment of the present application. To avoid repetition, details are not repeated here.
  • the apparatus for confirming ambiguity provided in the embodiment of the present application can realize each process in the embodiment of the method for confirming the ambiguity provided in the embodiment of the present application. To avoid repetition, details are not repeated here.
  • the embodiment of the present application also provides an electronic device 600, including a processor 601 and a memory 602.
  • the memory 602 stores programs or instructions that can run on the processor 601.
  • the program or instructions When executed by the processor 601, each step of the above-mentioned training method of the ambiguity confirmation model or the embodiment of the ambiguity confirmation method can be realized, and the same technical effect can be achieved. To avoid repetition, details are not repeated here.
  • the processor 601 may include a central processing unit (CPU), or a specific integrated circuit (Application Specific Integrated Circuit, ASIC), or may be configured to implement one or more of the embodiments of the present application. integrated circuit.
  • CPU central processing unit
  • ASIC Application Specific Integrated Circuit
  • the memory 602 may include a read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a disk storage medium device, an optical storage medium device, a flash memory Device, electrical, optical, or other physical/tangible memory storage device.
  • ROM read-only memory
  • RAM random access memory
  • disk storage medium device an optical storage medium device
  • flash memory Device electrical, optical, or other physical/tangible memory storage device.
  • memory 602 includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software comprising computer-executable instructions, and when the software is executed (e.g., by a or a plurality of processors), it is operable to execute the operations described with reference to the training method of the ambiguity confirmation model or the ambiguity confirmation method provided according to the embodiments of the present application.
  • software e.g., memory devices
  • software e.g., by a or a plurality of processors
  • FIG. 7 is a schematic diagram of a hardware structure of an electronic device implementing an embodiment of the present application.
  • the electronic device 700 includes, but is not limited to: a radio frequency unit 701, a network module 702, an audio output unit 703, an input unit 704, a sensor 705, a display unit 706, a user input unit 707, an interface unit 708, a memory 709, and a processor 710, etc. part.
  • the electronic device 700 can also include a power supply (such as a battery) for supplying power to various components, and the power supply can be logically connected to the processor 710 through the power management system, so that the management of charging, discharging, and function can be realized through the power management system. Consumption management and other functions.
  • a power supply such as a battery
  • the structure of the electronic device shown in FIG. 7 does not constitute a limitation to the electronic device.
  • the electronic device may include more or fewer components than shown in the figure, or combine certain components, or arrange different components, and details will not be repeated here. .
  • the processor 710 is used to: obtain sample positioning data; extract feature information of the sample positioning data, wherein the feature information includes: the difference between the wide-lane ambiguity and the ambiguity difference, the position difference between the first position and the second position , the position difference between the first position and the third position is at least one of them, and the ambiguity difference is the difference between the ambiguities of two frequency points, the first position is the position calculated according to the wide-lane ambiguity, and the second The position is the position calculated according to the ambiguity of any one of the two frequency points, and the third position is the position calculated according to the ambiguity of the ultra-wide lane; according to the feature information, the neural network model is trained to obtain Ambiguity Confirmation Model.
  • the feature information includes: the difference between the wide-lane ambiguity and the ambiguity difference, the position difference between the first position and the second position , the position difference between the first position and the third position is at least one of them, and the ambiguity difference is the difference between the ambiguities of two frequency points
  • the feature information further includes at least three of the following items:
  • the number of fixed ambiguities, the post-test unit weight error, the precision factor value, the number of reference satellite changes, the good or bad sign of the single-point solution positioning result, the number of effective pseudo-ranges, pseudo-range residuals, the number of satellites, according to the alternative The ratio of the smallest unit weight variance obtained from the ambiguity to the next smallest unit weight variance, the mean value of the fixed carrier double difference residual, and the average carrier-to-noise ratio.
  • the neural network model includes:
  • the binary classification model includes:
  • a binary classification model based on logistic regression A binary classification model based on logistic regression.
  • the processor 710 may also be used to: obtain target positioning data; extract feature information of the target positioning data; input the feature information into the above-mentioned ambiguity confirmation model to obtain the ambiguity corresponding to the target positioning data confirmation result.
  • the ambiguity corresponding to the acquired target positioning data can be confirmed through the ambiguity confirmation model, and the confirmation result of the ambiguity corresponding to the target positioning data can be obtained.
  • the confirmation of ambiguity does not depend on empirical models and the set observation thresholds, which can improve the accuracy of ambiguity confirmation.
  • the ambiguity confirmation model is trained according to at least one of the difference between wide-lane ambiguity and ambiguity difference, the position difference between the first position and the second position, and the position difference between the first position and the third position Obtained, wherein, the ambiguity difference is the difference between the ambiguities of the two frequency points, the first position is the position calculated according to the wide-lane ambiguity, and the second position is based on any one of the two frequency points
  • the position calculated based on the ambiguity of the frequency point, the third position is the position calculated according to the ultra-wide-lane ambiguity, and the difference between the wide-lane ambiguity and the ambiguity difference, the first position and the second position
  • the position difference of , the position difference between the first position and the third position are important features among the features related to ambiguity confirmation. Therefore, the ambiguity confirmation model trained by the above three features can accurately confirm the ambiguity , which can improve the accuracy of ambiguity confirmation.
  • the input unit 704 may include a graphics processor (Graphics Processing Unit, GPU) 7041 and a microphone 7042, and the graphics processor 7041 is used for the image capture device (such as the image data of the still picture or video obtained by the camera) for processing.
  • the display unit 706 may include a display panel 7061, and the display panel 7061 may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like.
  • the user input unit 707 includes at least one of a touch panel 7071 and other input devices 7072 .
  • the touch panel 7071 is also called a touch screen.
  • the touch panel 7071 may include two parts, a touch detection device and a touch controller.
  • Other input devices 7072 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, switch buttons, etc.), trackballs, mice, and joysticks, which will not be described in detail here.
  • the memory 709 can be used to store software programs as well as various data.
  • the memory 709 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instructions required by at least one function (such as a sound playing function, image playback function, etc.), etc.
  • memory 709 may include volatile memory or nonvolatile memory, or, memory 709 may include both volatile and nonvolatile memory.
  • the non-volatile memory can be read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), erasable programmable read-only memory (Erasable PROM, EPROM), electronically programmable Erase Programmable Read-Only Memory (Electrically EPROM, EEPROM) or Flash.
  • ROM Read-Only Memory
  • PROM programmable read-only memory
  • Erasable PROM Erasable PROM
  • EPROM erasable programmable read-only memory
  • Electrical EPROM Electrical EPROM
  • EEPROM electronically programmable Erase Programmable Read-Only Memory
  • Volatile memory can be random access memory (Random Access Memory, RAM), static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic RAM, DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDRSDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), synchronous connection dynamic random access memory (Synch link DRAM , SLDRAM) and Direct Memory Bus Random Access Memory (Direct Rambus RAM, DRRAM).
  • RAM Random Access Memory
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • DRAM synchronous dynamic random access memory
  • SDRAM double data rate synchronous dynamic random access memory
  • Double Data Rate SDRAM Double Data Rate SDRAM
  • DDRSDRAM double data rate synchronous dynamic random access memory
  • Enhanced SDRAM, ESDRAM enhanced synchronous dynamic random access memory
  • Synch link DRAM , SLDRAM
  • Direct Memory Bus Random Access Memory Direct Rambus
  • the processor 710 may include one or more processing units; optionally, the processor 710 integrates an application processor and a modem processor, wherein the application processor mainly processes operations related to the operating system, user interface, and application programs, etc., Modem processors mainly process wireless communication signals, such as baseband processors. It can be understood that the foregoing modem processor may not be integrated into the processor 710 .
  • the embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the above-mentioned training method of the ambiguity confirmation model or the embodiment of the ambiguity confirmation method are implemented.
  • a readable storage medium on which a program or instruction is stored, and when the program or instruction is executed by a processor, the above-mentioned training method of the ambiguity confirmation model or the embodiment of the ambiguity confirmation method are implemented.
  • Each process can achieve the same technical effect, so in order to avoid repetition, it will not be repeated here.
  • the processor is the processor in the electronic device in the foregoing embodiments.
  • the readable storage medium includes a computer-readable storage medium, and examples of the computer-readable storage medium include non-transitory computer-readable storage media, such as ROM, RAM, magnetic disks, or optical disks.
  • the embodiment of the present application further provides a chip, including a processor and a communication interface, the communication interface is coupled to the processor, the processor is used to run programs or instructions, and implement the training method of the above-mentioned ambiguity confirmation model or the embodiment of the ambiguity confirmation method
  • a chip including a processor and a communication interface
  • the communication interface is coupled to the processor
  • the processor is used to run programs or instructions, and implement the training method of the above-mentioned ambiguity confirmation model or the embodiment of the ambiguity confirmation method
  • chips mentioned in the embodiments of the present application may also be called system-on-chip, system-on-chip, system-on-a-chip, or system-on-a-chip.
  • An embodiment of the present application provides a computer program product, the program product is stored in a storage medium, and the program product is executed by at least one processor to implement each of the above embodiments of the ambiguity confirmation model training method or ambiguity confirmation method. process, and can achieve the same technical effect, in order to avoid repetition, it will not be repeated here.
  • the term “comprising”, “comprising” or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article or apparatus comprising a set of elements includes not only those elements, It also includes other elements not expressly listed, or elements inherent in the process, method, article, or device. Without further limitations, an element defined by the phrase “comprising a " does not preclude the presence of additional identical elements in the process, method, article, or apparatus comprising that element.
  • the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved. Functions are performed, for example, the described methods may be performed in an order different from that described, and various steps may also be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Remote Sensing (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Automation & Control Theory (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Medical Informatics (AREA)
  • Position Fixing By Use Of Radio Waves (AREA)

Abstract

本申请公开了一种模糊度确认模型的训练方法、模糊度确认方法及装置，属于定位技术领域。首先，获取样本定位数据；提取样本定位数据的特征信息；根据样本定位数据的特征信息，训练神经网络模型，得到模糊度确认模型。然后，将目标定位数据的特征信息输入到该模糊度确认模型，即可得到目标定位数据对应的模糊度的确认结果。通过本申请公开的方案，能够提高模糊度确认的准确度。

Description

模糊度确认模型的训练方法、模糊度确认方法及装置
相关申请的交叉引用
本申请主张在2021年12月21日在中国提交的中国专利申请号202111572245.8的优先权，其全部内容通过引用包含于此。
技术领域
本申请属于定位技术领域，具体涉及一种模糊度确认模型的训练方法、模糊度确认方法及装置。
背景技术
随着车道级导航、自动驾驶等技术的快速发展，高精度快速动态定位应用越来越广泛。而模糊度是高精度定位的关键。
模糊度确认结果影响定位结果的固定率和精度，例如，将实际真实有效的模糊度确认为非真实有效的模糊度，会影响定位的固定率，将实际非真实有效的模糊度确认为真实有效的模糊度，会影响定位精度。因此，准确地对模糊度进行确认具有重要意义。
相关技术中，主要是通过经验模型设定一些观测量阈值，将实时观测量与设定的观测量阈值进行比较来对模糊度进行确认。然而，该方式严重依赖经验模型及和所设定的观测量阈值，模糊度确认的准确度较低。
发明内容
本申请实施例的目的是提供一种模糊度确认模型的训练方法、模糊度确认方法及装置，能够解决模糊度确认准确度低的问题。
第一方面，本申请实施例提供了一种模糊度确认模型的训练方法，包括：
获取样本定位数据；
提取样本定位数据的特征信息，其中，特征信息包括：宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差至少其中之一，模糊度差值为两个频点的模糊度的差值，第一位置为根据宽巷模糊度进行定位计算得到的位置，第二位置为根据两个频点中的任一个频点的模糊度进行定位计算得到的位置，第三位置为根据超宽巷模糊度进行定位计算得到的位置；
根据特征信息，训练神经网络模型，得到模糊度确认模型。
第二方面，本申请实施例提供了一种模糊度确认方法，包括：
获取目标定位数据；
提取目标定位数据的特征信息；
将特征信息输入利用本申请实施例第一方面提供的模糊度确认模型的训练方法训练得到的模糊度确认模型，得到目标定位数据对应的模糊度的确认结果。
第三方面，本申请实施例提供了一种模糊度确认模型的训练装置，包括：
第一获取模块，用于获取样本定位数据；
第一提取模块，用于提取样本定位数据的特征信息，其中，特征信息包括：宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差至少其中之一，模糊度差值为两个频点的模糊度的差值，第一位置为根据宽巷模糊度进行定位计算得到的位置，第二位置为根据两个频点中的任一个频点的模糊度进行定位计算得到的位置，第三位置为根据超宽巷模糊度进行定位计算得到的位置；
训练模块，用于根据特征信息，训练神经网络模型，得到模糊度确认模型。
第四方面，本申请实施例提供了一种模糊度确认装置，包括：
第二获取模块，用于获取目标定位数据；
第二提取模块，用于提取目标定位数据的特征信息；
确认模块，用于将特征信息输入利用本申请实施例第一方面提供的模糊度确认模型的训练方法训练得到的模糊度确认模型，得到目标定位数据 对应的模糊度的确认结果。
第五方面，本申请实施例提供了一种电子设备，包括处理器和存储器，存储器存储可在处理器上运行的程序或指令，程序或指令被处理器执行时实现如第一方面或第二方面所述的方法的步骤。
第六方面，本申请实施例提供了一种可读存储介质，可读存储介质上存储程序或指令，程序或指令被处理器执行时实现如第一方面或第二方面所述的方法的步骤。
第七方面，本申请实施例提供了一种芯片，包括处理器和通信接口，通信接口和处理器耦合，处理器用于运行程序或指令，实现如第一方面或第二方面所述的方法的步骤。
第八方面，本申请实施例提供一种计算机程序产品，该程序产品被存储在存储介质中，该程序产品被至少一个处理器执行以实现如第一方面或第二方面所述的方法。
在本申请实施例中，通过获取样本定位数据；提取样本定位数据的特征信息，其中，特征信息包括：宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差至少其中之一，模糊度差值为两个频点的模糊度的差值，第一位置为根据宽巷模糊度进行定位计算得到的位置，第二位置为根据两个频点中的任一个频点的模糊度进行定位计算得到的位置，第三位置为根据超宽巷模糊度进行定位计算得到的位置；根据特征信息，训练神经网络模型，得到模糊度确认模型。进而利用该模糊度确认模型对获取到的目标定位数据对应的模糊度进行确认，即可得到目标定位数据对应的模糊度的确认结果。一方面，无需通过经验模型设定一些观测量阈值，即模糊度确认时不依赖经验模型及和所设定的观测量阈值，能够提高模糊度确认的准确度。另一方面，由于模糊度确认模型是根据上述宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差至少其中之一训练得到的，而宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差是与模糊度确认相关的特征中的重要特征，因此，通过上述三个特征训练得到的模糊度确认模型，能够准确地对模糊度进行确 认，能够提高模糊度确认的准确度。
附图说明
为了更清楚地说明本申请实施例的技术方案，下面将对本申请实施例中所需要使用的附图作简单的介绍，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本申请实施例提供的模糊度确认模型的训练方法的流程示意图；
图2是本申请实施例提供的模糊度确认方法的流程示意图；
图3是本申请实施例提供的模糊度确认的过程示意图；
图4是本申请实施例提供的模糊度确认模型的训练装置的结构示意图；
图5是本申请实施例提供的模糊度确认装置的结构示意图；
图6是本申请实施例提供的电子设备的结构示意图；
图7是实现本申请实施例的电子设备的硬件结构示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚地描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员获得的所有其他实施例，都属于本申请保护的范围。
本申请的说明书和权利要求书中的术语“第一”、“第二”等是用于区别类似的对象，而不用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换，以便本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施，且“第一”、“第二”等所区分的对象通常为一类，并不限定对象的个数，例如第一对象可以是一个，也可以是多个。此外，说明书以及权利要求中“和/或”表示所连接对象的至少其中之一，字符“/”，一般表示前后关联对象是一种“或”的关系。
下面结合附图，通过具体的实施例及其应用场景对本申请实施例提供 的模糊度确认模型的训练方法、模糊度确认方法及装置进行详细地说明。
图1是本申请实施例提供的模糊度确认模型的训练方法的流程示意图。如图1所示，模糊度确认模型的训练方法可以包括：
S101：获取样本定位数据；
S102：提取样本定位数据的特征信息，其中，特征信息包括：宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差至少其中之一，模糊度差值为两个频点的模糊度的差值，第一位置为根据宽巷模糊度进行定位计算得到的位置，第二位置为根据两个频点中的任一个频点的模糊度进行定位计算得到的位置，第三位置为根据超宽巷模糊度进行定位计算得到的位置；
S103：根据特征信息，训练神经网络模型，得到模糊度确认模型。
上述各步骤的具体实现方式将在下文中进行详细描述。
在本申请实施例中，通过获取样本定位数据；提取样本定位数据的特征信息，其中，特征信息包括：宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差至少其中之一，模糊度差值为两个频点的模糊度的差值，第一位置为根据宽巷模糊度进行定位计算得到的位置，第二位置为根据两个频点中的任一个频点的模糊度进行定位计算得到的位置，第三位置为根据超宽巷模糊度进行定位计算得到的位置；根据特征信息，训练神经网络模型，得到模糊度确认模型。进而利用该模糊度确认模型对获取到的目标定位数据对应的模糊度进行确认，即可得到目标定位数据对应的模糊度的确认结果。一方面，无需通过经验模型设定一些观测量阈值，即模糊度确认时不依赖经验模型及和所设定的观测量阈值，能够提高模糊度确认的准确度。另一方面，由于模糊度确认模型是根据上述宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差至少其中之一训练得到的，而宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差是与模糊度确认相关的特征中的重要特征，因此，通过上述三个特征训练得到的模糊度确认模型，能够准确地对模糊度进行确认，能够提高模糊度确认的准确度。
在本申请实施例的一些可能实现中，本申请实施例中的两个频点可以为满足组成宽巷条件的任意两个频点。
在本申请实施例的一些可能实现中，本申请实施例中的宽巷模糊度可以为将上述两个频点对应的载波相位观测值进行组合，得到载波相位组合观测值，根据该载波相位组合观测值固定的模糊度。
在本申请实施例的一些可能实现中，本申请实施例中的超宽巷模糊度可以为将满足组成超宽巷条件的至少两个频点的载波相位观测值进行组合，得到载波相位组合观测值，根据该载波相位组合观测值固定的模糊度。
在本申请实施例的一些可能实现中，可以预先对样本定位数据的模糊度进行标记。在S103中，可以将特征信息和其对应的模糊度实际标记输入到神经网络模型，得到对该特征信息对应的样本定位数据的模糊度的预测结果，根据模糊度实际标记和该预测结果训练神经网络模型，得到模糊度确认模型，其中，在训练网络模型时，模糊度实际标记作为其对应的样本定位数据的特征信息的期望输出。
在本申请实施例的一些可能实现中，在训练神经网络模型时，可以设定循环迭代训练次数阈值，当对神经网络模型训练的次数到达设定的循环迭代训练次数阈值时，停止训练，得到模糊度确认模型。
在本申请实施例的一些可能实现中，在训练神经网络模型时，还可以计算神经网络模型的损失值，当神经网络模型的损失值到达最小时，停止训练，得到模糊度确认模型。
在本申请实施例的一些可能实现中，在训练神经网络模型时，还可以根据模糊度实际标记和上述预测结果，计算神经网络模型预测的准确度，当神经网络模型预测的准确度大于准确度阈值时，停止训练，得到模糊度确认模型。
具体地，可以从样本定位数据中选取部分样本定位数据作为测试样本。当神经网络模型训练一段时间后，将测试样本的特征信息输入到此时的神经网络模型，得到对测试样本的模糊度的预测结果，根据该模糊度的预测结果和测试样本的模糊度的实际标记，计算此时神经网络模型预测的准确度，当神经网络模型预测的准确度不大于准确度阈值时，利用样本定位数 据的特征信息继续对神经网络模型进行训练，一段时间后，再将测试样本的特征信息输入到此时的神经网络模型，计算此时神经网络模型预测的准确度，直至神经网络模型预测的准确度大于准确度阈值，将此时的神经网络模型作为模糊度确认模型。
本申请实施例并不对根据特征信息，训练神经网络模型，得到模糊度确认模型的具体过程进行限定，根据特征信息，训练神经网络模型可以参考相关技术中训练神经网络模型的过程，本申请实施例在此不对其进行赘述。
在本申请实施例的一些可能实现中，本申请实施例中的特征信息还可以包括以下所列项中的至少三种：
固定模糊度数量、验后单位权中误差、精度因子值、参考卫星改变的个数、单点解定位结果的好坏标志、有效伪距个数、伪距残差、卫星数量、根据备选模糊度得到的最小单位权方差与次小单位权方差的比值、固定解载波双差残差的均值、平均载噪比。
在本申请实施例的一些可能实现中，由于对模糊度进行确认时，模糊度确认结果仅有两种情况，一种情况是模糊度真实有效(即模糊度确认通过)，另一种情况是模糊度不是真实有效的(即模糊度确认不通过)。因此，本申请实施例中的神经网络模型为二分类模型。利用二分类模型进行分类时，预测的结果是上述两种情况中的一种情况，非此即彼。
在本申请实施例的一些可能实现中，本申请实施例中的二分类模型可以为：基于逻辑回归的二分类模型、基于K最邻近算法的二分类模型、基于二叉决策树的二分类模型、基于支持向量机的二分类模型、基于朴素贝叶斯的二分类模型。
在本申请实施例中，本申请实施例中的二分类模型优选为：基于逻辑回归的二分类模型。
在本申请实施例的一些可能实现中，本申请实施例中的基于逻辑回归的二分类模型的分类函数为Sigmoid函数，Sigmoid函数如下公式(1)所示：
Figure PCTCN2022116375-appb-000001
本申请实施例中的基于逻辑回归的二分类模型的假设函数如下公式(2)所示：
Figure PCTCN2022116375-appb-000002
其中，公式(2)中，x为特征信息对应的向量，θ为需要拟合的权重向量。
在本申请实施例的一些可能实现中，可以通过最大似然估计求解θ。其中，通过最大似然估计求解θ是求使基于逻辑回归的二分类模型的对数似然函数取最大值时的θ，此时，求解得到的θ为基于逻辑回归的二分类模型的最佳参数。
在本申请实施例的一些可能实现中，可以采用梯度上升法求解θ。
图2是本申请实施例提供的模糊度确认方法的流程示意图。如图2所示，模糊度确认方法可以包括：
S201：获取目标定位数据；
S202：提取目标定位数据的特征信息；
S203：将特征信息输入利用本申请实施例提供的模糊度确认模型的训练方法训练得到的模糊度确认模型，得到目标定位数据对应的模糊度的确认结果。
在本申请实施例中，通过模糊度确认模型即可对获取到的目标定位数据对应的模糊度进行确认，得到目标定位数据对应的模糊度的确认结果。一方面，无需通过经验模型设定一些观测量阈值，即模糊度确认时不依赖经验模型及和所设定的观测量阈值，能够提高模糊度确认的准确度。另一方面，由于模糊度确认模型是根据宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差至少其中之一训练得到的，其中，模糊度差值为两个频点的模糊度的差值，第一位置为根据宽巷模糊度进行定位计算得到的位置，第二位置为根据两个频点中的任一个频点的模糊度进行定位计算得到的位置，第三位置为根据超宽巷模糊 度进行定位计算得到的位置，而宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差是与模糊度确认相关的特征中的重要特征，因此，通过上述三个特征训练得到的模糊度确认模型，能够准确地对模糊度进行确认，能够提高模糊度确认的准确度。
需要说明的是，在对目标定位数据提取特征信息时，提取的特征信息是与训练神经网络模型时相同的特征的信息。
在本申请实施例的一些可能实现中，模糊度的确认结果包括：模糊度真实有效(即模糊度正确，通过确认)、模糊度非真实有效(即模糊度错误，不通过确认)。
在本申请实施例的一些可能实现中，本申请实施例中确认的模糊度可以为整周模糊度(ambiguity of whole cycles)。整周模糊度又称整周未知数，是在全球定位系统技术的载波相位测量时，载波相位与基准相位之间相位差的首观测值所对应的整周未知数。
图3是本申请实施例提供的模糊度确认的过程示意图。其中，模糊度确认的过程包括两个阶段，分别为离线训练阶段和在线确认阶段。
在离线训练阶段，将样本定位数据的特征信息以及相应的模糊度实际标记输入到基于机器学习的二分类模型(即神经网络模型)，以对该二分类模型进行训练，得到用于模糊度确认的模糊度确认模型。
在在线确认阶段，提取实际定位数据的特征信息，将该特征信息输入到离线训练阶段训练得到的模糊度确认模型，即可得到对实际定位数据对应的模糊度的确认结果。
需要说明的是，本申请实施例提供的模糊度确认模型的训练方法，执行主体可以为模糊度确认模型的训练装置。本申请实施例中以模糊度确认模型的训练装置执行模糊度确认模型的训练方法为例，说明本申请实施例提供的模糊度确认模型的训练装置。
图4是本申请实施例提供的模糊度确认模型的训练装置的结构示意图。如图4所示，模糊度确认模型的训练装置400可以包括：
第一获取模块401，用于获取样本定位数据；
第一提取模块402，用于提取样本定位数据的特征信息，其中，特征 信息包括：宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差至少其中之一，模糊度差值为两个频点的模糊度的差值，第一位置为根据宽巷模糊度进行定位计算得到的位置，第二位置为根据两个频点中的任一个频点的模糊度进行定位计算得到的位置，第三位置为根据超宽巷模糊度进行定位计算得到的位置；
训练模块403，用于根据特征信息，训练神经网络模型，得到模糊度确认模型。
在本申请实施例中，通过获取样本定位数据；提取样本定位数据的特征信息，其中，特征信息包括：宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差至少其中之一，模糊度差值为两个频点的模糊度的差值，第一位置为根据宽巷模糊度进行定位计算得到的位置，第二位置为根据两个频点中的任一个频点的模糊度进行定位计算得到的位置，第三位置为根据超宽巷模糊度进行定位计算得到的位置；根据特征信息，训练神经网络模型，得到模糊度确认模型。进而利用该模糊度确认模型对获取到的目标定位数据对应的模糊度进行确认，即可得到目标定位数据对应的模糊度的确认结果。一方面，无需通过经验模型设定一些观测量阈值，即模糊度确认时不依赖经验模型及和所设定的观测量阈值，能够提高模糊度确认的准确度。另一方面，由于模糊度确认模型是根据上述宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差至少其中之一训练得到的，而宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差是与模糊度确认相关的特征中的重要特征，因此，通过上述三个特征训练得到的模糊度确认模型，能够准确地对模糊度进行确认，能够提高模糊度确认的准确度。
在本申请实施例的一些可能实现中，特征信息还包括以下所列项中的至少三种：
固定模糊度数量、验后单位权中误差、精度因子值、参考卫星改变的个数、单点解定位结果的好坏标志、有效伪距个数、伪距残差、卫星数量、根据备选模糊度得到的最小单位权方差与次小单位权方差的比值、固定解 载波双差残差的均值、平均载噪比。
在本申请实施例的一些可能实现中，神经网络模型，包括：
二分类模型。
在本申请实施例的一些可能实现中，二分类模型，包括：
基于逻辑回归的二分类模型。
需要说明的是，本申请实施例提供的模糊度确认方法，执行主体可以为模糊度确认装置。本申请实施例中以模糊度确认装置执行模糊度确认方法为例，说明本申请实施例提供的模糊度确认装置。
图5是本申请实施例提供的模糊度确认装置的结构示意图。如图5所示，模糊度确认装置500可以包括：
第二获取模块501，用于获取目标定位数据；
第二提取模块502，用于提取目标定位数据的特征信息；
确认模块503，用于将特征信息输入利用本申请实施例提供的模糊度确认模型的训练方法训练得到的模糊度确认模型，得到目标定位数据对应的模糊度的确认结果。
在本申请实施例中，通过模糊度确认模型即可对获取到的目标定位数据对应的模糊度进行确认，得到目标定位数据对应的模糊度的确认结果。一方面，无需通过经验模型设定一些观测量阈值，即模糊度确认时不依赖经验模型及和所设定的观测量阈值，能够提高模糊度确认的准确度。另一方面，由于模糊度确认模型是根据宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差至少其中之一训练得到的，其中，模糊度差值为两个频点的模糊度的差值，第一位置为根据宽巷模糊度进行定位计算得到的位置，第二位置为根据两个频点中的任一个频点的模糊度进行定位计算得到的位置，第三位置为根据超宽巷模糊度进行定位计算得到的位置，而宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差是与模糊度确认相关的特征中的重要特征，因此，通过上述三个特征训练得到的模糊度确认模型，能够准确地对模糊度进行确认，能够提高模糊度确认的准确度。
本申请实施例中的模糊度确认模型的训练装置或模糊度确认装置可以 是电子设备，也可以是电子设备中的部件，例如集成电路、或芯片。该电子设备可以是终端，也可以为除终端之外的其他设备。示例性的，电子设备可以为手机、平板电脑、笔记本电脑、掌上电脑、车载电子设备、移动上网装置(Mobile Internet Device，MID)、增强现实(augmented reality，AR)/虚拟现实(virtual reality，VR)设备、机器人、可穿戴设备、超级移动个人计算机(ultra-mobile personal computer，UMPC)、上网本或者个人数字助理(personal digital assistant，PDA)等，还可以为服务器、网络附属存储器(Network Attached Storage，NAS)、个人计算机(personal computer，PC)、电视机(television，TV)、柜员机或者自助机等，本申请实施例不作具体限定。
本申请实施例中的模糊度确认模型的训练装置或模糊度确认装置可以为具有操作系统的装置。该操作系统可以为安卓(Android)操作系统，可以为iOS操作系统，还可以为其他可能的操作系统，本申请实施例不作具体限定。
本申请实施例提供的模糊度确认模型的训练装置能够实现本申请实施例提供的模糊度确认模型的训练方法实施例中的各个过程，为避免重复，这里不再赘述。
本申请实施例提供的模糊度确认装置能够实现本申请实施例提供的模糊度确认方法实施例中的各个过程，为避免重复，这里不再赘述。
可选的，如图6所示，本申请实施例还提供一种电子设备600，包括处理器601和存储器602，存储器602存储有可在处理器601上运行的程序或指令，该程序或指令被处理器601执行时实现上述模糊度确认模型的训练方法或模糊度确认方法实施例的各个步骤，且能达到相同的技术效果，为避免重复，这里不再赘述。
在本申请实施例的一些可能实现中，处理器601可以包括中央处理器(CPU)，或者特定集成电路(Application Specific Integrated Circuit，ASIC)，或者可以被配置成实施本申请实施例的一个或多个集成电路。
在本申请实施例的一些可能实现中，存储器602可以包括只读存储器(Read-Only Memory，ROM)，随机存取存储器(Random Access  Memory，RAM)，磁盘存储介质设备，光存储介质设备，闪存设备，电气、光学或其他物理/有形的存储器存储设备。因此，通常，存储器602包括一个或多个编码有包括计算机可执行指令的软件的有形(非暂态)计算机可读存储介质(例如，存储器设备)，并且当该软件被执行(例如，由一个或多个处理器)时，其可操作来执行参考根据本申请实施例提供的模糊度确认模型的训练方法或模糊度确认方法所描述的操作。
图7是实现本申请实施例的电子设备的硬件结构示意图。
该电子设备700包括但不限于：射频单元701、网络模块702、音频输出单元703、输入单元704、传感器705、显示单元706、用户输入单元707、接口单元708、存储器709、以及处理器710等部件。
本领域技术人员可以理解，电子设备700还可以包括给各个部件供电的电源(比如电池)，电源可以通过电源管理系统与处理器710逻辑相连，从而通过电源管理系统实现管理充电、放电、以及功耗管理等功能。图7中示出的电子设备结构并不构成对电子设备的限定，电子设备可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置，在此不再赘述。
其中，处理器710用于：获取样本定位数据；提取样本定位数据的特征信息，其中，特征信息包括：宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差至少其中之一，模糊度差值为两个频点的模糊度的差值，第一位置为根据宽巷模糊度进行定位计算得到的位置，第二位置为根据两个频点中的任一个频点的模糊度进行定位计算得到的位置，第三位置为根据超宽巷模糊度进行定位计算得到的位置；根据特征信息，训练神经网络模型，得到模糊度确认模型。
在本申请实施例的一些可能实现中，特征信息还包括以下所列项中的至少三种：
固定模糊度数量、验后单位权中误差、精度因子值、参考卫星改变的个数、单点解定位结果的好坏标志、有效伪距个数、伪距残差、卫星数量、根据备选模糊度得到的最小单位权方差与次小单位权方差的比值、固定解载波双差残差的均值、平均载噪比。
在本申请实施例的一些可能实现中，神经网络模型，包括：
二分类模型。
在本申请实施例的一些可能实现中，二分类模型，包括：
基于逻辑回归的二分类模型。
在本申请实施例的一些可能实现中，处理器710还可以用于：获取目标定位数据；提取目标定位数据的特征信息；将特征信息输入上述模糊度确认模型，得到目标定位数据对应的模糊度的确认结果。
在本申请实施例中，通过模糊度确认模型即可对获取到的目标定位数据对应的模糊度进行确认，得到目标定位数据对应的模糊度的确认结果。一方面，无需通过经验模型设定一些观测量阈值，即模糊度确认时不依赖经验模型及和所设定的观测量阈值，能够提高模糊度确认的准确度。另一方面，由于模糊度确认模型是根据宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差至少其中之一训练得到的，其中，模糊度差值为两个频点的模糊度的差值，第一位置为根据宽巷模糊度进行定位计算得到的位置，第二位置为根据两个频点中的任一个频点的模糊度进行定位计算得到的位置，第三位置为根据超宽巷模糊度进行定位计算得到的位置，而宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、第一位置与第三位置的位置差是与模糊度确认相关的特征中的重要特征，因此，通过上述三个特征训练得到的模糊度确认模型，能够准确地对模糊度进行确认，能够提高模糊度确认的准确度。
应理解的是，本申请实施例中，输入单元704可以包括图形处理器(Graphics Processing Unit，GPU)7041和麦克风7042，图形处理器7041对在视频捕获模式或图像捕获模式中由图像捕获装置(如摄像头)获得的静态图片或视频的图像数据进行处理。显示单元706可包括显示面板7061，可以采用液晶显示器、有机发光二极管等形式来配置显示面板7061。用户输入单元707包括触控面板7071以及其他输入设备7072中的至少一种。触控面板7071，也称为触摸屏。触控面板7071可包括触摸检测装置和触摸控制器两个部分。其他输入设备7072可以包括但不限于物理键盘、功能键(比如音量控制按键、开关按键等)、轨迹球、鼠标、操作杆，在此 不再赘述。
存储器709可用于存储软件程序以及各种数据。存储器709可主要包括存储程序或指令的第一存储区和存储数据的第二存储区，其中，第一存储区可存储操作系统、至少一个功能所需的应用程序或指令(比如声音播放功能、图像播放功能等)等。此外，存储器709可以包括易失性存储器或非易失性存储器，或者，存储器709可以包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(Read-Only Memory，ROM)、可编程只读存储器(Programmable ROM，PROM)、可擦除可编程只读存储器(Erasable PROM，EPROM)、电可擦除可编程只读存储器(Electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(Random Access Memory，RAM)，静态随机存取存储器(Static RAM，SRAM)、动态随机存取存储器(Dynamic RAM，DRAM)、同步动态随机存取存储器(Synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(Double Data Rate SDRAM，DDRSDRAM)、增强型同步动态随机存取存储器(Enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(Synch link DRAM，SLDRAM)和直接内存总线随机存取存储器(Direct Rambus RAM，DRRAM)。本申请实施例中的存储器709包括但不限于这些和任意其它适合类型的存储器。
处理器710可包括一个或多个处理单元；可选的，处理器710集成应用处理器和调制解调处理器，其中，应用处理器主要处理涉及操作系统、用户界面和应用程序等的操作，调制解调处理器主要处理无线通信信号，如基带处理器。可以理解的是，上述调制解调处理器也可以不集成到处理器710中。
本申请实施例还提供一种可读存储介质，可读存储介质上存储有程序或指令，该程序或指令被处理器执行时实现上述模糊度确认模型的训练方法或模糊度确认方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
其中，处理器为上述实施例中的电子设备中的处理器。可读存储介质包括计算机可读存储介质，计算机可读存储介质的示例包括非暂态计算机 可读存储介质，如ROM、RAM、磁碟或者光盘等。
本申请实施例另提供了一种芯片，包括处理器和通信接口，通信接口和处理器耦合，处理器用于运行程序或指令，实现上述模糊度确认模型的训练方法或模糊度确认方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
应理解，本申请实施例提到的芯片还可以称为系统级芯片、系统芯片、芯片系统或片上系统芯片等。
本申请实施例提供一种计算机程序产品，该程序产品被存储在存储介质中，该程序产品被至少一个处理器执行以实现如上述模糊度确认模型的训练方法或模糊度确认方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。此外，需要指出的是，本申请实施方式中的方法和装置的范围不限按示出或讨论的顺序来执行功能，还可包括根据所涉及的功能按基本同时的方式或按相反的顺序来执行功能，例如，可以按不同于所描述的次序来执行所描述的方法，并且还可以添加、省去、或组合各种步骤。另外，参照某些示例所描述的特征可在其他示例中被组合。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对相关技术做出贡献的部分可以以计算机软件产品的形式体现出来，该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端(可以是手机，计算机，服务器，或者网络设备等)执行本申请各个实施例所述的 方法。
上面结合附图对本申请的实施例进行了描述，但是本申请并不局限于上述的具体实施方式，上述的具体实施方式仅仅是示意性的，而不是限制性的，本领域的普通技术人员在本申请的启示下，在不脱离本申请宗旨和权利要求所保护的范围情况下，还可做出很多形式，均属于本申请的保护之内。

Claims (12)

  1. 一种模糊度确认模型的训练方法，包括：
    获取样本定位数据；
    提取所述样本定位数据的特征信息，其中，所述特征信息包括：宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、所述第一位置与第三位置的位置差至少其中之一，所述模糊度差值为两个频点的模糊度的差值，所述第一位置为根据所述宽巷模糊度进行定位计算得到的位置，所述第二位置为根据所述两个频点中的任一个频点的模糊度进行定位计算得到的位置，所述第三位置为根据超宽巷模糊度进行定位计算得到的位置；
    根据所述特征信息，训练神经网络模型，得到模糊度确认模型。
  2. 根据权利要求1所述的方法，其中，所述特征信息还包括以下所列项中的至少三种：
    固定模糊度数量、验后单位权中误差、精度因子值、参考卫星改变的个数、单点解定位结果的好坏标志、有效伪距个数、伪距残差、卫星数量、根据备选模糊度得到的最小单位权方差与次小单位权方差的比值、固定解载波双差残差的均值、平均载噪比。
  3. 根据权利要求1所述的方法，其中，所述神经网络模型，包括：
    二分类模型。
  4. 根据权利要求3所述的方法，其中，所述二分类模型，包括：
    基于逻辑回归的二分类模型。
  5. 一种模糊度确认方法，包括：
    获取目标定位数据；
    提取所述目标定位数据的特征信息；
    将所述特征信息输入利用权利要求1至4任一项所述的模糊度确认模型的训练方法训练得到的模糊度确认模型，得到所述目标定位数据对应的模糊度的确认结果。
  6. 一种模糊度确认模型的训练装置，包括：
    第一获取模块，用于获取样本定位数据；
    第一提取模块，用于提取所述样本定位数据的特征信息，其中，所述特征信息包括：宽巷模糊度与模糊度差值的差值、第一位置与第二位置的位置差、所述第一位置与第三位置的位置差至少其中之一，所述模糊度差值为两个频点的模糊度的差值，所述第一位置为根据所述宽巷模糊度进行定位计算得到的位置，所述第二位置为根据所述两个频点中的任一个频点的模糊度进行定位计算得到的位置，所述第三位置为根据超宽巷模糊度进行定位计算得到的位置；
    训练模块，用于根据所述特征信息，训练神经网络模型，得到模糊度确认模型。
  7. 根据权利要求6所述的装置，其中，所述特征信息还包括以下所列项中的至少三种：
    固定模糊度数量、验后单位权中误差、精度因子值、参考卫星改变的个数、单点解定位结果的好坏标志、有效伪距个数、伪距残差、卫星数量、根据备选模糊度得到的最小单位权方差与次小单位权方差的比值、固定解载波双差残差的均值、平均载噪比。
  8. 根据权利要求6所述的装置，其中，所述神经网络模型，包括：
    二分类模型。
  9. 根据权利要求8所述的装置，其中，所述二分类模型，包括：
    基于逻辑回归的二分类模型。
  10. 一种模糊度确认装置，包括：
    第二获取模块，用于获取目标定位数据；
    第二提取模块，用于提取所述目标定位数据的特征信息；
    确认模块，用于将所述特征信息输入利用权利要求1至4任一项所述的模糊度确认模型的训练方法训练得到的模糊度确认模型，得到所述目标定位数据对应的模糊度的确认结果。
  11. 一种电子设备，包括：处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如权利要求1至4任一项所述的模糊度确认模型的训练方法或如权利要求5所述的模糊度确认方法的步骤。
  12. 一种可读存储介质，所述可读存储介质上存储程序或指令，所述程序或指令被处理器执行时实现如权利要求1至4任一项所述的模糊度确认模型的训练方法或如权利要求5所述的模糊度确认方法的步骤。
PCT/CN2022/116375 2021-12-21 2022-08-31 模糊度确认模型的训练方法、模糊度确认方法及装置 WO2023116043A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111572245.8A CN116304654A (zh) 2021-12-21 2021-12-21 模糊度确认模型的训练方法、模糊度确认方法及装置
CN202111572245.8 2021-12-21

Publications (1)

Publication Number Publication Date
WO2023116043A1 true WO2023116043A1 (zh) 2023-06-29

Family

ID=86822680

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/116375 WO2023116043A1 (zh) 2021-12-21 2022-08-31 模糊度确认模型的训练方法、模糊度确认方法及装置

Country Status (2)

Country Link
CN (1) CN116304654A (zh)
WO (1) WO2023116043A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117092679A (zh) * 2023-10-19 2023-11-21 北京凯芯微科技有限公司 一种用于rtk模糊度固定判决的人工神经网络的训练方法

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180299562A1 (en) * 2017-04-12 2018-10-18 Coherent Technical Services, Inc. Assured validation of carrier-phase integer ambiguities for safety-of-life applications
CN109191457A (zh) * 2018-09-21 2019-01-11 中国人民解放军总医院 一种病理图像质量有效性识别方法
CN110907973A (zh) * 2018-09-14 2020-03-24 千寻位置网络有限公司 网络rtk基线双差模糊度检验方法、装置及定位方法
CN112083463A (zh) * 2019-06-13 2020-12-15 千寻位置网络有限公司 模糊度是否固定正确的检测方法及装置、定位终端
US20210255335A1 (en) * 2020-02-13 2021-08-19 Mitsubishi Electric Research Laboratories, Inc. System and Method for GNSS Ambiguity Resolution

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20180299562A1 (en) * 2017-04-12 2018-10-18 Coherent Technical Services, Inc. Assured validation of carrier-phase integer ambiguities for safety-of-life applications
CN110907973A (zh) * 2018-09-14 2020-03-24 千寻位置网络有限公司 网络rtk基线双差模糊度检验方法、装置及定位方法
CN109191457A (zh) * 2018-09-21 2019-01-11 中国人民解放军总医院 一种病理图像质量有效性识别方法
CN112083463A (zh) * 2019-06-13 2020-12-15 千寻位置网络有限公司 模糊度是否固定正确的检测方法及装置、定位终端
US20210255335A1 (en) * 2020-02-13 2021-08-19 Mitsubishi Electric Research Laboratories, Inc. System and Method for GNSS Ambiguity Resolution

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117092679A (zh) * 2023-10-19 2023-11-21 北京凯芯微科技有限公司 一种用于rtk模糊度固定判决的人工神经网络的训练方法
CN117092679B (zh) * 2023-10-19 2024-01-30 北京凯芯微科技有限公司 一种用于rtk模糊度固定判决的人工神经网络的训练方法

Also Published As

Publication number Publication date
CN116304654A (zh) 2023-06-23

Similar Documents

Publication Publication Date Title
WO2022213879A1 (zh) 目标对象检测方法、装置、计算机设备和存储介质
CN111476306B (zh) 基于人工智能的物体检测方法、装置、设备及存储介质
US20190370633A1 (en) Intelligent image captioning
US10620713B1 (en) Methods and systems for touchless control with a mobile device
US20160307334A1 (en) Non-matching feature-based visual motion estimation for pose determination
CN113792207A (zh) 一种基于多层次特征表示对齐的跨模态检索方法
WO2023116043A1 (zh) 模糊度确认模型的训练方法、模糊度确认方法及装置
WO2021179036A1 (en) Systems and methods for image-based location determination
US9208404B2 (en) Object detection with boosted exemplars
CN105354841A (zh) 一种快速遥感影像匹配方法及系统
CN114449439B (zh) 一种地下管廊空间的定位方法及装置
CN109448018B (zh) 跟踪目标的定位方法、装置、设备及存储介质
CN108061878A (zh) 基于移动终端的融合定位方法、存储装置及移动终端
CN110728172B (zh) 基于点云的人脸关键点检测方法、装置、系统及存储介质
CN111161238A (zh) 图像质量评价方法及装置、电子设备、存储介质
CN116798027A (zh) 基于多尺度特征提取的三维点云实时云匹配方法及装置
CN116580056A (zh) 一种船舶检测与跟踪方法、装置、电子设备及存储介质
CN113344994B (zh) 图像配准方法、装置、电子设备及存储介质
CN115952248A (zh) 终端设备的位姿处理方法、装置、设备、介质及产品
CN114648556A (zh) 视觉跟踪方法、装置和电子设备
CN116295466A (zh) 地图生成方法、装置、电子设备、存储介质、及车辆
CN112614168B (zh) 一种目标人脸跟踪方法、装置、电子设备及存储介质
CN114298204A (zh) 设备屏幕划痕检测模型训练方法及设备屏幕划痕检测方法
CN111008210B (zh) 商品识别方法、装置、编解码器及存储装置
CN114913541A (zh) 基于正交匹配追踪的人体关键点检测方法、装置及介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22909368

Country of ref document: EP

Kind code of ref document: A1