阅读说明：本技术 基于载波观测值历元间双差分的高精度相对运动矢量算法 (High-precision relative motion vector algorithm based on double differences between carrier observed value epochs ) 是由 杨文豪 刘强 王超 陈文俊 喻炎森 滕玲 董方云 李健 李占刚 方帅 邢亚 于 2021-08-18 设计创作，主要内容包括：本发明公开了基于载波观测值历元间双差分的高精度相对运动矢量算法,包括以下步骤：S1、算法获取相邻历元中移动站和基准站接收机的载波、伪距观测值,并通过判断卫星仰角来选取合适的卫星观测值,去除或减弱可能存在的测量值多径效应误差影响,判断载波测量值中是否存在周跳,本发明通过构建历元间双差分载波观测模型,消除整周模糊度项,求解历元间接收机相对运动矢量,采用推位算法给出相对运动跟踪定位结果,基于速度信息约束和钟差漂移解,给出了相对运动矢量解正确性检验方法及错误解修复方法,解决了相对运动矢量解求取失败导致的推位算法误差项不断累积,使得最终的跟踪定位结果变差甚至根本无法使用的问题。(The invention discloses a high-precision relative motion vector algorithm based on double differences between epochs of a carrier observed value, which comprises the following steps: s1, obtaining carrier wave and pseudo range observation values of a mobile station and a reference station receiver in adjacent epochs by an algorithm, selecting a proper satellite observation value by judging a satellite elevation angle, removing or weakening possible influence of measurement value multipath effect errors, and judging whether cycle slip exists in a carrier measurement value.)

1. The high-precision relative motion vector algorithm based on the double difference between carrier observed value epochs is characterized in that: the method comprises the following steps:

s1, firstly, obtaining carrier and pseudo-range observed values of a mobile station and a reference station receiver in adjacent epochs by an algorithm, selecting a proper satellite observed value by judging a satellite elevation, removing or weakening the influence of a possibly existing measured value multipath effect error, and judging whether cycle slip exists in a carrier measured value or not;

s2, after all available satellite carrier observed values in adjacent epochs are obtained, calculating double-difference observed values between the epochs;

s3, calculating approximate coordinate values of the current epoch mobile station receiver based on the baseline vector between the previous epoch receivers, and calculating an observation matrix, a weight matrix and a double-difference observation residual error item between the epochs based on the approximate coordinate values;

s4, providing a relative motion vector solution between receivers in adjacent epochs, and judging whether the relative motion vector solution exceeds a threshold limit;

s5, if not, the correctness of the relative motion vector solution is continuously checked by solving the clock difference drift solution; if the solution of the relative motion solution is correct, the vector solution is superposed to the baseline vector of the two receivers of the previous epoch, and the positioning tracking is finished; if the relative motion vector is failed to be solved, adopting a correct value solved by the previous epoch to complete a carry-over algorithm;

s6, when the failure times of vector solution test exceed the threshold limit, the EKF algorithm through parallel operation reinitializes the positioning result of the positioning algorithm in the current epoch so as to reduce the long-time accumulation of the relative motion tracking positioning error;

and S7, repeating the steps for all the adjacent epoch carrier observed values, and continuously providing a relative motion tracking and positioning result.

2. The carrier observation epoch based double difference high precision relative motion vector algorithm of claim 1, wherein: the algorithm of the time domain double differential observation value in step S2 is that the time domain double differential observation model has the advantages that the reference satellite does not need to be determined, the observation value does not contain the carrier integer ambiguity term, and the observation value contains the information of the change of the true geometric distance between the satellite and the receiver between the adjacent epochs, because the observation model adopts the carrier phase observation value, the accurate relative motion tracking position solution can be given through the observation model, the time domain double difference observation model is the secondary difference of the epoch to the single difference observation value, the single difference carrier phase observation model is the primary difference of two receivers to the carrier measurement value of the same satellite at the same time, the single difference can be used for removing the satellite clock difference in the measurement value, under the condition of a short baseline, the atmospheric time delay error of the satellite signal can be eliminated, and the single-difference carrier phase observation equation can be expressed as follows:

in the formula, u and r represent user receivers, respectivelyAnd reference station receiver number; δ t_urRepresenting a single difference in clock difference between the receivers;

it can be seen from the equation of single-difference observation that the satellite clock difference and the signal atmospheric delay error can be cancelled, but the receiver clock difference error term is still not eliminated, and the whole-cycle ambiguity is still the unknown integer term to be solved, and only if the ambiguity value is correctly solved and fixed, the high-precision positioning result can be obtained, and it should be pointed out that although the single-difference operation can eliminate most of the measurement errors in the observed quantity, the single-difference post-measurement noise can be eliminatedThe mean square error of the carrier phase is changed into the mean square error of the observation noise of the original carrier phaseBased on the single difference observation equation, the time domain double difference observation can be written as:

in the formulaRepresenting a time domain double difference and,

according to the formula, if the receiver continuously locks the satellite signal, the single difference observed value between the receivers carries out secondary difference on adjacent epochs to remove the whole-cycle ambiguity term, and only excessive receiver clock difference drift terms are reservedAnd the unmodeled noise error, at the moment, the double-difference observation model between the epochs completely keeps the variation of the geometric real distance difference between the receiver and the satellite in the adjacent epochs, and only needs to use the difference value for facilitating the subsequent descriptionRepresenting a time domain differential observation model:

as can be seen from the time domain double-difference observation equation, the model contains the true geometric distance variation value between the satellite and the receiver, and can be directly used to find the relative motion vector, and review the single-difference observation equation, where the observation value is the single difference of the two receivers to the same satellite measurement value, and this value represents the difference between the two receivers and the geometric distance between the satellite, and since the distance between the navigation satellite and the receiver is very far, the unit observation direction vectors of the two receivers to the same satellite can be considered to be equal, this assumption has little influence on the resolution precision of the relative motion vector, at this time, the single-difference geometric distance value between the satellite and the receiver can be equal to the projection of the baseline vector of the two receivers on the unit observation direction vector of the satellite, so the single-difference geometric distance value can be replaced by the following formula:

in the formula, bl_urRepresents a baseline vector between receivers u and r;a unit observation direction vector representing that the receiver points to the satellite; ". represents the vector inner product, in which case the equation can be transformed to:

the time domain double difference observation model equation can be changed into:

since the unmodeled error term in the carrier observation is less than a few millimeters, the unmodeled error term can be removed from the observation equation:

assuming that unit observation direction vectors of a receiver between two adjacent epochs are equal, the influence P on positioning result error is set_errCan be represented by the following formula:

P_err≈10.636×10^-9×bl_ur

as can be seen from the equation, even if the positioning error value introduced by the assumption is less than 1cm when the two receivers are separated by 1000km, the error value can be almost ignored, so based on the assumption, the equation can be further modified as follows:

from the equation, the time domain double-difference observation value includes the baseline vector change value information of two receivers in adjacent epochs, and at this time, if the number of observation satellites is n, the observation system equation can be written as:

Φ＝ρ+C×t_ur

wherein the content of the first and second substances,

at the initial estimated position of the mobile receiver for the observation system equation (X)_b+b₀) Taylor series expansion is carried out to obtain:

in the formula, e represents an error vector (n × 1); x_bRepresenting reference station absolute coordinate vector values; b₀Representing an initial estimate of the baseline vector between the receivers; b represents a systematic Jacobian observation matrix (n × 4); δ is the residual vector (n × 1); Δ b represents the change in baseline vector delta between receivers in adjacent epochs, i.e., the relative motion vector solution;representing the initial receiver clock drift value, the optimal solution problem can be expressed as,

where W represents a weight matrix, which can be represented by,

W＝P^-1

in the formula (I), the compound is shown in the specification,the standard deviation of the nth satellite observation is expressed as:

wherein a, b and c represent carrier observation error factors; el represents the elevation value of the observation satellite; d represents the satellite clock error, in which case the solution to the optimization problem equation can be expressed as,

Δb＝-(B^TWB)^-1B^TWδ

at this time, Δ b is a change value of the baseline vector of the two receivers in the adjacent epoch, that is, a relative motion vector solution between the receivers in the adjacent epoch, and the relative motion vector solution of the current adjacent epoch is continuously superimposed on the positioning result of the previous epoch through a carry-over algorithm, so as to obtain a relative motion tracking positioning result between the receivers.

3. The carrier observation epoch based double difference high precision relative motion vector algorithm of claim 1, wherein: when the double-difference carrier observed value between epochs is adopted to solve the receiver relative motion vector solution in the step S4, the number of satellites required to be continuously locked by the receiver is more than four, and in a complex environment, the receiver may have a satellite lock losing phenomenon, so that cycle slip exists in the carrier phase observed value, and the influence of the cycle slip on the relative motion vector solving precision is monitored and repaired, which is a necessary condition for continuously obtaining a high-precision relative motion tracking positioning result.

4. The carrier observation epoch based double difference high precision relative motion vector algorithm of claim 1, wherein: in step S5, the double-difference observed value between epochs includes, in addition to the relative motion vector information between receivers in adjacent epochs, a clock difference drift term, which is a drift error that gradually occurs in the single-difference clock difference value between the receivers as the stability of the quartz clock used in the GNSS receiver is inferior to that of the high-precision atomic clock at the satellite end, and may be specifically expanded as follows:

Δ▽δt_ur＝δt_u(t_i)-δt_u(t_i-1)-(δt_r(t_i)-δt_r(t_i-1))

where, δ t represents the receiver clock difference,unit of meter; t is t_iRepresenting a corresponding epoch time; the sign Δ ∑ represents a time domain double difference calculation, although the error of the clock difference term of the receiver is large, the double difference clock difference drift value between the receivers in adjacent epochs is small and stable, so whether the relative motion vector solution of the current epoch is correct or not can be judged through the clock difference drift solution value,

the clock offset drift value solved based on the double-difference observed value between the epochs, the solution in each epoch is stable, when the solution of the relative motion vector solution fails to be solved due to the cycle slip influence of the carrier observed value and the like, the clock offset drift value has very obvious abnormal deviation items, and the abnormal value can reach 3.8 multiplied by 10¹⁷Therefore, the clock offset drift value can be used as a criterion for checking whether the solution of the relative motion vector is correct, and meanwhile, when the relative motion tracking and positioning are carried out by adopting a single-frequency receiver, the situation that the motion speed of an object basically does not exceed 300m/s does not exist, so that the correctness of the solution value of the relative motion vector can be effectively monitored in a mode of setting speed threshold value constraint.

5. The carrier observation epoch based double difference high precision relative motion vector algorithm of claim 1, wherein: in step S5, after the wrong relative motion vector solution is monitored, the wrong solution needs to be repaired, and in consideration of the situation that the sampling interval is short enough, for example, the sampling frequency is 5Hz, the receiver will basically keep the same motion state in the previous and subsequent epochs, and at this time, the wrong solution of the current epoch can be repaired by using the relative motion vector solution correctly calculated by the previous epoch, and the method is applied to the locating algorithm;

through an extended Kalman filtering algorithm of parallel operation, when the error frequency of solving the relative motion vector solution exceeds a threshold limit, the relative motion positioning result of the positioning algorithm in the current epoch is reinitialized so as to solve the problem that the error items of the positioning algorithm are continuously accumulated due to the failure of solving the relative motion vector solution, so that the final tracking positioning result is deteriorated or even cannot be used at all;

the Extended Kalman Filter (EKF) formula of the parallel operation is as follows:

in the formula, symbolAndrespectively representing a priori estimated value and a posteriori updated value;is shown at t_iThe state vector to be solved at the moment comprises a receiver baseline vector and a receiver movement speed;represents t_iOf time of dayA covariance matrix;represents t_iA Kalman filter gain at a time;and h (x) represents a carrier double-difference observed value and an observed predicted value respectively;representing a measurement error covariance matrix; h (x) represents the Jacobian observation matrix;andrespectively representing slave time t_i-1To time t_iThe state transition matrix of (a) and the covariance matrix of the process noise vector,andthe matrices can be found by the following equations, respectively:

wherein I represents an identity matrix and Q_vThe deployment can continue as follows:

in the formula, τ_rRepresents a GNSS signal sampling interval in seconds;respectively, the standard deviation of the receiver velocity measurement noise in the northeast direction.

Technical Field

The invention relates to the technical field of carrier observed values, in particular to a high-precision relative motion vector algorithm based on double differences between epochs of the carrier observed values.

Background

With the continuous expansion of the application range of global satellite navigation positioning systems, satellite navigation positioning becomes a preferred positioning mode for outdoor application, which brings huge economic value and social value, and accelerates the development of high-precision positioning technology, mass market application still gives priority to low-cost single-frequency receivers at present, most applications do not need precise absolute position positioning information in practice, but pay more attention to relative position and relative motion positioning information among nodes, such as anti-collision warning, precision agriculture, wireless sensor networks, unmanned aerial vehicle formation and the like, when a relative positioning method based on solving integer ambiguity adopts a single-frequency carrier observed value for positioning, the positioning precision is easily affected by the cycle slip of the carrier, and under the serious condition of cycle slip, the positioning solution is changed into a floating solution, the positioning precision is reduced, and in addition, even if satellite signals are relocked, the method still needs to spend long convergence time to give a high-precision relative motion tracking and positioning result again, a learner turns the research direction into a high-precision relative motion tracking and positioning method without solving the whole-cycle ambiguity to remove the influence of cycle slip on the positioning precision, Will doctor and the like propose a real-time relative motion tracking and positioning method based on single-frequency carrier observation quantity, the method removes the whole-cycle ambiguity item in the observed value by constructing a secondary difference observation model between epochs, meanwhile, the method does not need to set a reference station, all receivers can obtain the high-precision relative motion tracking and positioning result of each other, but the method only supports a GPS carrier observation value, the long-time dynamic tracking and positioning precision is still influenced by the cycle slip value in the carrier measurement, and the high-precision relative position and relative motion tracking and positioning information are needed to be provided for more and more market applications, the carrier measurement value is used as a preferred ranging mode, the cycle slip phenomenon of the measurement value is a main error source influencing the long-term relative motion tracking and positioning accuracy, the relative motion tracking and positioning method based on the elimination of the whole-cycle ambiguity is free from the whole-cycle ambiguity solving problem, the method is lower in calculation complexity and less in required calculation resources, the research of the long-term robust relative motion tracking and positioning method based on the algorithm is developed, the advantage that the algorithm positioning accuracy is not influenced by the cycle slip of the carrier observation value or is quickly recovered from the influence of the cycle slip is given to the algorithm, the requirement of mass market application on obtaining accurate relative motion and relative position information among devices by adopting a single-frequency receiver is met, and the method has important practical application value.

When the relative positioning method based on the integer ambiguity solution is used for positioning by adopting a single-frequency carrier observed value, the positioning precision is easily influenced by the cycle slip of the carrier observed value, and under the condition of serious cycle slip, the positioning solution is changed into a floating solution, so that the positioning precision is reduced. In addition, even if the satellite signal is locked again, the method still needs to take a long convergence time to give a high-precision relative motion tracking positioning result again, so a high-precision relative motion vector algorithm based on double differences between epochs of the carrier observed values is proposed to solve the above-mentioned problems.

Disclosure of Invention

The invention aims to provide a high-precision relative motion vector algorithm based on double differences between epochs of a carrier observed value so as to solve the problems in the background technology.

In order to achieve the purpose, the invention provides the following technical scheme: the high-precision relative motion vector algorithm based on the double difference between the carrier observation value epochs comprises the following steps:

s1, firstly, obtaining carrier and pseudo-range observed values of a mobile station and a reference station receiver in adjacent epochs by an algorithm, selecting a proper satellite observed value by judging a satellite elevation, removing or weakening the influence of a possibly existing measured value multipath effect error, and judging whether cycle slip exists in a carrier measured value or not;

s2, after all available satellite carrier observed values in adjacent epochs are obtained, calculating double-difference observed values between the epochs;

s3, calculating approximate coordinate values of the current epoch mobile station receiver based on the baseline vector between the previous epoch receivers, and calculating an observation matrix, a weight matrix and a double-difference observation residual error item between the epochs based on the approximate coordinate values;

s4, providing a relative motion vector solution between receivers in adjacent epochs, and judging whether the relative motion vector solution exceeds a threshold limit;

s5, if not, the correctness of the relative motion vector solution is continuously checked by solving the clock difference drift solution; if the solution of the relative motion solution is correct, the vector solution is superposed to the baseline vector of the two receivers of the previous epoch, and the positioning tracking is finished; if the relative motion vector is failed to be solved, adopting a correct value solved by the previous epoch to complete a carry-over algorithm;

s6, when the failure times of vector solution test exceed the threshold limit, the EKF algorithm through parallel operation reinitializes the positioning result of the positioning algorithm in the current epoch so as to reduce the long-time accumulation of the relative motion tracking positioning error;

and S7, repeating the steps for all the adjacent epoch carrier observed values, and continuously providing a relative motion tracking and positioning result.

Preferably, the algorithm of the time domain double differential observation in step S2 is that the time domain double differential observation model has the advantages that the reference satellite does not need to be determined, the observation does not contain the carrier integer ambiguity any more, and the observation value contains the information of the change of the true geometric distance between the satellite and the receiver between the adjacent epochs, because the observation model adopts the carrier phase observation value, the accurate relative motion tracking position solution can be given through the observation model, the time domain double difference observation model is the secondary difference of the epoch to the single difference observation value, the single difference carrier phase observation model is the primary difference of two receivers to the carrier measurement value of the same satellite at the same time, the single difference can be used for removing the satellite clock difference in the measurement value, under the condition of a short baseline, the atmospheric time delay error of the satellite signal can be eliminated, and the single-difference carrier phase observation equation can be expressed as follows:

in the formula, u and r respectively represent the numbers of a user receiver and a reference station receiver; δ t_urRepresenting a single difference in clock difference between the receivers;

it can be seen from the equation of single-difference observation that the satellite clock difference and the signal atmospheric delay error can be cancelled, but the receiver clock difference error term is still not eliminated, and the whole-cycle ambiguity is still the unknown integer term to be solved, and only if the ambiguity value is correctly solved and fixed, the high-precision positioning result can be obtained, and it should be pointed out that although the single-difference operation can eliminate most of the measurement errors in the observed quantity, the single-difference post-measurement noise can be eliminatedThe mean square error of the carrier phase is changed into the mean square error of the observation noise of the original carrier phaseMultiple, based on the equation of single difference observation, timeThe domain double differential observations can be written as:

where Δ ∑ represents a time domain double difference,

according to the formula, if the receiver continuously locks the satellite signal, the single difference observed values between the receivers are subjected to secondary difference on adjacent epochs to remove the whole-cycle ambiguity term, and only the excessive receiver clock difference drift term delta t is reserved_urAnd (v) and unmodeled noise error, wherein the variation of the geometric true distance difference between the receiver and the satellite in the adjacent epoch is completely reserved by the double-difference observation model between the epochs, and for convenience of subsequent description, the time-domain difference observation model is represented by delta ^:

as can be seen from the time domain double-difference observation equation, the model contains the true geometric distance variation value between the satellite and the receiver, and can be directly used to find the relative motion vector, and review the single-difference observation equation, where the observation value is the single difference of the two receivers to the same satellite measurement value, and this value represents the difference between the two receivers and the geometric distance between the satellite, and since the distance between the navigation satellite and the receiver is very far, the unit observation direction vectors of the two receivers to the same satellite can be considered to be equal, this assumption has little influence on the resolution precision of the relative motion vector, at this time, the single-difference geometric distance value between the satellite and the receiver can be equal to the projection of the baseline vector of the two receivers on the unit observation direction vector of the satellite, so the single-difference geometric distance value can be replaced by the following formula:

in the formula, bl_urRepresenting a baseline vector between receivers u and r；A unit observation direction vector representing that the receiver points to the satellite; ". represents the vector inner product, in which case the equation can be transformed to:

the time domain double difference observation model equation can be changed into:

since the unmodeled error term in the carrier observation is less than a few millimeters, the unmodeled error term can be removed from the observation equation:

assuming that unit observation direction vectors of a receiver between two adjacent epochs are equal, the influence P on positioning result error is set_errCan be represented by the following formula:

P_err≈10.636×10^-9×bl_ur

as can be seen from the equation, even if the positioning error value introduced by the assumption is less than 1cm when the two receivers are separated by 1000km, the error value can be almost ignored, so based on the assumption, the equation can be further modified as follows:

from the equation, the time domain double-difference observation value includes the baseline vector change value information of two receivers in adjacent epochs, and at this time, if the number of observation satellites is n, the observation system equation can be written as:

Φ＝ρ+C×t_ur

wherein the content of the first and second substances,

at the initial estimated position of the mobile receiver for the observation system equation (X)_b+b₀) Taylor series expansion is carried out to obtain:

in the formula, e represents an error vector (n × 1); x_bRepresenting reference station absolute coordinate vector values; b₀Representing an initial estimate of the baseline vector between the receivers; b represents a systematic Jacobian observation matrix (n × 4); δ is the residual vector (n × 1); Δ b represents the change in baseline vector delta between receivers in adjacent epochs, i.e., the relative motion vector solution;representing the initial receiver clock drift value, the optimal solution problem can be expressed as,

where W represents a weight matrix, which can be represented by,

W＝P^-1

in the formula (I), the compound is shown in the specification,the standard deviation of the nth satellite observation is expressed as:

wherein a, b and c represent carrier observation error factors; el represents the elevation value of the observation satellite; d represents the satellite clock error, in which case the solution to the optimization problem equation can be expressed as,

Δb＝-(B^TWB)^-1B^TWδ

at this time, Δ b is a change value of the baseline vector of the two receivers in the adjacent epoch, that is, a relative motion vector solution between the receivers in the adjacent epoch, and the relative motion vector solution of the current adjacent epoch is continuously superimposed on the positioning result of the previous epoch through a carry-over algorithm, so as to obtain a relative motion tracking positioning result between the receivers.

Preferably, when the double-difference carrier observed value between epochs is used to solve the receiver relative motion vector solution in step S4, the number of satellites that need to be continuously locked by the receiver is greater than four, and in a complex environment, the receiver may have a satellite out-of-lock phenomenon, so that cycle slip exists in the carrier phase observed value, and monitoring and repairing the influence of cycle slip on the relative motion vector solution accuracy is a necessary condition for continuously obtaining a high-accuracy relative motion tracking positioning result.

Preferably, the double-difference observed value between epochs in step S5 includes, in addition to the relative motion vector information between receivers in adjacent epochs, a clock difference drift term, which is caused by the fact that the stability of a quartz clock used in a GNSS receiver is inferior to that of a high-precision atomic clock at a satellite end, and as the single-difference clock difference value between receivers is measured, drift errors gradually occur, and the double-difference clock difference drift term can be specifically expanded as follows:

Δ▽δt_ur＝δt_u(t_i)-δt_u(t_i-1)-(δt_r(t_i)-δt_r(t_i-1))

wherein δ t represents the receiver clock error in meters; t is t_iRepresenting a corresponding epoch time; the sign Δ ∑ represents a time domain double difference calculation, although the error of the clock difference term of the receiver is large, the double difference clock difference drift value between the receivers in adjacent epochs is small and stable, so whether the relative motion vector solution of the current epoch is correct or not can be judged through the clock difference drift solution value,

the clock offset drift value solved based on the double-difference observed value between the epochs, the solution in each epoch is stable, when the solution of the relative motion vector solution fails to be solved due to the cycle slip influence of the carrier observed value and the like, the clock offset drift value has very obvious abnormal deviation items, and the abnormal value can reach 3.8 multiplied by 10¹⁷Therefore, the clock offset drift value can be used as a criterion for checking whether the solution of the relative motion vector is correct, and meanwhile, when the relative motion tracking and positioning are carried out by adopting a single-frequency receiver, the situation that the motion speed of an object basically does not exceed 300m/s does not exist, so that the correctness of the solution value of the relative motion vector can be effectively monitored in a mode of setting speed threshold value constraint.

Preferably, in step S5, after the erroneous relative motion vector solution is monitored, the erroneous solution needs to be repaired, and in consideration of the situation that the sampling interval is short enough, for example, the sampling frequency is 5Hz, the receiver keeps the same motion state in the previous and subsequent adjacent epochs, at this time, the erroneous solution of the current epoch can be repaired by using the relative motion vector solution correctly solved by the previous epoch, and the method is applied to the locating algorithm;

through an extended Kalman filtering algorithm of parallel operation, when the error frequency of solving the relative motion vector solution exceeds a threshold limit, the relative motion positioning result of the positioning algorithm in the current epoch is reinitialized so as to solve the problem that the error items of the positioning algorithm are continuously accumulated due to the failure of solving the relative motion vector solution, so that the final tracking positioning result is deteriorated or even cannot be used at all;

the Extended Kalman Filter (EKF) formula of the parallel operation is as follows:

in the formula, symbolAndrespectively representing a priori estimated value and a posteriori updated value;is shown at t_iThe state vector to be solved at the moment comprises a receiver baseline vector and a receiver movement speed;represents t_iOf time of dayA covariance matrix;represents t_iA Kalman filter gain at a time;and h (x) represents a carrier double-difference observed value and an observed predicted value respectively;representing a measurement error covariance matrix; h (x) represents the Jacobian observation matrix;andrespectively representing slave time t_i-1To time t_iThe state transition matrix of (a) and the covariance matrix of the process noise vector,andthe matrices can be found by the following equations, respectively:

wherein I represents an identity matrix and Q_vThe deployment can continue as follows:

in the formula, τ_rRepresents a GNSS signal sampling interval in seconds;standard deviation for respectively representing velocity measurement noise of receiver in north-east directionAnd (4) poor.

Compared with the prior art, the improved relative motion tracking and positioning algorithm has the beneficial effects that through constructing a double-differential carrier observation model between epochs, the whole-cycle ambiguity item is eliminated, the relative motion vector of a receiver between epochs is solved, a relative motion tracking and positioning result is given by adopting a positioning algorithm, and a relative motion vector solution correctness checking method and an error solution repairing method are given based on speed information constraint and clock offset solution. Through the extended Kalman filtering algorithm of parallel operation, when the error frequency of solving the relative motion vector solution exceeds the threshold limit, the relative motion positioning result of the positioning algorithm in the current epoch is reinitialized, so that the problem that the final tracking positioning result is poor or even cannot be used at all due to continuous accumulation of error items of the positioning algorithm caused by failure of solving the relative motion vector is solved.

Drawings

FIG. 1 is a schematic diagram of the clock offset drift solution and its abnormal solution result in the present invention;

FIG. 2 is a schematic diagram of a positioning result of an extended Kalman filtering algorithm reinitialization relocation algorithm in the present invention;

FIG. 3 is a flow chart of the M-Regtrack relative tracking positioning algorithm of the present invention;

FIG. 4 is a schematic diagram of the relationship between the single-difference geometric distance values and the baseline vector between satellite receivers according to the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

In the description of the present invention, it should be noted that the terms "upper", "lower", "inner", "outer", "front", "rear", "both ends", "one end", "the other end", and the like indicate orientations or positional relationships based on those shown in the drawings, and are only for convenience of description and simplicity of description, but do not indicate or imply that the referred device or element must have a specific orientation, be constructed in a specific orientation, and be operated, and thus, should not be construed as limiting the present invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.

In the description of the present invention, it is to be noted that, unless otherwise explicitly specified or limited, the terms "mounted," "disposed," "connected," and the like are to be construed broadly, such as "connected," which may be fixedly connected, detachably connected, or integrally connected; can be mechanically or electrically connected; they may be connected directly or indirectly through intervening media, or they may be interconnected between two elements. The specific meanings of the above terms in the present invention can be understood in specific cases to those skilled in the art.

Example one

Referring to fig. 1 to 4, an embodiment of the present invention includes: the high-precision relative motion vector algorithm based on the double difference between the carrier observation value epochs comprises the following steps:

s1, firstly, obtaining carrier and pseudo-range observed values of a mobile station and a reference station receiver in adjacent epochs by an algorithm, selecting a proper satellite observed value by judging a satellite elevation, removing or weakening the influence of a possibly existing measured value multipath effect error, and judging whether cycle slip exists in a carrier measured value or not;

s2, after all available satellite carrier observed values in adjacent epochs are obtained, calculating double-difference observed values between the epochs;

s3, calculating approximate coordinate values of the current epoch mobile station receiver based on the baseline vector between the previous epoch receivers, and calculating an observation matrix, a weight matrix and a double-difference observation residual error item between the epochs based on the approximate coordinate values;

s4, providing a relative motion vector solution between receivers in adjacent epochs, and judging whether the relative motion vector solution exceeds a threshold limit;

s5, if not, the correctness of the relative motion vector solution is continuously checked by solving the clock difference drift solution; if the solution of the relative motion solution is correct, the vector solution is superposed to the baseline vector of the two receivers of the previous epoch, and the positioning tracking is finished; if the relative motion vector is failed to be solved, adopting a correct value solved by the previous epoch to complete a carry-over algorithm;

s6, when the failure times of vector solution test exceed the threshold limit, the EKF algorithm through parallel operation reinitializes the positioning result of the positioning algorithm in the current epoch so as to reduce the long-time accumulation of the relative motion tracking positioning error;

and S7, repeating the steps for all the adjacent epoch carrier observed values, and continuously providing a relative motion tracking and positioning result.

Example two

Referring to fig. 1 to 4, an embodiment of the present invention includes: the high-precision relative motion vector algorithm based on the double difference between the carrier observation value epochs comprises the following steps:

s1, firstly, obtaining carrier and pseudo-range observed values of a mobile station and a reference station receiver in adjacent epochs by an algorithm, selecting a proper satellite observed value by judging a satellite elevation, removing or weakening the influence of a possibly existing measured value multipath effect error, and judging whether cycle slip exists in a carrier measured value or not;

s2, after all the available satellite carrier observed values in the adjacent epochs are obtained, calculating double-difference observed values between the epochs,

the algorithm of the time domain double differential observation value in step S2 is that the time domain double differential observation model has the advantages that the reference satellite does not need to be determined, the observation value does not contain the carrier integer ambiguity term, and the observation value contains the information of the change of the true geometric distance between the satellite and the receiver between the adjacent epochs, because the observation model adopts the carrier phase observation value, the accurate relative motion tracking position solution can be given through the observation model, the time domain double difference observation model is the secondary difference of the epoch to the single difference observation value, the single difference carrier phase observation model is the primary difference of two receivers to the carrier measurement value of the same satellite at the same time, the single difference can be used for removing the satellite clock difference in the measurement value, under the condition of a short baseline, the atmospheric time delay error of the satellite signal can be eliminated, and the single-difference carrier phase observation equation can be expressed as follows:

in the formula, u and r respectively represent the numbers of a user receiver and a reference station receiver; δ t_urRepresenting a single difference in clock difference between the receivers;

it can be seen from the equation of single-difference observation that the satellite clock difference and the signal atmospheric delay error can be cancelled, but the receiver clock difference error term is still not eliminated, and the whole-cycle ambiguity is still the unknown integer term to be solved, and only if the ambiguity value is correctly solved and fixed, the high-precision positioning result can be obtained, and it should be pointed out that although the single-difference operation can eliminate most of the measurement errors in the observed quantity, the single-difference post-measurement noise can be eliminatedThe mean square error of the carrier phase is changed into the mean square error of the observation noise of the original carrier phaseBased on the single difference observation equation, the time domain double difference observation can be written as:

where Δ ∑ represents a time domain double difference,

according to the formula, if the receiver continuously locks the satellite signal, the single difference observed values between the receivers are subjected to secondary difference on adjacent epochs to remove the whole-cycle ambiguity term, and only the excessive receiver clock difference drift term delta t is reserved_urT and unmodeled noise error, at which time the geometric true distance between the receiver and the satellite in the adjacent epoch is completely reserved by the double difference observation model between epochsFor convenience of subsequent description, the variation of the difference is represented by Δ ∑ only for the time domain difference observation model:

as can be seen from the time domain double-difference observation equation, the model contains the true geometric distance variation value between the satellite and the receiver, and can be directly used to find the relative motion vector, and review the single-difference observation equation, where the observation value is the single difference of the two receivers to the same satellite measurement value, and this value represents the difference between the two receivers and the geometric distance between the satellite, and since the distance between the navigation satellite and the receiver is very far, the unit observation direction vectors of the two receivers to the same satellite can be considered to be equal, this assumption has little influence on the resolution precision of the relative motion vector, at this time, the single-difference geometric distance value between the satellite and the receiver can be equal to the projection of the baseline vector of the two receivers on the unit observation direction vector of the satellite, so the single-difference geometric distance value can be replaced by the following formula:

in the formula, bl_urRepresents a baseline vector between receivers u and r;a unit observation direction vector representing that the receiver points to the satellite; ". represents the vector inner product, in which case the equation can be transformed to:

the time domain double difference observation model equation can be changed into:

since the unmodeled error term in the carrier observation is less than a few millimeters, the unmodeled error term can be removed from the observation equation:

assuming that unit observation direction vectors of a receiver between two adjacent epochs are equal, the influence P on positioning result error is set_errCan be represented by the following formula:

P_err≈10.636×10^-9×bl_ur

as can be seen from the equation, even if the positioning error value introduced by the assumption is less than 1cm when the two receivers are separated by 1000km, the error value can be almost ignored, so based on the assumption, the equation can be further modified as follows:

from the equation, the time domain double-difference observation value includes the baseline vector change value information of two receivers in adjacent epochs, and at this time, if the number of observation satellites is n, the observation system equation can be written as:

Φ＝ρ+C×t_ur

wherein the content of the first and second substances,

at the initial estimated position of the mobile receiver for the observation system equation (X)_b+b₀) Taylor series expansion is carried out to obtain:

in the formula, e represents an error vector (n × 1); x_bRepresenting reference station absolute coordinate vector values; b₀Representing an initial estimate of the baseline vector between the receivers; b represents a systematic Jacobian observation matrix (n × 4); δ is the residual vector (n × 1); Δ b represents the change in baseline vector delta between receivers in adjacent epochs, i.e., the relative motion vector solution;representing the initial receiver clock drift value, the optimal solution problem can be expressed as,

where W represents a weight matrix, which can be represented by,

W＝P^-1

in the formula (I), the compound is shown in the specification,the standard deviation of the nth satellite observation is expressed as:

wherein a, b and c represent carrier observation error factors; el represents the elevation value of the observation satellite; d represents the satellite clock error, in which case the solution to the optimization problem equation can be expressed as,

Δb＝-(B^TWB)^-1B^TWδ

at the moment, delta b is a base line vector change value of the two receivers in adjacent epochs, namely a relative motion vector solution between the receivers in the adjacent epochs, and the relative motion vector solution of the current adjacent epochs is continuously superposed into a positioning result of the previous epoch through a carry-over algorithm so as to obtain a relative motion tracking positioning result between the receivers;

s3, calculating approximate coordinate values of the current epoch mobile station receiver based on the baseline vector between the previous epoch receivers, and calculating an observation matrix, a weight matrix and a double-difference observation residual error item between the epochs based on the approximate coordinate values;

s4, giving out the relative motion vector solution between receivers in adjacent epochs, judging whether the relative motion vector solution exceeds the threshold limit,

when the double-difference carrier observed value between epochs is adopted to solve the receiver relative motion vector solution in the step S4, the number of satellites required to be continuously locked by the receiver is more than four, and in a complex environment, the receiver may have a satellite lock losing phenomenon, so that cycle slip exists in the carrier phase observed value, and the influence of the cycle slip on the relative motion vector solving precision is monitored and repaired, so that the method is a necessary condition for continuously obtaining a high-precision relative motion tracking positioning result;

s5, if not, the correctness of the relative motion vector solution is continuously checked by solving the clock difference drift solution; if the solution of the relative motion solution is correct, the vector solution is superposed to the baseline vector of the two receivers of the previous epoch, and the positioning tracking is finished; if the solving of the relative motion vector fails, the correct value solved by the previous epoch is adopted to complete the bit-pulling algorithm,

in step S5, the double-difference observed value between epochs includes, in addition to the relative motion vector information between receivers in adjacent epochs, a clock difference drift term, which is a drift error that gradually occurs in the single-difference clock difference value between the receivers as the stability of the quartz clock used in the GNSS receiver is inferior to that of the high-precision atomic clock at the satellite end, and may be specifically expanded as follows:

Δ▽δt_ur＝δt_u(t_i)-δt_u(t_i-1)-(δt_r(t_i)-δt_r(t_i-1))

wherein δ t represents the receiver clock error in meters; t is t_iRepresenting a corresponding epoch time; the sign Δ ∑ represents a time domain double difference calculation, although the error of the clock difference term of the receiver is large, the double difference clock difference drift value between the receivers in adjacent epochs is small and stable, so whether the relative motion vector solution of the current epoch is correct or not can be judged through the clock difference drift solution value,

the clock offset drift value solved based on the double-difference observed value between the epochs, the solution in each epoch is stable, when the solution of the relative motion vector solution fails to be solved due to the cycle slip influence of the carrier observed value and the like, the clock offset drift value has very obvious abnormal deviation items, and the abnormal value can reach 3.8 multiplied by 10¹⁷Therefore, the clock offset drift value can be used as a criterion for checking whether the solution of the relative motion vector is correct, and meanwhile, when the relative motion tracking and positioning are carried out by adopting a single-frequency receiver, the situation that the motion speed of an object basically does not exceed 300m/s does not exist, so that the correctness of the solution value of the relative motion vector can be effectively monitored in a mode of setting speed threshold value constraint.

In step S5, after the wrong relative motion vector solution is monitored, the wrong solution needs to be repaired, and in consideration of the situation that the sampling interval is short enough, for example, the sampling frequency is 5Hz, the receiver will basically keep the same motion state in the previous and subsequent epochs, and at this time, the wrong solution of the current epoch can be repaired by using the relative motion vector solution correctly calculated by the previous epoch, and the method is applied to the locating algorithm;

through an extended Kalman filtering algorithm of parallel operation, when the error frequency of solving the relative motion vector solution exceeds a threshold limit, the relative motion positioning result of the positioning algorithm in the current epoch is reinitialized so as to solve the problem that the error items of the positioning algorithm are continuously accumulated due to the failure of solving the relative motion vector solution, so that the final tracking positioning result is deteriorated or even cannot be used at all;

the Extended Kalman Filter (EKF) formula of the parallel operation is as follows:

in the formula, symbolAndrespectively representing a priori estimated value and a posteriori updated value;is shown at t_iThe state vector to be solved at the moment comprises a receiver baseline vector and a receiver movement speed;represents t_iOf time of dayA covariance matrix;represents t_iKalman filtering of momentsGain;and h (x) represents a carrier double-difference observed value and an observed predicted value respectively;representing a measurement error covariance matrix; h (x) represents the Jacobian observation matrix;andrespectively representing slave time t_i-1To time t_iThe state transition matrix of (a) and the covariance matrix of the process noise vector,andthe matrices can be found by the following equations, respectively:

wherein I represents an identity matrix and Q_vThe deployment can continue as follows:

in the formula, τ_rRepresents a GNSS signal sampling interval in seconds;respectively representing the speed of the receiver in the northeast directionMeasuring the standard deviation of the noise;

s6, when the failure times of vector solution test exceed the threshold limit, the EKF algorithm through parallel operation reinitializes the positioning result of the positioning algorithm in the current epoch so as to reduce the long-time accumulation of the relative motion tracking positioning error;

and S7, repeating the steps for all the adjacent epoch carrier observed values, and continuously providing a relative motion tracking and positioning result.

It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.