(19)
(11)EP 3 410 074 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
22.07.2020 Bulletin 2020/30

(21)Application number: 16909376.2

(22)Date of filing:  08.10.2016
(51)International Patent Classification (IPC): 
G01C 21/16(2006.01)
(86)International application number:
PCT/CN2016/101442
(87)International publication number:
WO 2018/014449 (25.01.2018 Gazette  2018/04)

(54)

METHOD AND DEVICE FOR IMPROVING PERFORMANCE OF RELATIVE-POSITION SENSOR, AND COMPUTER STORAGE MEDIUM

VERFAHREN UND VORRICHTUNG ZUR VERBESSERUNG DER LEISTUNG EINES SENSORS FÜR RELATIVE POSITION UND COMPUTERSPEICHERMEDIUM

PROCÉDÉ ET DISPOSITIF POUR AMÉLIORER LES PERFORMANCES D'UN CAPTEUR DE POSITION RELATIVE, ET SUPPORT DE STOCKAGE INFORMATIQUE


(84)Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

(30)Priority: 19.07.2016 CN 201610569714

(43)Date of publication of application:
05.12.2018 Bulletin 2018/49

(73)Proprietor: Ninebot (Beijing) Tech Co., Ltd.
Beijing 100192 (CN)

(72)Inventors:
  • DONG, Shiqian
    Beijing 100192 (CN)
  • REN, Guanjiao
    Beijing 100192 (CN)

(74)Representative: Regimbeau 
20, rue de Chazelles
75847 Paris Cedex 17
75847 Paris Cedex 17 (FR)


(56)References cited: : 
CN-A- 1 737 580
CN-A- 102 308 183
CN-A- 103 149 939
CN-A- 104 111 058
CN-A- 105 698 765
US-A1- 2011 209 544
CN-A- 102 171 628
CN-A- 103 134 489
CN-A- 103 221 788
CN-A- 105 651 242
US-A1- 2009 278 791
US-B1- 6 474 159
  
      
    Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


    Description

    TECHNICAL FIELD



    [0001] The disclosure relates to the field of electronic technology, and in particular to a method, apparatus and a computer storage medium for improving performance of a relative position sensor.

    BACKGROUND



    [0002] A relative position sensor is configured to measure a relative position (e.g., a relative angle and a relative distance) between two objects, is widely applied to the field of robots, and may be used for tracking the measured objects.

    [0003] However, performance of current relative position sensors are commonly poor, and measurement data are likely to fluctuate due to interference, which may cause that the measurement data is inaccurate and dynamic performance are poor, thus severely affecting application of relative position measurement sensors.

    [0004] In conclusion, how to improve performance of a relative position sensor has become a problem to be urgently solved at the present stage.

    [0005] CN 105 698 765 A discloses a method using combination of double IMUS (Inertial Measurement Units) and monocular vision to measure pose of target object under non-inertial system.

    [0006] US 6 474 159 B1 discloses a method for using inertial head-tracking systems on-board moving platforms by computing the motion of a "tracking" Inertial Measurement Unit (IMU) mounted on the object being tracked relative to a "reference" IMU rigidly attached to the moving platform.

    SUMMARY



    [0007] The embodiments of the disclosure provide a method, apparatus and a computer storage medium for improving performance of a relative position sensor, capable of solving the technical problems of inaccurate measurement data and poor dynamic performance caused by the fact that measurement data of a relative position sensor is likely to fluctuate due to interference in the conventional art.

    [0008] The disclosure is defined in the appended independent claims. Advantageous features are set out in the appended dependent claims. In the following, references to embodiments not falling within the scope of the claims are to be understood as examples useful for understanding the invention.

    [0009] One or more technical solutions provided in the embodiments of the disclosure at least have the technical effects or advantages as follows.

    [0010] In the embodiments of the disclosure, a method, apparatus and a computer storage medium for improving performance of a relative position sensor are disclosed. They are applied to a measurement apparatus. First measurement data measured by a relative position sensor is acquired, and second measurement data measured by a first auxiliary sensor and a second auxiliary sensor is acquired. An EKF is constructed on the basis of the first measurement data, the second measurement data, and the third measurement data. The first measurement data is corrected by using the EKF. Since the first measurement data measured by the relative position sensor is corrected by using the EKF, the technical problems of inaccurate measurement data and poor dynamic performance caused by the fact that measurement data of the relative position sensor is likely to fluctuate due to interference in the conventional art are effectively solved, thereby achieving the technical effect of increasing the accuracy of measurement data of the relative position sensor so as to improve dynamic performance of the relative position sensor.

    BRIEF DESCRIPTION OF THE DRAWINGS



    [0011] In order to more clearly illustrate the technical solutions in the embodiments of the disclosure, drawings required for describing the embodiments will be simply introduced. Apparently, the drawings described below are only some embodiments of the disclosure. On the premise of no creative work, those skilled in the art can obtain other drawings according to these drawings.

    Fig. 1 is a flowchart showing a method for improving performance of a relative position sensor according to an embodiment of the disclosure;

    Figs. 2 and 3 are schematic views illustrating a model of a system consisting of a measurement apparatus and a measured object according to an embodiment of the disclosure; and

    Fig. 4 is a schematic view illustrating a structure of a measurement apparatus according to an embodiment of the disclosure.


    DETAILED DESCRIPTION



    [0012] The embodiments of the disclosure provide a method and apparatus for improving performance of a relative position sensor, capable of solving the technical problems of inaccurate measurement data and poor dynamic performance caused by the fact that measurement data of the relative position sensor is likely to fluctuate due to interference in the conventional art.

    [0013] To solve the technical problems, the concept of the technical solution in the embodiments of the disclosure is as follows.

    [0014] A method for improving performance of a relative position sensor is applied to a measurement apparatus. The measurement apparatus is configured to measure a relative position between a measured object and the measurement apparatus. A relative position sensor and a first auxiliary sensor are disposed on the measurement apparatus. A second auxiliary sensor is disposed on the measured object. The method includes the steps as follows. First measurement data measured by a relative position sensor is acquired, and second measurement data measured by a first auxiliary sensor and a second auxiliary sensor is acquired. An EKF is constructed on the basis of the first measurement data and the second measurement data. The first measurement data is corrected by using the EKF.

    [0015] In order to better understand the above technical solution, the above technical solution will be elaborated below in conjunction with the drawings of the specification and specific implementations.

    [0016] It is to be noted that terms 'and/or' appearing herein are only an association relation for describing associated objects, which represents that three relations may exist. For example, A and/or B may represent three situations that A independently exists, A and B simultaneously exist, and B independently exists. In addition, a character '/' herein generally represents that associated objects are in an 'or' relation.

    Embodiment 1



    [0017] This embodiment provides a method for improving performance of a relative position sensor, applied to a measurement apparatus. The measurement apparatus is configured to measure a relative position between a measured object and the measurement apparatus, wherein a relative position sensor and a first auxiliary sensor are disposed on the measurement apparatus, and a second auxiliary sensor is disposed on the measured object.

    [0018] In a specific implementation process, the measurement apparatus may be a device such as a ground robot, a self-balance car, an unmanned aerial vehicle or an electric vehicle. Here, the specific type of the measurement apparatus is not specifically limited in this embodiment.

    [0019] In a specific implementation process, the measured object may be a stationary person or object.

    [0020] In a specific implementation process, a positioning apparatus may be disposed on the measured object (or the measured object carries the positioning apparatus), the second auxiliary sensor is disposed in the positioning apparatus, and the measurement apparatus actually measures a relative position between the positioning apparatus and the measurement apparatus, wherein the positioning apparatus may be a smart phone, a pad, a remote control key, fitness equipment, a personal digital assistant or a game console.

    [0021] In a specific implementation process, an Inertial Measurement Unit (IMU) is an electronic system consisting of one or more acceleration sensors and angular speed sensors, a microprocessor and a peripheral circuit.

    [0022] In the embodiment of the disclosure, the first auxiliary sensor or the second auxiliary sensor includes: the IMU (integrated with a gyroscope, an accelerometer, an electronic compass and other devices), a speed measurement sensor (e.g., a coded disc or a light flow sensor), and the like. Generally, the gyroscope, the accelerometer and the electronic compass may be regarded as IMUs.

    [0023] As shown in Fig. 1, the method for improving performance of a relative position sensor includes the steps as follows.

    [0024] In Step S101, first measurement data measured by a relative position sensor is acquired, and second measurement data measured by a first auxiliary sensor and a second auxiliary sensor is acquired.

    [0025] In a specific implementation process, the first measurement data includes:
    a relative angle

    between the measurement apparatus and the measured object, and a relative distance

    between the measurement apparatus and the measured object. Here,

    and

    are noise-containing data measured by the relative position sensor disposed on the measurement apparatus.

    [0026] In a specific implementation process, as shown in Figs. 2 and 3, the second measurement data includes:

    a component

    of an acceleration of the measured object in an X axis, and a component

    of the acceleration of the measured object in a Y axis, wherein

    and

    are noise-containing data measured by the second auxiliary sensor (e.g., a gyroscope and/or an accelerometer) disposed on the measured object;

    a forward speed

    of the measurement apparatus, and a rotation speed

    of the measurement apparatus, which are noise-containing data measured by the first auxiliary sensor (e.g., a coded disc) disposed on the measurement apparatus; and

    an angle θs between a positive direction of the measurement apparatus and the Earth's magnetic north pole, and an angle θt between a positive direction of the measured object and the Earth's magnetic north pole, where θs is data measured by the first auxiliary sensor (e.g., an electronic compass) disposed on the measurement apparatus, and θt is data measured by the second auxiliary sensor (e.g., an electronic compass) disposed on the measured object.



    [0027] In a specific implementation process, the measurement apparatus may communicate with a positioning apparatus on the measured object by using a wireless communication technology, so as to acquire the data measured by the second auxiliary sensor. The wireless communication technology may be an Ultra Wideband (UWB) technology, a Wireless Fidelity (WiFi) technology, a Bluetooth technology or the like.

    [0028] In Step S102, an EKF is constructed on the basis of the first measurement data and the second measurement data.

    [0029] The EKF (Extended Kalman Filter) is a nonlinear system optimal state estimator, which can obtain a measurement result having accuracy higher than that of a single sensor by combining measurement data of a plurality of sensors which are linear or nonlinear and inaccurate. Essentially, the EKF is a recursive least square method.

    [0030] In the embodiment of the disclosure, when the relative position sensor measures the relative position between the measured object and the measurement apparatus, the first measurement data and the second measurement data are combined by using the EKF to correct the first measurement data, so as to increase the accuracy, precision and dynamic performance of the relative position sensor.

    [0031] As an optional implementation, before Step S102, the method further includes the steps as follows.

    [0032] A system consisting of the measurement apparatus and the measured object is mathematically modelled to obtain a mathematical model:

    where

    is a bias of the acceleration of the measured object in the X axis, and

    is a bias of the acceleration of the measured object in the Y axis.

    is a ratio of a forward speed, measured by the first auxiliary sensor (e.g., a coded disc or a light flow sensor), of the measurement apparatus to an actual forward speed of the measurement apparatus. And

    is a ratio of a rotation speed, measured by the first auxiliary sensor (e.g., the coded disc or the light flow sensor), of the measurement apparatus to an actual rotation speed of the measurement apparatus. Here, generally,

    and

    are not equal to 1, are correlated to a wheel diameter and interval of the measurement apparatus, and can reflect the change in the geometric size of the measurement apparatus.

    is a component of a speed, along a pointing direction of the second auxiliary sensor (e.g., an IMU disposed on the measured object), of the measured object in the X axis. And

    is a component of the speed, along the pointing direction of the second auxiliary sensor (e.g., the IMU disposed on the measured object), of the measured object in the Y axis (

    coincides with

    in Fig. 2, and

    coincides with

    in Fig. 2).

    is a magnitude of a projection of a forward speed vector of the measurement apparatus onto a direction of a radius vector of the measurement apparatus and the measured object, and

    is a magnitude of a projection of the forward speed vector of the measurement apparatus onto a direction normal to the radius vector direction of the measurement apparatus and the measured object.

    is a magnitude of a projection of a speed vector of the measured object onto a direction of a radius vector connecting the measurement apparatus and the measured object, and

    is a magnitude of a projection of the speed vector of the measured object onto a direction normal to the radius vector direction connecting the measurement apparatus and the measured object.

    [0033] T(•) is a rotation transformation matrix in a two-dimensional space, which represents that a vector post-multiplied the rotation transformation matrix is rotated anticlockwise by (•).

    [0034] In a specific implementation process, the EKF may be substantially implemented by two steps as follows.

    [0035] The first step is prediction.

    [0036] An established model based on a system difference equation/differential equation is equivalent to 'simulation' of a corresponding system in a real world. If a series of input quantities (or called as 'driving forces') of this simulation system is given, the state of the simulation system will continuously change. Since the simulation system has perfectly modelled an actual system and the driving forces 'drive' the actual system to run while 'driving' the state of the simulation system to be updated, the state change of the model and the state change of the actual system are almost simultaneous and identical. Therefore, before the state of the actual system is collected by using various sensors (i.e., the relative position sensor, the first auxiliary sensor and the second auxiliary sensor), the system state may have been acquired from the simulation system, which is called as 'state prediction'. Conventionally, a system state equation is linear. When the EKF is applied to nonlinear system state prediction, first-order Taylor expansion is performed on a nonlinear system at a current state to obtain an approximate system state update differential equation.

    [0037] The second step is correction.

    [0038] Under ideal conditions, state feedback control may be supposed to be directly performed using a prediction value in the first step. However, two problems may exist in reality. The first one is a problem of initial value indeterminacy, and the second one is a problem of modelling inaccuracy. The two problems will make the state of the simulation system deviate from an actual state gradually. In order to solve the problems, in this embodiment, the EKF continuously corrects the state of the simulation system by using a difference value between an actual system output collected by the relative position sensor and an output of the simulation system. Since the data collected by the relative position sensor has errors, the magnitude of the correction meets a criterion of making it optimally estimated after calculation. That is, a variance of the noise caused by the noise in the magnitude of the correction in the output of the simulation system is minimized.

    [0039] From the above two-step algorithm, it can be seen that the EKF is an algorithm which depends on system modelling and indirectly estimates the system state using the second measurement data. Due to the existence of the system model and an optimal estimation method, compared with a method for filtering data of a relative position sensor using a Finite Impulse Response (FIR) filter, an Infinite Impulse Response (IIR) filter and the like, the EKF has the advantages of real-time response and high accuracy, and may further increase the accuracy by combining data of a plurality of sensors.

    [0040] As an optional implementation, the step of constructing the extended Kalman filter on the basis of the first measurement data and the second measurement data includes the steps as follows.

    [0041] Firstly, a state variable

    is constructed.

    [0042] Here, a real speed of a measured target (i.e.,

    ) and a real relative position (a real relative angle sα between the measurement apparatus and the measured object, and a real relative distance sρ between the measurement apparatus and the measured object) of the measured target are taken as state variables to be observed. In addition, the time-varying bias of an inertial sensor and the change in the geometric size of the measurement apparatus (e.g., the change in the geometric size of a body of a self-balance car) are taken as variable to be observed. Such complete modelling may greatly increase the accuracy of the measurement.

    [0043] Then, an input variable

    is constructed.

    [0044] Finally, on the basis of the state variable and the input variable, the mathematical model is adjusted to obtain the extended Kalman filter:

    where v is a measurement noise vector fitting Gaussian distribution having a mean value of 0 and a covariance of R, ti is time, = f(x,u,ti) is a differential equation model of the system (i.e., an equation 1 of the mathematical model),

    is an output of the system, and h is a measurement matrix:



    [0045] In Step S103, the first measurement data is corrected by using the extended Kalman filter.

    [0046] As an optional implementation, Step S103 includes the steps as follows.

    [0047] Firstly, the extended Kalman filter is initialized.

    [0048] Then, the first measurement data is corrected by performing a recursive algorithm, so as to obtain optimal state estimations for

    and



    [0049] As an optional implementation, the extended Kalman filter may be initialized on the basis of the following equations:



    where (0) is an estimation value of an initial state x(0) of the system, P(0) is a transition probability matrix of the state of the system, and E[•] represents an expectation of •.

    [0050] As an optional implementation, the first measurement data may be corrected by performing the recursive algorithm on the basis of the following equation:



    [0051] In the above equation, the first item f(x̂,y,ti) part represents 'prediction'. That is, the state of the simulation system is updated by taking the equation (1) as sensor data of a 'driving force' part of the simulation system. The second item K[ym - h(x, v0,ti)] represents 'correction'. K represents a correction amplitude by which the state of a simulation system is corrected according to an error between the simulation system and an actual system. This correction needs to meet an optimal estimation criterion, and K needs to be continuously updated in accordance with the following equations:





    where

    is a derivative of an estimation value of a variable of the state of the system; K a correction amplitude by which the state of a simulation system is corrected according to an error between the simulation system and an actual system; ym is a measurement result containing noise, including the relative distance

    and the relative angle

    output by the relative position sensor; is an intermediate variable; P is a propagation probability matrix; and is a derivative of the propagation probability matrix.







    [0052] The above first-order partial derivative matrix is a matrix for local linearization of a system under a current state, where A is a partial derivative matrix of a state transition function f with respect to a state variable x, C is a partial derivative matrix of a measurement matrix h with respect to a state variable x, and M is a partial derivative matrix of the measurement matrix h with respect to a measurement noise v.

    [0053] In the above equation, ym is a measurement result containing noise, including a relative distance and a relative angle output by the relative position sensor. Similar to a traditional state observer, the EKF updates the state by using an input quantity u of a state space equation, and continuously corrects the state variable of the EKF by using offset between an estimation output result and an actual output. Intuitively, compared with a quickly jumping measurement noise, the 'rate' of a correction state is low enough, and compared with a real state change, the 'rate' is high enough. Therefore, a real value may be extracted, almost without delay, from measurement data polluted by the noise.

    [0054] The technical solution in the embodiment of the disclosure at least has the technical effects or advantages as follows.

    [0055] In the embodiment of the disclosure, a method for improving performance of a relative position sensor is disclosed, which is applied to a measurement apparatus. The measurement apparatus is configured to measure a relative position between a measured object and the measurement apparatus. A relative position sensor and a first auxiliary sensor are disposed on the measurement apparatus. A second auxiliary sensor is disposed on the measured object. The method includes the steps as follows. First measurement data measured by a relative position sensor is acquired, and second measurement data measured by a first auxiliary sensor and a second auxiliary sensor is acquired. An extended Kalman filter is constructed on the basis of the first measurement data and the second measurement data. The first measurement data is corrected by using the extended Kalman filter. Since the first measurement data measured by the relative position sensor is corrected by using the extended Kalman filter, the technical problems of inaccurate measurement data and poor dynamic performance caused by the fact that measurement data of the relative position sensor is likely to fluctuate due to interference in the conventional art are effectively solved, thereby achieving the technical effect of increasing the accuracy of measurement data of the relative position sensor so as to improve dynamic performance of the relative position sensor.

    Embodiment 2



    [0056] On the basis of the same inventive concept, another embodiment of the disclosure provides a measurement apparatus implementing the method for improving performance of a relative position sensor in the embodiment of the disclosure.

    [0057] As shown in Fig. 4, a measurement apparatus is configured to measure a relative position between a measured object and the measurement apparatus. A relative position sensor and a first auxiliary sensor are disposed on the measurement apparatus. A second auxiliary sensor is disposed on the measured object. The measurement apparatus includes an acquisition unit 401, a construction unit 402 and a correction unit 403.

    [0058] The acquisition unit 401 is configured to acquire first measurement data measured by a relative position sensor, and acquire second measurement data measured by a first auxiliary sensor and a second auxiliary sensor.

    [0059] The construction unit 402 is configured to construct an extended Kalman filter on the basis of the first measurement data and the second measurement data.

    [0060] The correction unit 403 is configured to correct the first measurement data by using the extended Kalman filter.

    [0061] As an optional implementation, the first measurement data includes:
    a relative angle

    between the measurement apparatus and the measured object, and a relative distance

    between the measurement apparatus and the measured object.

    [0062] As an optional implementation, the second measurement data includes:

    a component

    of an acceleration of the measured object in an X axis, and a component

    of the acceleration of the measured object in a Y axis, here,

    and

    are noise-containing data measured by the second auxiliary sensor (e.g., a gyroscope and/or an accelerometer) disposed on the measured object;

    a forward speed

    of the measurement apparatus, and a rotation speed

    of the measurement apparatus, which are noise-containing data measured by the first auxiliary sensor (e.g., a coded disc) disposed on the measurement apparatus; and

    an angle θs between a positive direction of the measurement apparatus and the Earth's magnetic north pole, and an angle θt between a positive direction of the measured object and the Earth's magnetic north pole, where θs is data measured by the first auxiliary sensor (e.g., an electronic compass) disposed on the measurement apparatus, and θt is data measured by the second auxiliary sensor (e.g., an electronic compass) disposed on the measured object.



    [0063] As an optional implementation, the measurement apparatus further includes a modelling unit.

    [0064] The modelling unit is configured to mathematically model, before the EKF is constructed on the basis of the first measurement data and the second measurement data, a system consisting of the measurement apparatus and the measured object to obtain a mathematical model:

    where

    is a bias of the acceleration of the measured object in the X axis, and

    is a bias of the acceleration of the measured object in the Y axis.

    is a ratio of a forward speed, measured by the first auxiliary sensor (e.g., a coded disc or a light flow sensor), of the measurement apparatus to an actual forward speed of the measurement apparatus, and

    is a ratio of a rotation speed, measured by the first auxiliary sensor (e.g., the coded disc or the light flow sensor), of the measurement apparatus to an actual rotation speed of the measurement apparatus. Here, generally,

    and

    are not equal to 1, are correlated to a wheel diameter and interval of the measurement apparatus, and may reflect the change in the geometric size of the measurement apparatus.

    is a component of a speed, along a pointing direction of the second auxiliary sensor (e.g., an IMU disposed on the measured object), of the measured object in the X axis, and

    is a component of the speed, along the pointing direction of the second auxiliary sensor (e.g., the IMU disposed on the measured object), of the measured object in the Y axis (

    coincides with

    in Fig. 2, and

    coincides with

    in Fig. 2).

    is a magnitude of a projection of a forward speed vector of the measurement apparatus onto a direction of a radius vector of the measurement apparatus and the measured object, and

    is a magnitude of a projection of the forward speed vector of the measurement apparatus onto a direction normal to the radius vector direction of the measurement apparatus and the measured object.

    is a magnitude of a projection of a speed vector of the measured object onto a direction of a radius vector connecting the measurement apparatus and the measured object, and

    is a magnitude of a projection of the speed vector of the measured object onto a direction normal to the radius vector direction connecting the measurement apparatus and the measured object.

    [0065] T(•) is a rotation transformation matrix in a two-dimensional space, which represents that a vector post-multiplied the rotation transformation matrix is rotated anticlockwise by (•).

    [0066] As an optional implementation, the construction unit 402 is specifically configured to:

    construct a state variable

    construct an input variable

    and

    adjust, on the basis of the state variable and the input variable, the mathematical model to the following expression, so as to obtain the EKF:

    where v is a measurement noise vector fitting Gaussian distribution having a mean value of 0 and a covariance of R, ti is time, = f(x,u,ti) is a differential equation model of the system (i.e., an equation 1 of the mathematical model),

    is an output of the system, and h is a measurement matrix:



    [0067] As an optional implementation, the correction unit 403 is specifically configured to:

    [0068] initialize the extended Kalman filter; and correct the first measurement data by executing a recursive algorithm, so as to obtain optimal state estimations for

    and



    [0069] As an optional implementation, the correction unit 403 is specifically configured to initialize the extended Kalman filter on the basis of the following equations:



    where (0) is an estimation value of an initial state x(0) of the system, P(0) is a transition probability matrix of the state of the system, and E[•] represents an expectation of •.

    [0070] As an optional implementation, the correction unit 403 is specifically configured to correct the first measurement data by executing the recursive algorithm on the basis of the following equation:



    [0071] In the above equation, the first item f(x̂,y,ti) part represents 'prediction'. That is, the state of the simulation system is updated by taking the equation (1) as sensor data of a 'driving force' part of the simulation system. The second item K[ym - h(x,v0,ti)] represents 'correction'. K represents a correction amplitude by which the state of a simulation system is corrected according to an error between the simulation system and an actual system. This correction needs to meet an optimal estimation criterion, and K needs to be continuously updated in accordance with the following equations:





    where

    is a derivative of an estimation value of a variable of the state of the system; K is a correction amplitude by which the state of a simulation system is corrected according to an error between the simulation system and an actual system; ym is a measurement result containing noise, including the relative distance

    and the relative angle

    output by the relative position sensor; is an intermediate variable; P is a propagation probability matrix; and is a derivative of the propagation probability matrix.







    [0072] The above first-order partial derivative matrix is a matrix for local linearization of a system under a current state, where A is a partial derivative matrix of a state transition function f with respect to a state variable x, C is a partial derivative matrix of a measurement matrix h with respect to a state variable x, and M is a partial derivative matrix of the measurement matrix h with respect to a measurement noise v.

    [0073] Since the measurement apparatus introduced in this embodiment is an apparatus adopted for implementing the method for improving performance of a relative position sensor in the embodiment of the disclosure, on the basis of the method for improving performance of a relative position sensor introduced in the embodiment of the disclosure, those skilled in the art can know a specific implementation and various variation of the measurement apparatus of this embodiment. Thus, how to implement the method in the embodiment of the disclosure by the measurement apparatus will be no longer introduced herein in detail.

    [0074] The technical solution in the embodiment of the disclosure at least has the technical effects or advantages as follows.

    [0075] In the embodiment of the disclosure, a measurement apparatus is disclosed. The measurement apparatus is configured to measure a relative position between a measured object and the measurement apparatus. A relative position sensor and a first auxiliary sensor are disposed on the measurement apparatus. A second auxiliary sensor is disposed on the measured object. The measurement apparatus includes: an acquisition unit configured to acquire first measurement data measured by a relative position sensor, and acquire second measurement data measured by a first auxiliary sensor and a second auxiliary sensor; a construction unit configured to construct an EKF on the basis of the first measurement data and the second measurement data; and a correction unit configured to correct the first measurement data by using the EKF. Since the first measurement data measured by the relative position sensor is corrected by using the extended Kalman filter, the technical problems of inaccurate measurement data and poor dynamic performance caused by the fact that measurement data of the relative position sensor is likely to fluctuate due to interference in the conventional art are effectively solved, thereby achieving the technical effect of increasing the accuracy of measurement data of the relative position sensor so as to improve dynamic performance of the relative position sensor.

    [0076] Another embodiment of the disclosure provides a computer storage medium storing computer-executable instructions configured to:

    acquire first measurement data measured by a relative position sensor, and acquire second measurement data measured by a first auxiliary sensor and a second auxiliary sensor;

    construct an EKF on the basis of the first measurement data and the second measurement data; and

    correct the first measurement data by using the EKF.



    [0077] Furthermore, processing executed by the computer storage medium is the same as processing in the embodiment 1, which will not be elaborated herein.

    [0078] Those skilled in the art shall understand that the embodiments of the disclosure may be provided as a method, a system or a computer program product. Thus, forms of hardware embodiments, software embodiments or embodiments integrating software and hardware may be adopted in the disclosure. Moreover, a form of the computer program product implemented on one or more computer available storage media (including, but are not limited to, a disk memory, a Compact Disc Read-Only Memory (CD-ROM) an optical memory and the like) containing computer available program codes may be adopted in the disclosure.

    [0079] The disclosure is described with reference to flow charts and/or block diagrams of the method, the device (system) and the computer program product according to the embodiments of the disclosure. It will be understood that each flow and/or block in the flow charts and/or the block diagrams and a combination of the flows and/or the blocks in the flow charts and/or the block diagrams may be implemented by computer program instructions. These computer program instructions may be provided for a general computer, a dedicated computer, an embedded processor or processors of other programmable data processing devices to generate a machine, such that an apparatus for implementing functions designated in one or more flows of the flow charts and/or one or more blocks of the block diagrams is generated via instructions executed by the computers or the processors of the other programmable data processing devices.

    [0080] These computer program instructions may also be stored in a computer readable memory capable of guiding the computers or the other programmable data processing devices to work in a specific manner, such that a manufactured product including an instruction apparatus is generated via the instructions stored in the computer readable memory, and the instruction apparatus implements the functions designated in one or more flows of the flow charts and/or one or more blocks of the block diagrams.

    [0081] These computer program instructions may also be loaded to the computers or the other programmable data processing devices, such that processing implemented by the computers is generated by executing a series of operation steps on the computers or the other programmable devices, and therefore the instructions executed on the computers or the other programmable devices provide a step of implementing the functions designated in one or more flows of the flow charts and/or one or more blocks of the block diagrams.

    [0082] Although the preferred embodiments of the disclosures have been described, once those skilled in the art obtains a basic creativity concept, those skilled in the art may change and modify these embodiments additionally.

    INDUSTRIAL APPLICABILITY



    [0083] The embodiment of the disclosure provides a method for improving performance of a relative position sensor. The method includes the steps as follows. First measurement data measured by a relative position sensor is acquired, and second measurement data measured by a first auxiliary sensor and a second auxiliary sensor is acquired. An EKF is constructed on the basis of the first measurement data and the second measurement data. The first measurement data is corrected by using the EKF. Since the first measurement data measured by the relative position sensor is corrected by using the EKF, the technical problems of inaccurate measurement data and poor dynamic performance caused by the fact that measurement data of the relative position sensor is likely to fluctuate due to interference in the conventional art are effectively solved, thereby achieving the technical effect of increasing the accuracy of measurement data of the relative position sensor so as to improve dynamic performance of the relative position sensor.


    Claims

    1. A method for improving performance of a relative position sensor, the method being applied to a measurement apparatus configured to measure a relative position between a measured object and the measurement apparatus, the relative position sensor and a first auxiliary sensor being disposed on the measurement apparatus, and a second auxiliary sensor being disposed on the measured object, the method comprising:

    acquiring (S101) first measurement data measured by the relative position sensor, acquiring second measurement data measured by the first auxiliary sensor, and acquiring third measurement data measured by the second auxiliary sensor;

    constructing (S102) an Extended Kalman Filter (EKF) on the basis of the first measurement data, the second measurement data, and the third measurement data; and

    correcting (S103) the first measurement data by using the EKF;

    wherein the first measurement data comprises: a relative angle

    between the measurement apparatus and the measured object, and a relative distance

    between the measurement apparatus and the measured object; characterized in that

    the second measurement data comprises: a forward speed

    of the measurement apparatus, a rotation speed

    of the measurement apparatus, and an angle θs between a positive direction of the measurement apparatus and a magnetic north pole of Earth;

    the third measurement data comprises: a component

    of an acceleration of the measured object in an X axis, a component

    of the acceleration of the measured object in a Y axis, and an angle θt between a positive direction of the measured object and the magnetic north pole of Earth; and

    before the EKF is constructed on the basis of the first measurement data, the second measurement data, and the third measurement data, the method further comprises:
    mathematically modelling a system consisting of the measurement apparatus and the measured object to obtain a mathematical model:

    where

    is a bias of the acceleration of the measured object in the X axis, and

    is a bias of the acceleration of the measured object in the Y axis;



    is a ratio of a forward speed, measured by the first auxiliary sensor, of the measurement apparatus to an actual forward speed of the measurement apparatus, and

    is a ratio of a rotation speed, measured by the first auxiliary sensor, of the measurement apparatus to an actual rotation speed of the measurement apparatus;



    is a component of a speed, along a pointing direction of the second auxiliary sensor, of the measured object in the X axis, and

    is a component of the speed, along the pointing direction of the second auxiliary sensor, of the measured object in the Y axis;



    is a magnitude of a projection of a forward speed vector of the measurement apparatus onto a direction of a radius vector of the measurement apparatus and the measured object, and

    is a magnitude of a projection of the forward speed vector of the measurement apparatus onto a direction normal to the direction of the radius vector of the measurement apparatus and the measured object;



    is a magnitude of a projection of a speed vector of the measured object onto a direction of a radius vector connecting the measurement apparatus and the measured object, and

    is a magnitude of a projection of the speed vector of the measured object onto a direction normal to the direction of the radius vector connecting the measurement apparatus and the measured object; and

    T(•) is a rotation transformation matrix in a two-dimensional space.


     
    2. The method for improving performance of a relative position sensor according to claim 1, wherein constructing the EKF on the basis of the first measurement data, the second measurement data, and the third measurement data comprises:

    constructing a state variable

    constructing an input variable

    and

    adjusting, on the basis of the state variable and the input variable, the mathematical model to obtain the EKF:

    where v is a measurement noise vector fitting Gaussian distribution having a mean value of 0 and a covariance of R, ti is time, = f(x,u,ti) is a differential equation model of the system,

    is an output of the system, and h is a measurement matrix.


     
    3. The method for improving performance of a relative position sensor according to claim 2, wherein correcting the first measurement data by using the EKF comprises:

    initializing the EKF; and

    correcting the first measurement data by executing a recursive algorithm, so as to obtain optimal state estimations for

    and


     
    4. The method for improving performance of a relative position sensor according to claim 3, wherein initializing the EKF comprises:
    initializing the EKF on the basis of the following equations:



    where (0) is an estimation value of an initial state x(0) of the system, P(0) is a transition probability matrix of a state of the system, and E[•] represents an expectation of •,

    wherein correcting the first measurement data by executing the recursive algorithm comprises:
    correcting the first measurement data by executing the recursive algorithm on the basis of the following equation:

    where K is continuously updated in accordance with the following equations:





    where







    is a derivative of an estimation value of a variable of the state of the system; K is a correction amplitude by which a state of a simulation system is corrected according to an error between the simulation system and an actual system; ym is a measurement result containing noise, comprising the relative distance

    and the relative angle

    output by the relative position sensor; is an intermediate variable; P is a propagation probability matrix; and is a derivative of the propagation probability matrix.


     
    5. A measurement apparatus configured to measure a relative position between a measured object and the measurement apparatus, a relative position sensor and a first auxiliary sensor being disposed on the measurement apparatus, and a second auxiliary sensor being disposed on the measured object, the measurement apparatus comprising:

    an acquisition unit (401) configured to acquire first measurement data measured by the relative position sensor, acquire second measurement data measured by the first auxiliary sensor, and acquire third measurement data measured by the second auxiliary sensor;

    a construction unit (402) configured to construct an Extended Kalman Filter (EKF) on the basis of the first measurement data, the second measurement data, and the third measurement data; and

    a correction unit (403) configured to correct the first measurement data by using the EKF; characterized in that

    the first measurement data comprises: a relative angle

    between the measurement apparatus and the measured object, and a relative distance

    between the measurement apparatus and the measured object;

    the second measurement data comprises: a forward speed

    of the measurement apparatus, and a rotation speed

    of the measurement apparatus, and an angle θs between a positive direction of the measurement apparatus and a magnetic north pole of Earth;

    the third measurement data comprises: a component

    of an acceleration of the measured object in an X axis, a component

    of the acceleration of the measured object in a Y axis, and an angle θt between a positive direction of the measured object and the magnetic north pole of Earth; and

    in that the measurement apparatus further comrpises: a modelling unit configured to mathematically model, before the EKF is constructed on the basis of the first measurement data and the second measurement data, a system consisting of the measurement apparatus and the measured object to obtain a mathematical model:

    where

    is a bias of the acceleration of the measured object in the X axis, and

    is a bias of the acceleration of the measured object in the Y axis;



    is a ratio of a forward speed, measured by the first auxiliary sensor, of the measurement apparatus to an actual forward speed of the measurement apparatus, and

    is a ratio of a rotation speed, measured by the first auxiliary sensor, of the measurement apparatus to an actual rotation speed of the measurement apparatus;



    is a component of a speed, along a pointing direction of the second auxiliary sensor, of the measured object in the X axis, and

    is a component of the speed, along the pointing direction of the second auxiliary sensor, of the measured object in the Y axis;



    is a magnitude of a projection of a forward speed vector of the measurement apparatus onto a direction of a radius vector of the measurement apparatus and the measured object, and

    is a magnitude of a projection of the forward speed vector of the measurement apparatus onto a direction normal to the direction of the radius vector of the measurement apparatus and the measured object;



    is a magnitude of a projection of a speed vector of the measured object onto a direction of a radius vector connecting the measurement apparatus and the measured object, and

    is a magnitude of a projection of the speed vector of the measured object onto a direction normal to the direction of the radius vector connecting the measurement apparatus and the measured object; and

    T(•) is a rotation transformation matrix in a two-dimensional space.


     
    6. The measurement apparatus according to claim 5, wherein the construction unit is configured to:

    construct a state variable

    construct an input variable

    and

    adjust, on the basis of the state variable and the input variable, the mathematical model to obtain the EKF:

    where v is a measurement noise vector fitting Gaussian distribution having a mean value of 0 and a covariance of R, ti is time, = f(x,u,ti) is a differential equation model of the system,

    is an output of the system, and h is a measurement matrix,
    wherein the correction unit is configured to:
    correct the first measurement data by executing a recursive algorithm, so as to obtain optimal state estimations for

    and


     
    7. The measurement apparatus according to claim 6, wherein the correction unit is configured to initialize the EKF on the basis of the following equations:



    where (0) is an estimation value of an initial state x(0) of the system, P(0) is a transition probability matrix of a state of the system, and E[•] represents an expectation of •.
     
    8. The measurement apparatus according to claim 7, wherein the correction unit is configured to correct the first measurement data by executing the recursive algorithm on the basis of the following equation:

    where Kis continuously updated in accordance with the following equations:





    where








    is a derivative of an estimation value of a variable of the state of the system; Kis a correction amplitude by which a state of a simulation system is corrected according to an error between the simulation system and an actual system; ym is a measurement result containing noise, comprising the relative distance

    and the relative angle

    output by the relative position sensor; is an intermediate variable; P is a propagation probability matrix; and is a derivative of the propagation probability matrix.
     
    9. A computer storage medium storing computer-executable instructions that when executed by a processor, causes the processor to execute the method according to any one of claims 1-4.
     


    Ansprüche

    1. Verfahren zur Verbesserung der Leistung eines Relativpositionssensors, wobei das Verfahren auf eine Messvorrichtung angewendet wird, die dazu konfiguriert ist, eine Relativposition zwischen einem gemessenen Objekt und der Messvorrichtung zu messen, wobei der Relativpositionssensors und ein erster Hilfssensor an der Messvorrichtung angeordnet sind, und wobei ein zweiter Hilfssensor am gemessenen Objekt angeordnet ist, wobei das Verfahren umfasst:

    Erfassen (S101) von ersten Messdaten, die vom Relativpositionssensor gemessen werden, Erfassen von zweiten Messdaten, die vom ersten Hilfssensor gemessen werden, und Erfassen von dritten Messdaten, die vom zweiten Hilfssensor gemessen werden;

    Entwerfen (S102) eines erweiterten Kalman-Filters (EKF) auf der Basis der ersten Messdaten, der zweiten Messdaten und der dritten Messdaten; und

    Korrigieren (S103) der ersten Messdaten unter Verwendung des EKF;

    wobei die ersten Messdaten umfassen: einen Relativwinkel

    zwischen der Messvorrichtung und dem gemessenen Objekt, und einen Relativabstand

    zwischen der Messvorrichtung und dem gemessenen Objekt; dadurch gekennzeichnet, dass

    die zweiten Messdaten umfassen: eine Vorwärtsgeschwindigkeit

    der Messvorrichtung, eine Drehgeschwindigkeit

    der Messvorrichtung, und einen Winkel θs, zwischen einer positiven Richtung der Messvorrichtung und einem magnetischen Nordpol der Erde;

    die dritten Messdaten umfassen: eine Komponente

    einer Beschleunigung des gemessenen Objekts auf einer X-Achse, eine Komponente

    der Beschleunigung des gemessenen Objekts auf einer Y-Achse, und einen Winkel θi zwischen einer positiven Richtung des gemessenen Objekts und dem magnetischen Nordpol der Erde; und

    das Verfahren, bevor das EKF auf der Basis der ersten Messdaten, der zweiten Messdaten und der dritten Messdaten entworfen wird, weiter umfasst:
    mathematisches Modellieren eines Systems, das aus der Messvorrichtung und dem gemessenen Objekt besteht, um ein mathematisches Modell zu erhalten:

    wobei

    ein Bias der Beschleunigung des gemessenen Objekts auf der X-Achse ist, und

    ein Bias der Beschleunigung des gemessenen Objekts auf der Y-Achse ist;



    ein Verhältnis einer vom ersten Hilfssensor gemessenen Vorwärtsgeschwindigkeit der Messvorrichtung zu einer tatsächlichen Vorwärtsgeschwindigkeit der Messvorrichtung ist, und

    ein Verhältnis einer vom ersten Hilfssensor gemessenen Drehgeschwindigkeit der Messvorrichtung zu einer tatsächlichen Drehgeschwindigkeit der Messvorrichtung ist;



    eine Komponente einer Geschwindigkeit des gemessenen Objekts entlang einer Zeigerichtung des zweiten Hilfssensors auf der X-Achse ist, und

    eine Komponente der Geschwindigkeit des gemessenen Objekts entlang der Zeigerichtung des zweiten Hilfssensors auf der Y-Achse ist;



    eine Größe einer Projektion eines Vorwärtsgeschwindigkeitsvektors der Messvorrichtung auf eine Richtung eines Radiusvektors der Messvorrichtung und des gemessenen Objekts ist, und

    eine Größe einer Projektion des Vorwärtsgeschwindigkeitsvektors der Messvorrichtung auf eine Richtung normal zur Richtung des Radiusvektors der Messvorrichtung und des gemessenen Objekts ist;



    eine Größe einer Projektion eines Geschwindigkeitsvektors des gemessenen Objekts auf eine Richtung eines Radiusvektors ist, der die Messvorrichtung und das gemessene Objekt verbindet, und

    eine Größe einer Projektion des Geschwindigkeitsvektors des gemessenen Objekts auf eine Richtung normal zur Richtung des Radiusvektors ist, der die Messvorrichtung und das gemessene Objekt verbindet; und

    T(•) eine Dreh-Transformationsmatrix in einem zweidimensionalen Raum ist.


     
    2. Verfahren zur Verbesserung der Leistung eines Relativpositionssensors nach Anspruch 1, wobei das Entwerfen des EKF auf der Basis der ersten Messdaten, der zweiten Messdaten und der dritten Messdaten umfasst:

    Entwerfen einer Zustandsvariablen

    Entwerfen einer Eingangsvariablen

    und

    Anpassen des mathematischen Modells auf der Basis der Zustandsvariablen und der Eingangsvariablen, um das EKF zu erhalten:

    wobei v ein der Gaußschen Verteilung entsprechender Messrauschvektor ist, der einen Mittelwert von 0 und eine Kovarianz von R aufweist, ti die Zeit ist, = f(x,u,ti) ein Differentialgleichungsmodell des Systems ist,

    ein Ausgang des Systems ist, und h eine Messmatrix ist.


     
    3. Verfahren zur Verbesserung der Leistung eines Relativpositionssensors nach Anspruch 2, wobei das Korrigieren der ersten Messdaten unter Verwendung des EKF umfasst:

    Initialisieren des EKF; und

    Korrigieren der ersten Messdaten durch Ausführen eines rekursiven Algorithmus, um optimale Zustandsschätzungen für

    und

    zu erhalten.


     
    4. Verfahren zur Verbesserung der Leistung eines Relativpositionssensors nach Anspruch 3, wobei das Initialisieren des EKF umfasst:
    Initialisieren des EKF auf der Basis der folgenden Gleichungen:



    wobei (0) ein Schätzwert eines anfänglichen Zustands x(0) des Systems ist, P(0) eine Übergangswahrscheinlichkeitsmatrix eines Zustands des Systems ist, und E[•] eine Erwartung von • darstellt,

    wobei das Korrigieren der ersten Messdaten durch Ausführen des rekursiven Algorithmus umfasst:
    Korrigieren der ersten Messdaten durch Ausführen des rekursiven Algorithmus auf der Basis der folgenden Gleichung:

    wobei K kontinuierlich entsprechend den folgenden Gleichungen aktualisiert wird:





    wobei







    eine Ableitung eines Schätzwerts einer Variablen des Zustands des Systems ist; K eine Korrekturamplitude ist, um die ein Zustand eines Simulationssystems entsprechend einem Fehler zwischen dem Simulationssystem und einem tatsächlichen System korrigiert wird; ym ein Rauschen enthaltendes Messergebnis ist, das den Relativabstand

    und den Relativwinkel

    umfasst, die vom Relativpositionssensor ausgegeben werden; eine Zwischenvariable ist; P eine Ausbreitungswahrscheinlichkeitsmatrix ist; und eine Ableitung der Ausbreitungswahrscheinlichkeitsmatrix ist.


     
    5. Messvorrichtung, die dazu konfiguriert ist, eine Relativposition zwischen einem gemessenen Objekt und der Messvorrichtung zu messen, wobei ein Relativpositionssensors und ein erster Hilfssensor an der Messvorrichtung angeordnet sind, und wobei ein zweiter Hilfssensor am gemessenen Objekt angeordnet ist, wobei die Messvorrichtung umfasst:

    eine Erfassungseinheit (401), die dazu konfiguriert ist, erste Messdaten zu erfassen, die vom Relativpositionssensor gemessen werden, zweite Messdaten zu erfassen, die vom ersten Hilfssensor gemessen werden, und dritte Messdaten zu erfassen, die vom zweiten Hilfssensor gemessen werden;

    eine Entwurfseinheit (402), die dazu konfiguriert ist, ein erweitertes Kalman-Filter (EKF) auf der Basis der ersten Messdaten, der zweiten Messdaten und der dritten Messdaten zu entwerfen; und

    eine Korrektureinheit (403), die dazu konfiguriert ist, die ersten Messdaten unter Verwendung des EKF zu korrigieren; dadurch gekennzeichnet, dass

    die ersten Messdaten umfassen: einen Relativwinkel

    zwischen der Messvorrichtung und dem gemessenen Objekt, und einen Relativabstand

    zwischen der Messvorrichtung und dem gemessenen Objekt;

    die zweiten Messdaten umfassen: eine Vorwärtsgeschwindigkeit

    der Messvorrichtung, und eine Drehgeschwindigkeit

    der Messvorrichtung, und einen Winkel θs zwischen einer positiven Richtung der Messvorrichtung und einem magnetischen Nordpol der Erde;

    die dritten Messdaten umfassen: eine Komponente

    einer Beschleunigung des gemessenen Objekts auf einer X-Achse, eine Komponente

    der Beschleunigung des gemessenen Objekts auf einer Y-Achse, und einen Winkel θt zwischen einer positiven Richtung des gemessenen Objekts und dem magnetischen Nordpol der Erde; und

    dadurch, dass die Messvorrichtung weiter umfasst:
    eine Modelliereinheit, die dazu konfiguriert ist, bevor das EKF auf der Basis der ersten Messdaten und der zweiten Messdaten entworfen wird, mathematisch ein System zu modellieren, das aus der Messvorrichtung und dem gemessenen Objekt besteht, um ein mathematisches Modell zu erhalten:

    wobei

    ein Bias der Beschleunigung des gemessenen Objekts auf der X-Achse ist, und

    ein Bias der Beschleunigung des gemessenen Objekts auf der Y-Achse ist;



    ein Verhältnis einer vom ersten Hilfssensor gemessenen Vorwärtsgeschwindigkeit der Messvorrichtung zu einer tatsächlichen Vorwärtsgeschwindigkeit der Messvorrichtung ist, und

    ein Verhältnis einer vom ersten Hilfssensor gemessenen Drehgeschwindigkeit der Messvorrichtung zu einer tatsächlichen Drehgeschwindigkeit der Messvorrichtung ist;



    eine Komponente einer Geschwindigkeit des gemessenen Objekts entlang einer Zeigerichtung des zweiten Hilfssensors auf der X-Achse ist, und

    eine Komponente der Geschwindigkeit des gemessenen Objekts entlang der Zeigerichtung des zweiten Hilfssensors auf der Y-Achse ist;



    eine Größe einer Projektion eines Vorwärtsgeschwindigkeitsvektors der Messvorrichtung auf eine Richtung eines Radiusvektors der Messvorrichtung und des gemessenen Objekts ist, und

    eine Größe einer Projektion des Vorwärtsgeschwindigkeitsvektors der Messvorrichtung auf eine Richtung normal zur Richtung des Radiusvektors der Messvorrichtung und des gemessenen Objekts ist;



    eine Größe einer Projektion eines Geschwindigkeitsvektors des gemessenen Objekts auf eine Richtung eines Radiusvektors ist, der die Messvorrichtung und das gemessene Objekt verbindet, und

    eine Größe einer Projektion des Geschwindigkeitsvektors des gemessenen Objekts auf eine Richtung normal zur Richtung des Radiusvektors ist, der die Messvorrichtung und das gemessene Objekt verbindet; und

    T(•) eine Dreh-Transformationsmatrix in einem zweidimensionalen Raum ist.


     
    6. Messvorrichtung nach Anspruch 5, wobei die Entwurfseinheit dazu konfiguriert ist:

    eine Zustandsvariable

    zu entwerfen;

    eine Eingangsvariable

    zu entwerfen; und

    auf der Basis der Zustandsvariablen und der Eingangsvariablen das mathematische Modell anzupassen, um das EKF zu erhalten:

    wobei v ein der Gaußschen Verteilung entsprechender Messrauschvektor ist, der einen Mittelwert von 0 und eine Kovarianz von R aufweist, ti die Zeit ist, = f(x,u,ti) ein Differentialgleichungsmodell des Systems ist,

    ein Ausgang des Systems ist, und h eine Messmatrix ist,

    wobei die Korrektureinheit dazu konfiguriert ist:
    die ersten Messdaten durch Ausführen eines rekursiven Algorithmus zu korrigieren, um optimale Zustandsschätzungen für

    und

    zu erhalten.


     
    7. Messvorrichtung nach Anspruch 6, wobei die Korrektureinheit dazu konfiguriert ist, das EKF auf der Basis der folgenden Gleichungen zu initialisieren:



    wobei (0) ein Schätzwert eines anfänglichen Zustands x(0) des Systems ist, P(0) eine Übergangswahrscheinlichkeitsmatrix eines Zustands des Systems ist, und E[•] eine Erwartung von • darstellt.
     
    8. Messvorrichtung nach Anspruch 7, wobei die Korrektureinheit dazu konfiguriert ist, die ersten Messdaten durch Ausführen des rekursiven Algorithmus auf der Basis der folgenden Gleichung zu korrigieren:

    wobei K kontinuierlich entsprechend den folgenden Gleichungen aktualisiert wird:





    wobei







    eine Ableitung eines Schätzwerts einer Variablen des Zustands des Systems ist; K eine Korrekturamplitude ist, um die ein Zustand eines Simulationssystems entsprechend einem Fehler zwischen dem Simulationssystem und einem tatsächlichen System korrigiert wird; ym ein Rauschen enthaltendes Messergebnis ist, das den Relativabstand

    und den Relativwinkel

    umfasst, die vom Relativpositionssensor ausgegeben werden; eine Zwischenvariable ist; P eine Ausbreitungswahrscheinlichkeitsmatrix ist; und eine Ableitung der Ausbreitungswahrscheinlichkeitsmatrix ist.
     
    9. Computerspeichermedium, das computerausführbare Anweisungen speichert, das, wenn es von einem Prozessor ausgeführt wird, den Prozessor dazu bringt, das Verfahren nach einem der Ansprüche 1 bis 4 auszuführen.
     


    Revendications

    1. Méthode pour améliorer les performances d'un capteur de position relative, la méthode étant appliquée à un appareil de mesure configuré pour mesurer une position relative entre un objet mesuré et l'appareil de mesure, le capteur de position relative et un premier capteur auxiliaire étant disposés sur l'appareil de mesure, et un second capteur auxiliaire étant disposé sur l'objet mesuré, la méthode comprenant :

    l'acquisition (S101) de premières données de mesure mesurées par le capteur de position relative, l'acquisition de deuxièmes données de mesure mesurées par le premier capteur auxiliaire, et l'acquisition de troisièmes données de mesure mesurées par le second capteur auxiliaire ;

    la construction (S102) d'un filtre de Kalman étendu (EKF) sur la base des premières données de mesure, des deuxièmes données de mesure et des troisièmes données de mesure ; et

    la correction (S103) des premières données de mesure en utilisant l'EKF ;

    dans laquelle les premières données de mesure comprennent : un angle relatif

    entre l'appareil de mesure et l'objet mesuré, et une distance relative

    entre l'appareil de mesure et l'objet mesuré ; caractérisée en ce que

    les deuxièmes données de mesure comprennent : une vitesse d'avancement

    de l'appareil de mesure, une vitesse de rotation

    de l'appareil de mesure, et un angle θs entre une direction positive de l'appareil de mesure et un pôle nord magnétique de la Terre ;

    les troisièmes données de mesure comprennent : une composante

    d'une accélération de l'objet mesuré dans un axe X, une composante

    de l'accélération de l'objet mesuré dans un axe Y, et un angle θt entre une direction positive de l'objet mesuré et le pôle nord magnétique de la Terre ; et

    avant que l'EKF soit construit sur la base des premières données de mesure, des deuxièmes données de mesure et des troisièmes données de mesure, la méthode comprend en outre :
    la modélisation mathématique d'un système consistant en l'appareil de mesure et l'objet mesuré pour obtenir un modèle mathématique :



    est un biais de l'accélération de l'objet mesuré dans l'axe X, et

    est un biais de l'accélération de l'objet mesuré dans l'axe Y ;



    est un rapport d'une vitesse d'avancement, mesurée par le premier capteur auxiliaire, de l'appareil de mesure sur une vitesse d'avancement réelle de l'appareil de mesure, et

    est un rapport d'une vitesse de rotation, mesurée par le premier capteur auxiliaire, de l'appareil de mesure sur une vitesse de rotation réelle de l'appareil de mesure ;



    est une composante d'une vitesse, le long d'une direction de pointage du second capteur auxiliaire, de l'objet mesuré dans l'axe X, et

    est une composante de la vitesse, le long de la direction de pointage du second capteur auxiliaire, de l'objet mesuré dans l'axe Y ;



    est une grandeur d'une projection d'un vecteur de vitesse d'avancement de l'appareil de mesure sur une direction d'un vecteur de rayon de l'appareil de mesure et de l'objet mesuré, et

    est une grandeur d'une projection du vecteur de vitesse d'avancement de l'appareil de mesure sur une direction normale à la direction du vecteur de rayon de l'appareil de mesure et de l'objet mesuré ;



    est une grandeur d'une projection d'un vecteur de vitesse de l'objet mesuré sur une direction d'un vecteur de rayon reliant l'appareil de mesure et l'objet mesuré, et

    est une grandeur d'une projection du vecteur de vitesse de l'objet mesuré sur une direction normale à la direction du vecteur de rayon reliant l'appareil de mesure et l'objet mesuré ; et

    T(•) est une matrice de transformation de rotation dans un espace bidimensionnel.


     
    2. Méthode pour améliorer les performances d'un capteur de position relative selon la revendication 1, dans laquelle la construction de l'EKF sur la base des premières données de mesure, des deuxièmes données de mesure et des troisièmes données de mesure comprend :

    la construction d'une variable d'état

    la construction d'une variable d'entrée

    et

    l'ajustement, sur la base de la variable d'état et de la variable d'entrée, du modèle mathématique pour obtenir l'EKF :

    v est une distribution gaussienne d'adaptation de vecteur de bruit de mesure ayant une valeur moyenne de 0 et une covariance de R, ti est le temps, = f(x,u,ti) est un modèle d'équation différentielle du système, y =

    est une sortie du système et h est une matrice de mesure.


     
    3. Méthode pour améliorer les performances d'un capteur de position relative selon la revendication 2, dans laquelle la correction des premières données de mesure en utilisant l'EKF comprend :

    l'initialisation de l'EKF ; et

    la correction des premières données de mesure par l'exécution d'un algorithme récursif, de façon à obtenir des estimations d'état optimal pour

    et


     
    4. Méthode pour améliorer les performances d'un capteur de position relative selon la revendication 3, dans laquelle l'initialisation de l'EKF comprend :
    l'initialisation de l'EKF sur la base des équations suivantes :



    (0) est une valeur d'estimation d'un état initial x(0) du système, P(0) est une matrice de probabilité de transition d'un état du système, et E[•] représente une espérance de •,
    dans laquelle la correction des premières données de mesure par l'exécution de l'algorithme récursif comprend :
    la correction des premières données de mesure par l'exécution de l'algorithme récursif sur la base de l'équation suivante :

    où K est continuellement mis à jour conformément aux équations suivantes :













    est une dérivée d'une valeur d'estimation d'une variable de l'état du système ; K est une amplitude de correction par laquelle un état d'un système de simulation est corrigé en fonction d'une erreur entre le système de simulation et un système réel ; ym est un résultat de mesure contenant du bruit, comprenant la distance relative

    et l'angle relatif

    délivrés en sortie par le capteur de position relative ; est une variable intermédiaire ; P est une matrice de probabilité de propagation ; et est une dérivée de la matrice de probabilité de propagation.
     
    5. Appareil de mesure configuré pour mesurer une position relative entre un objet mesuré et l'appareil de mesure, un capteur de position relative et un premier capteur auxiliaire étant disposés sur l'appareil de mesure, et un second capteur auxiliaire étant disposé sur l'objet mesuré, l'appareil de mesure comprenant :

    une unité d'acquisition (401) configurée pour acquérir des premières données de mesure mesurées par le capteur de position relative, acquérir des deuxièmes données de mesure mesurées par le premier capteur auxiliaire, et acquérir des troisièmes données de mesure mesurées par le second capteur auxiliaire ;

    une unité de construction (402) configurée pour construire un filtre de Kalman étendu (EKF) sur la base des premières données de mesure, des deuxièmes données de mesure et des troisièmes données de mesure ; et

    une unité de correction (403) configurée pour corriger les premières données de mesure en utilisant l'EKF ; caractérisée en ce que

    les premières données de mesure comprennent : un angle relatif

    entre l'appareil de mesure et l'objet mesuré, et une distance relative

    entre l'appareil de mesure et l'objet mesuré ;

    les deuxièmes données de mesure comprennent : une vitesse d'avancement

    de l'appareil de mesure, une vitesse de rotation

    de l'appareil de mesure, et un angle θs entre une direction positive de l'appareil de mesure et un pôle nord magnétique de la Terre ;

    les troisièmes données de mesure comprennent : une composante

    d'une accélération de l'objet mesuré dans un axe X, une composante

    de l'accélération de l'objet mesuré dans un axe Y, et un angle θt entre une direction positive de l'objet mesuré et le pôle nord magnétique de la Terre ; et

    en ce que l'appareil de mesure comprend en outre : une unité de modélisation configurée pour modéliser mathématiquement, avant que l'EKF soit construit sur la base des premières données de mesure, des deuxièmes données de mesure, un système consistant en l'appareil de mesure et l'objet mesuré pour obtenir un modèle mathématique :



    est un biais de l'accélération de l'objet mesuré dans l'axe X, et

    est un biais de l'accélération de l'objet mesuré dans l'axe Y ;



    est un rapport d'une vitesse d'avancement, mesurée par le premier capteur auxiliaire, de l'appareil de mesure sur une vitesse d'avancement réelle de l'appareil de mesure, et

    est un rapport d'une vitesse de rotation, mesurée par le premier capteur auxiliaire, de l'appareil de mesure sur une vitesse de rotation réelle de l'appareil de mesure ;



    est une composante d'une vitesse, le long d'une direction de pointage du second capteur auxiliaire, de l'objet mesuré dans l'axe X, et

    est une composante de la vitesse, le long de la direction de pointage du second capteur auxiliaire, de l'objet mesuré dans l'axe Y ;



    est une grandeur d'une projection d'un vecteur de vitesse d'avancement de l'appareil de mesure sur une direction d'un vecteur de rayon de l'appareil de mesure et de l'objet mesuré, et

    est une grandeur d'une projection du vecteur de vitesse d'avancement de l'appareil de mesure sur une direction normale à la direction du vecteur de rayon de l'appareil de mesure et de l'objet mesuré ;



    est une grandeur d'une projection d'un vecteur de vitesse de l'objet mesuré sur une direction d'un vecteur de rayon reliant l'appareil de mesure et l'objet mesuré, et

    est une grandeur d'une projection du vecteur de vitesse de l'objet mesuré sur une direction normale à la direction du vecteur de rayon reliant l'appareil de mesure et l'objet mesuré ; et

    T(•) est une matrice de transformation de rotation dans un espace bidimensionnel.


     
    6. Appareil de mesure selon la revendication 5, dans lequel l'unité de construction est configurée pour :

    construire une variable d'état

    construire une variable d'entrée

    et

    ajuster, sur la base de la variable d'état et de la variable d'entrée, le modèle mathématique pour obtenir l'EKF :

    où v est une distribution gaussienne d'adaptation de vecteur de bruit de mesure ayant une valeur moyenne de 0 et une covariance de R, ti est le temps, = f(x,u,ti) est un modèle d'équation différentielle du système, y =

    est une sortie du système et h est une matrice de mesure,

    dans lequel l'unité de correction est configurée pour :
    corriger les premières données de mesure par l'exécution d'un algorithme récursif, de façon à obtenir des estimations d'état optimal pour

    et


     
    7. Appareil de mesure selon la revendication 6, dans lequel l'unité de correction est configurée pour initialiser l'EKF sur la base des équations suivantes :



    (0) est une valeur d'estimation d'un état initial x(0) du système, P(0) est une matrice de probabilité de transition d'un état du système, et E[•] représente une espérance de •.
     
    8. Appareil de mesure selon la revendication 7, dans lequel l'unité de correction est configurée pour corriger les premières données de mesure par l'exécution de l'algorithme récursif sur la base de l'équation suivante :

    où K est continuellement mis à jour conformément aux équations suivantes :













    est une dérivée d'une valeur d'estimation d'une variable de l'état du système ; K est une amplitude de correction par laquelle un état d'un système de simulation est corrigé en fonction d'une erreur entre le système de simulation et un système réel ; ym est un résultat de mesure contenant du bruit, comprenant la distance relative

    et l'angle relatif

    délivrés en sortie par le capteur de position relative ; est une variable intermédiaire ; P est une matrice de probabilité de propagation ; et est une dérivée de la matrice de probabilité de propagation.
     
    9. Support de stockage d'ordinateur stockant des instructions exécutables par ordinateur qui, lorsqu'elles sont exécutées par un processeur, amènent le processeur à exécuter la méthode selon l'une quelconque des revendications 1 à 4.
     




    Drawing











    Cited references

    REFERENCES CITED IN THE DESCRIPTION



    This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

    Patent documents cited in the description