CN102819029B

CN102819029B - Supercompact combination satellite navigation receiver

Info

Publication number: CN102819029B
Application number: CN201210273837.4A
Authority: CN
Inventors: 高法钦
Original assignee: Zhejiang University of Technology ZJUT
Current assignee: Zhejiang University of Technology ZJUT
Priority date: 2012-08-03
Filing date: 2012-08-03
Publication date: 2014-01-22
Anticipated expiration: 2032-08-03
Also published as: CN102819029A

Abstract

The invention discloses an ultra-tight combination satellite navigation receiver, which includes a radio frequency front end, a signal preprocessor, a baseband prefilter and a navigation filter, a micro inertial navigation module, a pseudorange and pseudorange rate calculation module and an ephemeris solution Calculation and star selection calculation module, an adaptive filter is set between the baseband prefilter and the navigation filter, the output of the baseband prefilter is used as an input, and the adaptive filter is connected, and then the output of the adaptive filter is used as The observation data is input to the navigation filter for data fusion, which can significantly improve the tracking loop stability and tracking accuracy, thereby improving the adaptability of GNSS positioning to the environment and motion state. Further, the signal preprocessor, the baseband prefilter, the adaptive filter and the navigation filter, the pseudorange and pseudorange rate calculation modules constitute a vector tracking loop, and the navigation filter adopts a federated filter, which can improve the accuracy of the tracking loop. Stability and tracking accuracy, thereby improving performance such as GNSS positioning reliability and positioning accuracy.

Description

An ultra-tight combination satellite navigation receiver

技术领域 technical field

本发明涉及卫星导航技术领域，尤其涉及一种全球导航卫星系统(GNSS)与微惯导系统(MEMS-IMU)超紧组合的卫星导航接收机。The invention relates to the technical field of satellite navigation, in particular to a satellite navigation receiver with an ultra-tight combination of a Global Navigation Satellite System (GNSS) and a Micro Inertial Navigation System (MEMS-IMU).

背景技术 Background technique

全球导航卫星系统(GNSS)是一种以卫星为基础的无线电导航系统，能为陆、海、空的各类载体提供全天候、不间断、高精度、实时导航定位服务。目前，应用最广的全球导航卫星系统是美国的GPS，已经渗透到国民经济与日常生活的各个领域，如海上航行、城市交通管理、商业物流管理、船舶远洋导航、精密受时、大地测量、精细农业等。目前，我国也参与了将于近两年建成的Galileo系统的建设，并正在自主研发全球卫星定位系统Compass(北斗二代)，该系统去年底已经在我国及其周边地区提供定位服务。因此，研究Compass与GPS、Galileo等系统的多模组合导航技术必将成为国内未来一段时间内的研究重点。Global Navigation Satellite System (GNSS) is a satellite-based radio navigation system that can provide all-weather, uninterrupted, high-precision, real-time navigation and positioning services for various carriers on land, sea and air. At present, the most widely used global navigation satellite system is the GPS of the United States, which has penetrated into various fields of the national economy and daily life, such as maritime navigation, urban traffic management, commercial logistics management, ship ocean navigation, precise timing, geodesy, precision agriculture, etc. At present, my country has also participated in the construction of the Galileo system, which will be completed in the past two years, and is independently developing the global satellite positioning system Compass (the second generation of Beidou), which has provided positioning services in my country and its surrounding areas at the end of last year. Therefore, research on the multi-mode integrated navigation technology of Compass, GPS, Galileo and other systems will definitely become the focus of domestic research in the future.

GPS信号(扩频信号)到达地面接收机时已相当微弱，大约为-130dBmW，比接收机内部热噪声低20～30dB，特别的，在室内、城市、森林等复杂环境(本文中统称为室内环境)中，GPS接收信噪比更低，而室内环境恰是人类活动的主要环境之一。When the GPS signal (spread spectrum signal) arrives at the ground receiver, it is quite weak, about -130dBmW, which is 20-30dB lower than the internal thermal noise of the receiver. Especially, in complex environments such as indoors, cities, forests (collectively referred to as indoor environment), the signal-to-noise ratio of GPS reception is lower, and the indoor environment is just one of the main environments for human activities.

GNSS组合接收机可以充分利用各GNSS系统的卫星信号资源，因而可以提高系统可用性和定位输出的连续性。IMU(微机电惯性测量组件MEMS-IMU，以下简称为IMU)为自主式定位导航系统，基本不受环境的影响，且其动态性能较好。因此，将GNSS组合接收机与IMU系统组合，可以解决上述问题，提高定位的精度、定位连续性及服务保障能力。The GNSS combined receiver can make full use of the satellite signal resources of each GNSS system, thus improving system availability and continuity of positioning output. IMU (MEMS-IMU, hereinafter referred to as IMU) is an autonomous positioning and navigation system, which is basically not affected by the environment, and its dynamic performance is good. Therefore, combining the GNSS integrated receiver with the IMU system can solve the above problems and improve positioning accuracy, positioning continuity and service guarantee capabilities.

GNSS/IMU超紧组合导航理论的研究是很有必要的，是目前国内外研究的热点之一。目前，虽然国内外已经有相关文献研究GPS/IMU(惯导)超紧组合导航方式，在组合系统实现方案、建模等方面已做了一定研究工作，但超紧组合研究仍存在许多问题需要深入研究并解决，尤其是基于矢量跟踪的超紧组合导航方式，其工作的可靠性等方面仍面临许多问题迫切需要解决。为了提高组合导航系统性能，能够在民航、铁路交通等对可靠性要求较高领域中得到推广应用，本专利将给出一种GNSS/IMU超紧组合导航矢量跟踪接收机基带信息融合处理方案，解决上述关键科学问题。The research on the theory of GNSS/IMU ultra-tight integrated navigation is very necessary, and it is one of the hot research topics at home and abroad. At present, although there are relevant literatures at home and abroad to study the GPS/IMU (inertial navigation) ultra-tight integrated navigation method, and some research work has been done on the implementation scheme and modeling of the integrated system, there are still many problems in the ultra-tight combination research that need to be solved. In-depth research and solutions, especially the super-tight integrated navigation method based on vector tracking, the reliability of its work, etc. still face many problems that need to be solved urgently. In order to improve the performance of the integrated navigation system, it can be popularized and applied in fields with high reliability requirements such as civil aviation and railway transportation. This patent will provide a GNSS/IMU ultra-tight integrated navigation vector tracking receiver baseband information fusion processing scheme. Solve the key scientific problems mentioned above.

发明内容 Contents of the invention

本发明的目的是为了克服现有GNSS接收机灵敏度低和容易受干扰而导致卫星信号失锁，因而不能给出定位结果的弊端，通过设计与微惯导的超紧组合方案、改进矢量跟踪方案、采取措施提高GNSS组合接收机跟踪精度和稳定性能，提高GNSS定位可靠性和定位精度等性能。The purpose of the present invention is to overcome the low sensitivity of existing GNSS receivers and the disadvantages that satellite signals are out of lock due to interference, and thus cannot provide positioning results, by designing an ultra-tight combination scheme with micro-inertial navigation and improving the vector tracking scheme , Take measures to improve the tracking accuracy and stability of GNSS integrated receivers, and improve the performance of GNSS positioning reliability and positioning accuracy.

本发明提供了一种超紧组合导航卫星接收机，包括射频前端，信号预处理器，基带预滤波器和导航滤波器，以及微惯导模块，伪距和伪距率计算模块，星历解算与选星计算模块，其中信号预处理模块包括相关器和本地信号生成器，所述基带预滤波器与导航滤波器之间设置有自适应滤波器。The invention provides an ultra-tight integrated navigation satellite receiver, including a radio frequency front end, a signal preprocessor, a baseband prefilter and a navigation filter, a micro-inertial navigation module, a pseudorange and pseudorange rate calculation module, an ephemeris solution Calculation and star selection calculation module, wherein the signal preprocessing module includes a correlator and a local signal generator, and an adaptive filter is set between the baseband prefilter and the navigation filter.

所述本地信号生成器，相关器，基带预滤波器，自适应滤波器和导航滤波器，伪距和伪距率计算模块构成矢量跟踪环路。伪距和伪距率计算模块接收微惯导力学编排模块的输出信息、导航滤波器输出的时钟误差等信息、以及星历解算与选星计算模块的GNSS星历参数，生成延迟调整信号，输入本地信号生成器，调整本地伪码和载波，本方案即可提高跟踪环路稳定性，又可提高跟踪环路精度。The local signal generator, correlator, baseband pre-filter, adaptive filter, navigation filter, pseudo-range and pseudo-range rate calculation module constitute a vector tracking loop. The pseudo-range and pseudo-range rate calculation module receives the output information of the micro-inertial navigation mechanics arrangement module, the clock error output by the navigation filter, and the GNSS ephemeris parameters of the ephemeris calculation and star selection calculation module, and generates a delay adjustment signal, Input the local signal generator, adjust the local pseudo code and carrier, this scheme can improve the stability of the tracking loop and improve the accuracy of the tracking loop.

所述导航滤波器为联邦滤波器，包括主滤波器和至少一个子滤波器。所述联邦滤波器的子滤波器包括纯GNSS模式下的GNSS滤波器，超紧组合模式下的GNSS/IMU组合导航滤波器和纯IMU微惯导滤波器。用联邦滤波器作为导航滤波器，引入矢量跟踪环路，将定位信息也作为矢量跟踪的输入，有利于更精确的恢复载波和伪码。The navigation filter is a federated filter, including a main filter and at least one sub-filter. The sub-filters of the federated filter include a GNSS filter in a pure GNSS mode, a GNSS/IMU integrated navigation filter in an ultra-tight combination mode and a pure IMU micro inertial navigation filter. The federated filter is used as the navigation filter, the vector tracking loop is introduced, and the positioning information is also used as the input of the vector tracking, which is conducive to more accurate recovery of the carrier and pseudo code.

进一步地，所述超紧组合卫星导航接收机包括至少一路导航信号接收通道，每个导航信号接收通道都设置有信号预处理器和基带预滤波器，每一路基带预滤波器都通过一个自适应滤波器后接入导航滤波器。Further, the ultra-tight combination satellite navigation receiver includes at least one navigation signal receiving channel, and each navigation signal receiving channel is provided with a signal preprocessor and a baseband prefilter, and each baseband prefilter is passed through an adaptive Access the navigation filter after the filter.

所述的自适应滤波器的状态方程为：The state equation of the adaptive filter is:

$\frac{d d}{dt dt} [\begin{matrix} δρ δρ \\ δ δ \overset{\cdot &Center Dot;}{ρ ρ} \\ δ δ \overset{\cdot &Center Dot; \cdot &Center Dot;}{ρ ρ} \\ δN δ N \\ {Δ Δ}_{ion ion} \\ {\overset{\cdot &Center Dot;}{Δ Δ}}_{ion ion} \end{matrix}] = = [\begin{matrix} δ δ \overset{\cdot &Center Dot;}{ρ ρ} \\ δ δ \overset{\cdot &Center Dot; \cdot &Center Dot;}{ρ ρ} \\ 00 \\ 00 \\ {\overset{\cdot &Center Dot;}{Δ Δ}}_{ion ion} \\ 00 \end{matrix}] + + [\begin{matrix} 00 \\ 00 \\ {w w}_{a a} \\ 00 \\ 00 \\ {w w}_{ion ion} \end{matrix}]$

观测方程为：The observation equation is:

$[\begin{matrix} δρ δρ \\ δ δ \overset{\cdot &Center Dot;}{ρ ρ} \\ δ δ \overset{\cdot &Center Dot; \cdot &Center Dot;}{ρ ρ} \\ δN δ N \\ {Δ Δ}_{ion ion} \\ {\overset{\cdot &Center Dot;}{Δ Δ}}_{ion ion} \end{matrix}] = = [\begin{matrix} δρ δρ \\ δ δ \overset{\cdot &Center Dot;}{ρ ρ} \\ δ δ \overset{\cdot &Center Dot; \cdot &Center Dot;}{ρ ρ} \\ δN δ N \\ {Δ Δ}_{ion ion} \\ {\overset{\cdot &Center Dot;}{Δ Δ}}_{ion ion} \end{matrix}] + + [\begin{matrix} {w w}_{ρ ρ} \\ {w w}_{\overset{\cdot &Center Dot;}{ρ ρ}} \\ {w w}_{\overset{\cdot &Center Dot; \cdot &Center Dot;}{ρ ρ}} \\ {w w}_{N N} \\ {w w}_{i i} \\ {w w}_{io io} \end{matrix}]$

其中δρ为伪距残差，

为伪距率残差，

为伪距加速度残差，δN为载波相位整周残差，Δ_ion为电离层延时，为电离层延时率，w_a为伪距加速度过程噪声，w_ion为电离层噪声，w_ρ、

w_N、W_i、w_io分别为伪距、伪距率、伪距加速度、载波相位整周模糊度、电离层延时、电离层延时率的观测噪声。where δρ is the pseudorange residual,

is the pseudorange rate residual,

is the pseudorange acceleration residual, δN is the carrier phase integer residual, _Δion is the ionospheric delay, is the ionospheric delay rate, w _a is the pseudorange acceleration process noise, w _ion is the ionospheric noise, w _ρ ,

w _N , W _i , and w _io are observation noises of pseudorange, pseudorange rate, pseudorange acceleration, carrier phase integer ambiguity, ionospheric delay, and ionospheric delay rate, respectively.

进一步地，所述自适应滤波器为模糊逻辑自适应滤波器，包括模糊逻辑算法模块和卡尔曼滤波器。所述模糊逻辑算法模块接收伪距滤波新息平均值和均方差，输出为驱动噪声方差加权系数，通过实时调整所述卡尔曼滤波器的驱动噪声方差的参数值，实现自适应滤波。所述模糊逻辑算法模块包括：Further, the adaptive filter is a fuzzy logic adaptive filter, including a fuzzy logic algorithm module and a Kalman filter. The fuzzy logic algorithm module receives the mean value and mean square error of pseudorange filter innovation, and outputs the driving noise variance weighting coefficient, and realizes adaptive filtering by adjusting the parameter value of the driving noise variance of the Kalman filter in real time. The fuzzy logic algorithm module includes:

隶属度仿真单元，用于确定输入的伪距滤波新息平均值和均方差的隶属度函数，以及输出的驱动噪声方差加权系数的隶属度函数；The membership degree simulation unit is used to determine the membership degree function of the input pseudorange filtering innovation mean value and mean square error, and the membership degree function of the output driving noise variance weighting coefficient;

模糊推理规则单元，用于设置模糊推理规则；A fuzzy inference rule unit, configured to set fuzzy inference rules;

噪声方差加权单元，用于根据所述隶属度函数和模糊推理规则解模糊计算得到驱动噪声方差加权系数。The noise variance weighting unit is used to defuzzify and calculate the driving noise variance weighting coefficient according to the membership function and the fuzzy inference rule.

本发明模糊逻辑算法模块根据对特定导航应用系统的仿真分析和实际导航数据分析，确定模糊逻辑算法模块输入/输出量的模糊划分和隶属度函数，输入/输出量的函数关系，以及模糊逻辑推理规则，计算输入的伪距滤波新息平均值和均方差的隶属度函数值，以及驱动噪声方差加权系数的隶属度函数值，并利用模糊逻辑规则计算出驱动方差加权系数。采用上述模糊逻辑算法实时调整卡尔曼滤波器驱动噪声方差加权系数，使的滤波器参数随着载体机动情况而调整，提高了滤波估计精度和滤波稳定性，因而有助于提高跟踪环路的跟踪精度和稳定性。The fuzzy logic algorithm module of the present invention determines the fuzzy division and membership function of the input/output quantity of the fuzzy logic algorithm module, the functional relationship of the input/output quantity, and fuzzy logic reasoning according to the simulation analysis of the specific navigation application system and the actual navigation data analysis The rule calculates the membership function value of the input pseudo-range filter innovation average value and mean square error, and the membership function value of the driving noise variance weighting coefficient, and calculates the driving variance weighting coefficient using fuzzy logic rules. The above-mentioned fuzzy logic algorithm is used to adjust the Kalman filter driving noise variance weighting coefficient in real time, so that the filter parameters can be adjusted according to the maneuvering situation of the carrier, which improves the filtering estimation accuracy and filtering stability, thus helping to improve the tracking of the tracking loop precision and stability.

本发明公开的超紧组合卫星导航接收机，通过采用自适应滤波器，以及矢量跟踪环路结构下导航滤波器采用联邦滤波器等措施，提高GNSS定位的可靠性和定位精度等性能，并且，当信号受到遮挡、存在一定的环境干扰时也可稳定的给出定位结果。The ultra-tight combination satellite navigation receiver disclosed by the present invention improves the performance of GNSS positioning reliability and positioning accuracy by adopting adaptive filters, and the navigation filter under the vector tracking loop structure using federated filters, etc., and, When the signal is blocked or there is certain environmental interference, it can also provide stable positioning results.

附图说明 Description of drawings

图1是本发明超紧导航卫星接收机的结构示意图；Fig. 1 is the structural representation of ultra-tight navigation satellite receiver of the present invention;

图2是本发明自适应滤波器结构示意图；Fig. 2 is a structural schematic diagram of an adaptive filter of the present invention;

图3是本发明自适应滤波器模糊逻辑算法模块的结构示意图；Fig. 3 is the structural representation of adaptive filter fuzzy logic algorithm module of the present invention;

图4是模糊逻辑算法输入量滤波新息平均值隶属度函数示意图；Fig. 4 is a schematic diagram of the membership degree function of the fuzzy logic algorithm input quantity filtering innovation mean value;

图5是模糊逻辑算法输入量滤波新息均方差隶属度函数示意图；Fig. 5 is a schematic diagram of the membership degree function of the fuzzy logic algorithm input quantity filtering innovation mean square error;

图6是模糊逻辑算法输出量隶属度函数示意图；Fig. 6 is a schematic diagram of the membership degree function of the fuzzy logic algorithm output;

图7是联邦滤波器结构示意图。Fig. 7 is a schematic diagram of a federated filter structure.

具体实施方式 Detailed ways

下面结合附图和实施例对本发明技术方案做进一步详细说明，以下实施例不构成对本发明的限定。The technical solution of the present invention will be described in further detail below in conjunction with the accompanying drawings and embodiments, and the following embodiments do not constitute a limitation of the present invention.

基于矢量跟踪的GNSS/IMU超紧组合卫星导航接收机如图1所示，包括依次相连的射频前端1，信号预处理器2，基带预滤波器3，自适应滤波器4和导航滤波器5，以及微惯导模块6，伪距和伪距率计算模块7，星历解算与选星计算模块8。The GNSS/IMU ultra-tight integrated satellite navigation receiver based on vector tracking is shown in Fig. 1, including the RF front-end 1, signal preprocessor 2, baseband prefilter 3, adaptive filter 4 and navigation filter 5 connected in sequence , and micro-inertial navigation module 6, pseudo-range and pseudo-range rate calculation module 7, ephemeris calculation and star selection calculation module 8.

其中信号预处理器2包括相关器21和本地信号生成器22，相关器21累积输出的I/Q信号作为基带预滤波器3的观测输入，本地信号生成器22生成本地载波和伪码。The signal preprocessor 2 includes a correlator 21 and a local signal generator 22. The I/Q signal accumulated and output by the correlator 21 is used as the observation input of the baseband prefilter 3. The local signal generator 22 generates a local carrier and pseudo code.

其中微惯导模块6包括微惯导力学编排模块61和微惯导IMU模块62，微惯导IMU62模块输出的载体加速度、角速度信息，输入到微惯刀力学编排模块61，进行解算，得到载体的位置、速度和姿态信息。微惯刀力学编排模块61输出端一方面连接伪距和伪距率计算模块7，一方面连接导航定位结果输出端，输出导航定位结果。Wherein the micro-inertial navigation module 6 includes a micro-inertial navigation mechanics arrangement module 61 and a micro-inertial navigation IMU module 62, and the carrier acceleration and angular velocity information output by the micro-inertial navigation IMU62 module are input to the micro-inertial knife mechanics arrangement module 61 for calculation, and obtain The position, velocity and attitude information of the carrier. On the one hand, the output end of the micro-knife mechanics arrangement module 61 is connected to the pseudo-range and pseudo-range rate calculation module 7, and on the other hand, it is connected to the output end of the navigation positioning result to output the navigation positioning result.

如图1所示，本地信号生成器22，相关器21，基带预滤波器3，自适应滤波器4和导航滤波器5，伪距和伪距率计算模块7构成矢量跟踪环路。伪距和伪距率计算模块7接收微惯导力学编排模块61的输出信息，导航滤波器5输出的接收机时钟误差以及星历解算与选星计算模块8的GNSS星历参数，生成延迟调整信号，输入到本地信号生成器22，调整本地伪码和载波。As shown in Fig. 1, local signal generator 22, correlator 21, baseband pre-filter 3, adaptive filter 4 and navigation filter 5, pseudorange and pseudorange rate calculation module 7 constitute a vector tracking loop. The pseudo-range and pseudo-range rate calculation module 7 receives the output information of the micro-inertial navigation dynamics arrangement module 61, the receiver clock error output by the navigation filter 5, and the GNSS ephemeris parameters of the ephemeris calculation and star selection calculation module 8, and generates a delay The adjusted signal is input to the local signal generator 22, and the local pseudo code and carrier are adjusted.

来自卫星射频前端1的卫星信号经过相关器21后，输出同相(I)、正交相(Q)信号，进入基带预滤波器3进行预滤波，输出为伪距误差、伪距率误差和伪距加速度误差，以及电离层延迟误差，电离层延迟误差变化率信息。基带预滤波器3的作用是估计跟踪误差、并降低导航计算频率(由250～1000Hz降低到1～10Hz)和计算量。After the satellite signal from the satellite radio frequency front end 1 passes through the correlator 21, the output in-phase (I), quadrature phase (Q) signal enters the baseband pre-filter 3 for pre-filtering, and the output is pseudo-range error, pseudo-range rate error and pseudo-range error. Acceleration error, ionospheric delay error, ionospheric delay error change rate information. The function of the baseband pre-filter 3 is to estimate the tracking error, and reduce the navigation calculation frequency (from 250-1000 Hz to 1-10 Hz) and calculation amount.

具体地，本发明基带预滤波器3接收信号预处理器2输出的伪码相位误差和载波相位误差信号，为每个通道建立预滤波器，通过线性Kalman滤波器估计系统的状态量，预滤波器状态方程为：Specifically, the baseband pre-filter 3 of the present invention receives the pseudo code phase error and the carrier phase error signal output by the signal pre-processor 2, establishes a pre-filter for each channel, estimates the state quantity of the system through a linear Kalman filter, and pre-filters The state equation of the device is:

$\frac{d d}{dt dt} [\begin{matrix} δρ δρ \\ δ δ \overset{\cdot \cdot}{ρ ρ} \\ δ δ \overset{\cdot \cdot \cdot \cdot}{ρ ρ} \\ δN δN \\ {Δ Δ}_{ion ion} \\ {\overset{\cdot \cdot}{Δ Δ}}_{ion ion} \end{matrix}] = = [\begin{matrix} δ δ \overset{\cdot \cdot}{ρ ρ} \\ δ δ \overset{\cdot \cdot \cdot &Center Dot;}{ρ ρ} \\ 00 \\ 00 \\ {\overset{\cdot \cdot}{Δ Δ}}_{ion ion} \\ 00 \end{matrix}] + + [\begin{matrix} 00 \\ 00 \\ {w w}_{a a} \\ 00 \\ 00 \\ {w w}_{ion ion} \end{matrix}]$

其中，δρ为伪距残差，

为伪距率残差，

为伪距加速度残差，δN为载波相位整周残差，Δ_ion为电离层延时，为电离层延时率，w_a为伪距加速度过程噪声，w_ion为电离层噪声。Among them, δρ is the pseudorange residual,

is the pseudorange rate residual,

is the pseudorange acceleration residual, δN is the carrier phase integer residual, _Δion is the ionospheric delay, is the ionospheric delay rate, w _a is the pseudorange acceleration process noise, and w _ion is the ionospheric noise.

预滤波器的观测方程为：The observation equation of the pre-filter is:

$[\begin{matrix} Δρ Δρ \\ Δφ Δφ \end{matrix}] = = [\begin{matrix} δρ δρ + + {Δ Δ}_{ion ion} \\ δρ δρ - - {Δ Δ}_{ion ion} + + δN δ N \end{matrix}]$

其中，Δρ为伪距残差测量值，Δφ为载波相位残差测量值。Among them, Δρ is the pseudorange residual measurement value, and Δφ is the carrier phase residual measurement value.

本发明为了提高GNSS系统在高动态情况下的定位性能，不是将基带预滤波器3输出直接输入至导航滤波器5进行数据融合，而是，先进行自适应滤波，将基带预滤波器3输出作为输入，接入自适应滤波器4，然后，将自适应滤波4的输出作为观测量输入至导航滤波器5进行数据融合，自适应滤波器4对观测输入信号的剧烈变化有自适应能力，可显著提高跟踪环路稳定性，由此提高GNSS定位对环境和运动状态突变的适应能力。In order to improve the positioning performance of the GNSS system under high dynamic conditions, the present invention does not directly input the output of the baseband pre-filter 3 to the navigation filter 5 for data fusion, but performs adaptive filtering first, and outputs the output of the baseband pre-filter 3 As an input, the adaptive filter 4 is connected, and then, the output of the adaptive filter 4 is input to the navigation filter 5 as observations for data fusion. The adaptive filter 4 has the ability to adapt to the drastic changes of the observed input signal, The stability of the tracking loop can be significantly improved, thereby improving the adaptability of GNSS positioning to sudden changes in the environment and motion state.

具体地，每个通道的自适应滤波器4状态方程为：Specifically, the adaptive filter 4 state equation for each channel is:

$\frac{d d}{dt dt} [\begin{matrix} δρ δρ \\ δ δ \overset{\cdot \cdot}{ρ ρ} \\ δ δ \overset{\cdot \cdot \cdot \cdot}{ρ ρ} \\ δN δN \\ {Δ Δ}_{ion ion} \\ {\overset{\cdot &Center Dot;}{Δ Δ}}_{ion ion} \end{matrix}] = = [\begin{matrix} δ δ \overset{\cdot \cdot}{ρ ρ} \\ δ δ \overset{\cdot &Center Dot; \cdot \cdot}{ρ ρ} \\ 00 \\ 00 \\ {\overset{\cdot \cdot}{Δ Δ}}_{ion ion} \\ 00 \end{matrix}] + + [\begin{matrix} 00 \\ 00 \\ {w w}_{a a} \\ 00 \\ 00 \\ {w w}_{ion ion} \end{matrix}]$

其中，δρ为伪距残差，

为伪距率残差，

为伪距加速度残差，δN为载波相位整周残差，Δ_ion为电离层延时，

为电离层延时率，w_a为伪距加速度过程噪声，w_ion为电离层噪声。Among them, δρ is the pseudorange residual,

is the pseudorange rate residual,

is the pseudorange acceleration residual, δN is the carrier phase integer residual, _Δion is the ionospheric delay,

is the ionospheric delay rate, w _a is the pseudorange acceleration process noise, and w _ion is the ionospheric noise.

观测方程为：The observation equation is:

$[\begin{matrix} δρ δρ \\ δ δ \overset{\cdot &Center Dot;}{ρ ρ} \\ δ δ \overset{\cdot &Center Dot; \cdot &Center Dot;}{ρ ρ} \\ δN δN \\ {Δ Δ}_{ion ion} \\ {\overset{\cdot &Center Dot;}{Δ Δ}}_{ion ion} \end{matrix}] = = [\begin{matrix} δρ δρ \\ δ δ \overset{\cdot &Center Dot;}{ρ ρ} \\ δ δ \overset{\cdot &Center Dot; \cdot &Center Dot;}{ρ ρ} \\ δN δ N \\ {Δ Δ}_{ion ion} \\ {\overset{\cdot \cdot}{Δ Δ}}_{ion ion} \end{matrix}] + + [\begin{matrix} {w w}_{ρ ρ} \\ {w w}_{\overset{\cdot \cdot}{ρ ρ}} \\ {w w}_{\overset{\cdot \cdot \cdot \cdot}{ρ ρ}} \\ {w w}_{N N} \\ {w w}_{i i} \\ {w w}_{io io} \end{matrix}]$

其中，w_ρ、

w_N、w_i、w_io分别为伪距、伪距率、伪距加速度、载波相位整周模糊度、电离层延时、电离层延时率等参量的观测噪声。Among them, w _ρ ,

w _N , w _i , and w _io are observation noises of parameters such as pseudorange, pseudorange rate, pseudorange acceleration, carrier phase integer ambiguity, ionospheric delay, and ionospheric delay rate, respectively.

本发明的自适应滤波器4为模糊逻辑自适应滤波器，如图2所示，包括模糊逻辑算法模块41与卡尔曼滤波器42，其中模糊逻辑算法模块41的输入为伪距滤波新息平均值的绝对值和均方差，模糊逻辑算法模块41的输出为滤波器驱动噪声方差加权系数，通过实时调整卡尔曼滤波器42驱动噪声方差的参数值，实现自适应滤波。Adaptive filter 4 of the present invention is fuzzy logic adaptive filter, as shown in Figure 2, comprises fuzzy logic algorithm module 41 and Kalman filter 42, and wherein the input of fuzzy logic algorithm module 41 is pseudorange filtering innovation average The absolute value and mean square error of the value, the output of the fuzzy logic algorithm module 41 is the filter driving noise variance weighting coefficient, and the adaptive filtering is realized by adjusting the parameter value of the driving noise variance of the Kalman filter 42 in real time.

模糊逻辑算法模块41通过对系统的大量仿真，并经总结获得输入模糊量隶属度函数、模糊逻辑推理规则，最终经过模糊逻辑算法计算出驱动噪声方差加权系数。如图3所示，模糊逻辑算法模块41包括隶属度仿真单元411，模糊推理规则单元412和噪声方差加权单元413。The fuzzy logic algorithm module 41 obtains the input fuzzy quantity membership function and fuzzy logic reasoning rules through a large number of system simulations, and finally calculates the driving noise variance weighting coefficient through the fuzzy logic algorithm. As shown in FIG. 3 , the fuzzy logic algorithm module 41 includes a membership simulation unit 411 , a fuzzy inference rule unit 412 and a noise variance weighting unit 413 .

隶属度仿真单元411通过仿真分析获得模糊逻辑算法输入/输出论域划分上的隶属度函数，用于确定输入的伪距滤波新息平均值和均方差的隶属度函数，以及输出的驱动噪声方差加权系数的隶属度函数。The membership degree simulation unit 411 obtains the membership degree function on the input/output discourse domain division of the fuzzy logic algorithm through simulation analysis, and is used to determine the membership degree function of the input pseudorange filter innovation average value and mean square error, and the output driving noise variance The membership function of the weighting coefficient.

如图4所示，具体地，无机动时，滤波新息平均值绝对值小于10米的隶属度几乎等于1，而且，接近0米处的隶属度最大，随着滤波新息平均值绝对值的增大，隶属度逐渐减小。因此，本发明将[0，10]米划分为一个区间，对应模糊集‘小’的隶属度，表示为

在0米处最大，随着滤波新息平均值绝对值的增大，模糊集‘小’的隶属度逐渐减小。As shown in Figure 4, specifically, when there is no maneuvering, the membership degree of the filter innovation average value less than 10 meters is almost equal to 1, and the membership degree close to 0 meters is the largest, as the filter innovation average value absolute value increases, the degree of membership gradually decreases. Therefore, the present invention divides [0,10] meters into an interval, corresponding to the degree of membership of the fuzzy set 'small', expressed as

Maximum at 0 meters, as the absolute value of the mean value of the filtered innovation increases, the degree of membership of the fuzzy set is 'small' slowing shrieking.

当滤波新息平均值绝对值大于60米时，说明肯定出现了大机动；当新息平均值大于10米且小于60米时，仍有可能发生大于4g的大机动；发生大于4g的大机动时，滤波新息平均值绝对值小于10米的隶属度很小。基于这一分析结论，本发明将[10～150]米划分为一个区间，作为模糊集‘大’的隶属度

的定义域，在区间[10～60]米内，随着滤波新息平均值绝对值的增大，隶属度

逐渐增大；滤波新息平均值绝对值大于60米时其隶属度

恒等于1。When the absolute value of the filtered innovation average value is greater than 60 meters, it means that there must be a large maneuver; when the average innovation value is greater than 10 meters and less than 60 meters, a large maneuver greater than 4g may still occur; a large maneuver greater than 4g may occur When , the membership degree of filter innovation average value less than 10 meters is very small. Based on this analysis conclusion, the present invention divides [10～150] meters into an interval, as the membership degree of fuzzy set 'big'

The definition domain of , within the interval [10-60] meters, with the increase of the absolute value of the mean value of filter innovation, the degree of membership

Gradually increase; the membership degree when the absolute value of the filter innovation average value is greater than 60 meters

Constantly equal to 1.

小于4个g的加减速机动、圆周飞行机动、爬升/俯冲机动时的滤波新息平均值绝对值一般在区间[0～60]米取值；基于对某歼击机长航飞行时的分析可知其加/减速常可达12米/秒2左右，仿真结果表明，此时滤波新息平均值绝对值在10米附近的隶属度最大，越偏离10米隶属度越小。根据这一分析，本发明将[0～60]米划分为一个区间，对应模糊集‘中’的隶属度取值等于10米时，属于‘较小’的隶属度

最大；随着偏离10米，属于模糊集‘较小’的隶属度逐渐减小。The absolute value of the mean value of filter innovation during acceleration and deceleration maneuvers, circular flight maneuvers, and climb/dive maneuvers of less than 4 g is generally taken in the interval [0-60] meters; Acceleration/deceleration can often reach about 12 m/s 2 , and the simulation results show that the membership degree of the absolute value of the average value of filter innovation at this time is the largest near 10 meters, and the membership degree is smaller as it deviates from 10 meters. According to this analysis, the present invention divides [0～60] meters into an interval, corresponding to the membership degree of the fuzzy set '中' When the value is equal to 10 meters, it belongs to the 'smaller' degree of membership

Maximum; with a deviation of 10 meters, the degree of membership in the fuzzy set is 'smaller' slowing shrieking.

需要说明的是，本发明中，采用简单函数三角函数近似代替各语言变量的隶属度函数，虽然和实际存在一定的差距，但不会对运算结果产生较大影响。It should be noted that, in the present invention, the simple function trigonometric function is used to approximate the membership function of each linguistic variable. Although there is a certain gap with the actual situation, it will not have a great impact on the calculation result.

同理可得模糊逻辑算法另一输入，滤波新息均方差论域上的各模糊集隶属度

如图5所示；以及模糊逻辑算法输出α隶属度

如图6所示。In the same way, another input of the fuzzy logic algorithm can be obtained, which is the membership degree of each fuzzy set on the domain of mean square error of filter innovation

As shown in Figure 5; and fuzzy logic algorithm output α degree of membership

As shown in Figure 6.

模糊推理规则单元412经仿真分析，得到具有完备性的模糊推理规则。The fuzzy inference rule unit 412 obtains complete fuzzy inference rules through simulation analysis.

具体地，模糊推理规则表达滤波新息平均值绝对值、均方差与滤波器驱动噪声方差加权系数在取值上的对应关系。模糊推理规则由规则前件(或称为‘条件’)和规则结论两部分组成，滤波新息平均值绝对值、均方差对应规则前件，滤波器驱动噪声方差加权系数对应规则结论部分。Specifically, the fuzzy inference rules express the corresponding relationship between the absolute value of the average value of the filter innovation, the mean square error, and the weighting coefficient of the variance of the filter driving noise. Fuzzy inference rules are composed of rule antecedents (or called 'conditions') and rule conclusions. The absolute value and mean square error of the filter innovation mean value correspond to the rule antecedents, and the filter driving noise variance weighting coefficient corresponds to the rule conclusion part.

模糊推理规则形式为：The form of fuzzy inference rules is:

IF

and

then

IF

and

then

其中，(p₁，p₂)∈{({s，m，b}，{s，m，b})}，k∈{s，m，b}，上标s、m、b分别表示各模糊量，有关各模糊量参见图4-图6。上述规则的含义是：如果输入平均值绝对值Δρ₁属于模糊集

(

为其隶属度)、均方差Δρ₂属于模糊集

(

为其隶属度)，那么，输出(方差Q的)调整系数α属于模糊集B^(k)(

为其隶属度)。规则的条件部分为(

and)，结论部分为

该规则可简写为：(p₁，p₂)→k。Among them, (p ₁ , p ₂ ) ∈ {({s, m, b}, {s, m, b})}, k ∈ {s, m, b}, superscript s, m, b represent each Blur amount, see Figure 4-Figure 6 for each blur amount. The meaning of the above rules is: if the absolute value of the input mean value Δρ ₁ belongs to the fuzzy set

(

Its membership degree), the mean square error Δρ ₂ belongs to the fuzzy set

(

Its membership degree), then, the output (of the variance Q) adjustment coefficient α belongs to the fuzzy set B ^(k) (

for its degree of membership). The condition part of the rule is (

and ), the conclusion part is

This rule can be abbreviated as: (p ₁ , p ₂ )→k.

推理规则的举例说明：经仿真和分析可知，如果滤波新息平均值绝对值“较大”、均方差“偏大”，那么，当前一段时间有加速度大于2g的等效机动发生，需要增大滤波器驱动噪声均方差Q，其调整系数α取值为“小”，则可得推理规则如下：Example of inference rules: After simulation and analysis, it can be seen that if the absolute value of the mean value of the filter innovation is "large" and the mean square error is "large", then there is an equivalent maneuver with an acceleration greater than 2g in the current period of time, and it is necessary to increase The mean square error Q of filter driving noise, and the value of its adjustment coefficient α is "small", then the inference rules can be obtained as follows:

IF

and

then

简写为：(b，b)→sIF

and

then

Abbreviated as: (b, b) → s

经仿真分析，得到具有完备性的9条模糊推理规则如下：After simulation analysis, nine complete fuzzy inference rules are obtained as follows:

(s，s)→m (m，s)→s (b，s)→s (s，m)→m (m，m)→s (b，m)→s(s, s)→m (m, s)→s (b, s)→s (s, m)→m (m, m)→s (b, m)→s

(s，b)→b (m，b)→s (b，b)→s (1)(s,b)→b (m,b)→s (b,b)→s (1)

将结论相同的模糊推理规则合并为1条规则，合并后的模糊推理规则集中共有3条，其结论部分分别为“小、中、大”。设这3条规则的前件隶属度取值分别为μ^(s)、μ^(m)、μ^(b)，根据9条模糊推理规则集，可得：The fuzzy inference rules with the same conclusion are merged into one rule, and there are 3 rules in the merged fuzzy inference rule set, and the conclusion parts are "small, medium and large". Assuming the antecedent membership degrees of these three rules are respectively μ ^(s) , μ ^(m) and μ ^(b) , according to the nine fuzzy inference rule sets, we can get:

μ^(s)＝μ_(m，s)+μ_(m，m)+μ_(m，b)+μ_(b，s)+μ_(b，m)+μ_(b，b)； (2)μ ^(s) = μ _{(m, s)} + μ _{(m, m)} + μ _{(m, b)} + μ _{(b, s)} + μ _{(b, m)} + μ _{(b, b)} ; (2)

μ^(m)＝μ_(s，s)+μ_(s，m)μ^(b)＝μ_(s，b)；μ ^(m) = μ _{(s, s)} + μ _{(s, m)} μ ^(b) = μ _{(s, b)} ;

其中，μ_(m，s)表示规则(m，s)→s的前件隶属度，其计算公式如下：Among them, μ _{(m, s)} represents the membership degree of the antecedent of the rule (m, s) → s, and its calculation formula is as follows:

${μ μ}_{((m m,, s the s))} = = {μ μ}_{{A A}_{11}}^{m m} (({Δρ Δρ}_{11})) \times \times {μ μ}_{{A A}_{22}}^{s the s} (({Δρ Δρ}_{22})) - - - - - - ((33))$

需要说明的是，其它规则前件的隶属度的计算参考上式，在此不再赘述。It should be noted that the calculation of the degree of membership of the antecedents of other rules refers to the above formula, which will not be repeated here.

噪声方差加权单元413用于根据隶属度函数和模糊推理规则解模糊计算得到驱动噪声方差加权系数α，计算表达式为：The noise variance weighting unit 413 is used to defuzzify and calculate the drive noise variance weighting coefficient α according to the membership function and fuzzy inference rules, and the calculation expression is:

$α α = = \frac{{α α}_{00} \times \times {μ μ}^{((s the s))} + + {α α}_{11} \times \times {μ μ}^{((m m))} + + {α α}_{22} \times \times {μ μ}^{((b b))}}{{μ μ}^{((s the s))} + + {μ μ}^{((m m))} + + {μ μ}^{((b b))}} - - - - - - ((44))$

其中α₀，α₁，α₂分别为模糊逻辑算法输出隶属度在模糊集为小，中，大三种情况下隶属度最大时对应α的取值，对应到图6，分别为0.7，1和1.4，上述计算结果即为模糊逻辑算法的最终输出的滤波器驱动噪声方差加权系数α。Among them, α ₀ , α ₁ , and α ₂ are the values of α corresponding to the maximum membership degree when the fuzzy logic algorithm output membership degree is small, medium and large respectively, corresponding to Figure 6, which are 0.7 and 1 respectively and 1.4, the above calculation result is the filter driving noise variance weighting coefficient α of the final output of the fuzzy logic algorithm.

本发明模糊逻辑算法的一个实施例，设模糊逻辑算法的输入为(5，80)，即平均值绝对值等于5米，均方差为80米²，由式(2)计算可得：An embodiment of the fuzzy logic algorithm of the present invention, the input of the fuzzy logic algorithm is established as (5, 80), that is, the mean absolute value equals 5 meters, and the mean square error is 80 meters , calculated by formula ( ² ):

${μ μ}^{((s the s))} = = {μ μ}_{((m m,, s the s))} + + {μ μ}_{((m m,, m m))} + + {μ μ}_{((m m,, b b))} + + {μ μ}_{((b b,, s the s))} + + {μ μ}_{((b b,, m m))} + + {μ μ}_{((b b,, b b))} = = {μ μ}_{{A A}_{11}}^{m m} ((55)) \times \times {μ μ}_{{A A}_{22}}^{s the s} ((8080)) + + {μ μ}_{{A A}_{11}}^{m m} ((55)) \times \times {μ μ}_{{A A}_{22}}^{m m} ((8080))$

$+ + {μ μ}_{{A A}_{11}}^{m m} ((55)) \times \times {μ μ}_{{A A}_{22}}^{b b} ((8080)) + + {μ μ}_{{A A}_{11}}^{b b} ((55)) \times \times {μ μ}_{{A A}_{22}}^{s the s} ((8080)) + + {μ μ}_{{A A}_{11}}^{b b} ((55)) \times \times {μ μ}_{{A A}_{22}}^{m m} ((8080)) + + {μ μ}_{{A A}_{11}}^{b b} ((55)) \times \times {μ μ}_{{A A}_{22}}^{b b} ((8080))$

$= = 0.5 0.5 \times \times 0.79 0.79 + + 0.5 0.5 \times \times 0.21 0.21 + + 0.5 0.5 \times \times 00 + + 00 \times \times 0.79 0.79 + + 00 \times \times 0.21 0.21 + + 00 \times \times 00 = = 0.5 0.5$

${μ μ}^{((m m))} = = {μ μ}_{((s the s,, s the s))} + + {μ μ}_{((s the s,, m m))} = = {μ μ}_{{A A}_{11}}^{s the s} ((55)) \times \times {μ μ}_{{A A}_{22}}^{s the s} ((8080)) + + {μ μ}_{{A A}_{11}}^{s the s} ((55)) \times \times {μ μ}_{{A A}_{22}}^{m m} ((8080))$

$= = 0.5 0.5 \times \times 0.79 0.79 + + 0.5 0.5 \times \times 0.21 0.21 = = 0.5 0.5$

${μ μ}^{((b b))} = = {μ μ}_{((s the s,, b b))} = = {μ μ}_{{A A}_{11}}^{s the s} ((55)) \times \times {μ μ}_{{A A}_{22}}^{b b} ((8080)) = = 0.5 0.5 \times \times 00 = = 00$

然后，根据式(4)计算驱动噪声方差加权系数α：Then, calculate the driving noise variance weighting coefficient α according to formula (4):

$α α = = \frac{{α α}_{00} \times \times {μ μ}^{((s the s))} + + {α α}_{11} \times \times {μ μ}^{((m m))} + + {α α}_{22} \times \times {μ μ}^{((b b))}}{{μ μ}^{((s the s))} + + {μ μ}^{((m m))} + + {μ μ}^{((b b))}} = = \frac{0.7 0.7 \times \times 0.5 0.5 + + 11 \times \times 0.5 0.5 + + 1.4 1.4 \times \times 00}{0.5 0.5 + + 0.5 0.5 + + 00} = = 0.85 0.85$

本发明采用上述模糊逻辑算法实时调整卡尔曼滤波器驱动噪声方差加权系数，使的滤波器参数随着载体机动情况而调整，提高了滤波估计精度和滤波稳定性，因而有助于提高跟踪环路的跟踪精度和稳定性。The present invention adopts the above-mentioned fuzzy logic algorithm to adjust the Kalman filter driving noise variance weighting coefficient in real time, so that the filter parameters can be adjusted according to the carrier maneuvering situation, which improves the filtering estimation accuracy and filtering stability, thus helping to improve the tracking loop tracking accuracy and stability.

目前国内外主要采用EKF的集中式滤波对GNSS超紧组合矢量跟踪接收机的导航滤波器进行建模，本发明提出了采用联邦滤波理论设计GNSS矢量跟踪接收机的导航滤波器设计方案，提高导航滤波器容错能力和滤波精度。At present, the centralized filtering of EKF is mainly used to model the navigation filter of the GNSS super-tight combination vector tracking receiver at home and abroad. The present invention proposes the design scheme of the navigation filter of the GNSS vector tracking receiver using the federated filtering theory to improve the navigation efficiency. Filter fault tolerance and filtering accuracy.

本发明中导航滤波器5采用分布式的联邦滤波结构，总体设计方案如图7所示，包括联邦滤波子滤波器和联邦滤波主滤波器，各通道自适应滤波器4的输出作为联邦滤波子滤波器的观测输入，联邦主滤波器对子滤波器的状态进行信息分配。各子滤波器的状态估计输出端连接联邦主滤波器的观测量输入端，联邦主滤波器信息分配系数输出端反馈连接各子滤波器估计误差方差。In the present invention, the navigation filter 5 adopts a distributed federated filtering structure, and the overall design scheme is as shown in Figure 7, including a federated filtering sub-filter and a federated filtering main filter, and the output of each channel adaptive filter 4 is used as a federated filtering sub-filter The observation input of the filter, the federated main filter assigns information to the state of the sub-filters. The state estimation output end of each sub-filter is connected to the observation input end of the federated main filter, and the output end of the information distribution coefficient of the federated main filter is connected to the estimated error variance of each sub-filter in feedback.

具体地，联邦滤波器是一种两级数据融合结构，子滤波器和主滤波器并行工作，如图7所示，图中微惯导系统IMU的输出X_k一方面直接给联邦主滤波器，另一方面它可以输给各子滤波器作为观测值，GNSS各通道的输出Z只给相应的子滤波器，各子滤波器的局部估计

及其协方差阵P_i送入主滤波器和主滤波器的估计值一起进行融合以得到全局最优估计

此外，从图7中还可以看到，由子滤波器与主滤波器合成的全局估计值

及其相应的协方差阵P_g被放大为后再反馈到子滤波器以重置子滤波器的估计值。Specifically, the federated filter is a two-stage data fusion structure, and the sub-filter and the main filter work in parallel, as shown in Figure 7. In the figure, the output X _k of the IMU of the micro-inertial navigation system directly feeds the federated main filter , on the other hand, it can be output to each sub-filter as an observation value, the output Z of each GNSS channel is only given to the corresponding sub-filter, and the local estimation of each sub-filter

and its covariance matrix P _i are sent to the main filter and the estimated value of the main filter is fused together to obtain the global optimal estimate

In addition, it can also be seen from Figure 7 that the global estimated value synthesized by the sub-filter and the main filter

and its corresponding covariance matrix P _g is amplified as Then feed back to the sub-filter to reset the estimated value of the sub-filter.

主滤波器包括时间更新单元与最优融合单元，主滤波器的预报误差及其协方差P_m送入最优融合单元，同时主滤波器预报误差

的协方差P_m也可重置为全局协方差阵的倍，即为用来重置主滤波器的估计值。The main filter includes a time update unit and an optimal fusion unit, and the prediction error of the main filter and its covariance P _m are sent to the optimal fusion unit, and the prediction error of the main filter is

The covariance P _m of can also be reset to the global covariance matrix times, that is Used to reset the estimated value of the main filter.

其中，β_i(i＝1，2，…，N，m)称为信息分配系数，N为子滤波器的个数，采用动态最优信息分配系数的方法确定各子滤波器的信息分配系数，不同的β_i值可以获得联邦滤波器的不同结构和不同特性。在k时刻选取信息分配系数为：Among them, β _i (i=1, 2, ..., N, m) is called the information distribution coefficient, N is the number of sub-filters, and the method of dynamic optimal information distribution coefficient is used to determine the information distribution coefficient of each sub-filter , different β _i values can obtain different structures and different characteristics of the federated filter. Select the information distribution coefficient at time k as:

${β β}_{i i ((k k))} = = \frac{trace trace [[{(({P P}_{ii i ((k k,, k k - - 11))}^{- - 11}))}^{* *}]]}{trace trace [[{M m}_{i i} {P P}_{g g ((k k,, k k - - 11))}^{- - 11} {M m}_{i i}^{T T}]]}$

其中， ${(P_{ii (k, k - 1)}^{- 1})}^{*} = {[Φ_{k, k - 1} P_{ii (- 1)} Φ_{k, k - 1}^{T} + Γ_{k - 1} Q_{i (k - 1)} Γ_{K - 1}^{T}]}^{- 1},$ M_i是子系统状态X_ik与总系统状态X_k之间存在的关系矩阵。Φ_k，k-1为k时刻下的状态转移矩阵，Q_i(k-1)为第i个子滤波器(k-1)时刻下的驱动噪声方差阵，P_ii(k，k-1)、P_ii(k-1)分别为第i个子滤波器(k-1)时刻下的预测误差方差矩阵和估计误差方差矩阵，Γ为驱动噪声系数矩阵。in, ${(P_{i (k, k - 1)}^{- 1})}^{*} = {[Φ_{k, k - 1} P_{i (- 1)} Φ_{k, k - 1}^{T} + Γ_{k - 1} Q_{i (k - 1)} Γ_{K - 1}^{T}]}^{- 1},$ M _i is the relationship matrix between the subsystem state X _ik and the total system state X _k . Φ _{k, k-1} is the state transition matrix at time k, Q _i(k-1) is the driving noise variance matrix at the time of the ith sub-filter (k-1), P _{ii(k, k-1)} , P _ii(k-1) are respectively the prediction error variance matrix and estimation error variance matrix at the moment of the ith sub-filter (k-1), and Γ is the driving noise coefficient matrix.

本发明中联邦滤波有3个子滤波器：纯GNSS模式下的GNSS滤波器、超紧组合模式下的GNSS/IMU组合导航滤波器、纯IMU微惯导滤波器。In the present invention, the federated filter has three sub-filters: a GNSS filter in pure GNSS mode, a GNSS/IMU combined navigation filter in super-tight combination mode, and a pure IMU micro inertial navigation filter.

其中，纯GNSS模式下的GNSS滤波器采用Singer模型建模，GNSS/IMU组合导航滤波器和纯IMU微惯导滤波器均采用惯性导航系统误差传播模型建模，这些建模技术均为已有技术，因此，详细建模过程不在此赘述。Among them, the GNSS filter in the pure GNSS mode is modeled by the Singer model, and the GNSS/IMU integrated navigation filter and the pure IMU micro-inertial navigation filter are modeled by the error propagation model of the inertial navigation system. These modeling techniques are all existing technology, therefore, the detailed modeling process is not repeated here.

本发明超紧组合卫星导航接收机开启导航系统后，初始化过程包括IMU微惯导初始化和纯GNSS导航初始化，一旦GNSS接收机捕获到卫星信号且IMU微惯导完成初始化对准过程，系统运行超紧组合导航滤波定位算法，以提高动态性能和抗干扰性能。IMU微惯导初始化主要实现惯性设备加电初始化以及航向、姿态、位置参数的初始化，为INS解算和超紧组合导航解算做准备。纯GNSS导航模式的初始化利用了已存历书、星历等信息，实现GNSS接收机的快速捕获。After the ultra-tight combination satellite navigation receiver of the present invention starts the navigation system, the initialization process includes IMU micro-inertial navigation initialization and pure GNSS navigation initialization. Once the GNSS receiver captures satellite signals and the IMU micro-inertial navigation completes the initialization alignment process, the system runs beyond Tightly combined navigation filter positioning algorithm to improve dynamic performance and anti-jamming performance. The IMU micro-inertial navigation initialization mainly realizes the power-on initialization of the inertial equipment and the initialization of the heading, attitude, and position parameters, and prepares for the INS calculation and ultra-tight integrated navigation calculation. The initialization of the pure GNSS navigation mode utilizes the stored almanac, ephemeris and other information to realize the fast acquisition of the GNSS receiver.

正常状态下系统主要依靠超紧组合导航子滤波器；当GNSS信号失锁无法定位时，系统主要依靠纯IMU微惯导子滤波器；当IMU微惯导出现故障时或系统初始化时，系统主要依靠纯GNSS导航子滤波器。Under normal conditions, the system mainly relies on the ultra-tight integrated navigation sub-filter; when the GNSS signal is out of lock and cannot be positioned, the system mainly relies on the pure IMU micro-inertial navigation sub-filter; when the IMU micro-inertial navigation fails or the system is initialized, the system mainly relies on Rely on pure GNSS navigation sub-filters.

以上实施例仅用以说明本发明的技术方案而非对其进行限制，在不背离本发明精神及其实质的情况下，熟悉本领域的技术人员当可根据本发明作出各种相应的改变和变形，但这些相应的改变和变形都应属于本发明所附的权利要求的保护范围。The above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Without departing from the spirit and essence of the present invention, those skilled in the art can make various corresponding changes and changes according to the present invention. deformation, but these corresponding changes and deformations should belong to the scope of protection of the appended claims of the present invention.

Claims

1. a hypercompact combined satellite navigation receiver, comprise radio-frequency front-end, signal preprocessor, base band prefilter and Navigation Filter, and micro-inertial navigation module, pseudorange and pseudorange rates computing module, ephemeris resolves and selects star computing module, wherein signal pre-processing module comprises correlator and local signal maker, it is characterized in that, between described base band prefilter and Navigation Filter, is provided with sef-adapting filter;

Described Navigation Filter is Federated Filters, comprises senior filter and at least one subfilter; Wherein, the output of sef-adapting filter is as the observation input of subfilter, and senior filter carries out information distribution to the state of each subfilter; The state estimation output terminal of each subfilter connects the observation input end of senior filter, and senior filter output information partition factor feeds back to each subfilter, in order to the state estimation value of the subfilter of resetting;

The expression formula of described information distribution coefficient is as follows:

β_{i (k)} = \frac{trace [{(P_{ii (k, k - 1)}^{- 1})}^{*}]}{trace [M_{i} P_{g (k, k - 1)}^{- 1} M_{i}^{T}]}

Wherein: β _{i (k)}during for k, inscribe the information distribution coefficient of i subfilter,

{(P_{ii (k, k - 1)}^{- 1})}^{*} = {[Φ_{k, k - 1} P_{ii (k - 1)} Φ_{k, k - 1}^{T} + Γ_{k - 1} Q_{i (k - 1)} Γ_{k - 1}^{T}]}^{- 1},

M _isubsystem state X _ikwith total system state X _kbetween the relational matrix that exists, Φ _{k, k-1}the state-transition matrix of inscribing during for k, Q _{i (k-1)}the driving noise variance matrix of inscribing while being i subfilter k-1, P _{ii (k, k-1)}, P _{ii (k-1)}the predicated error variance matrix of inscribing while being respectively i subfilter k-1 and estimation error variance matrix, P _{g (k, k-1)}for P _{ii (k, k-1)}corresponding covariance matrix, the driving noise figure matrix that Γ k-1 inscribes while being k-1;

Described subfilter comprises the GNSS wave filter under pure GNSS pattern, the micro-inertial navigation wave filter of the GNSS/IMU integrated navigation wave filter under hypercompact integrated mode and pure IMU;

The state equation of described sef-adapting filter is:

\frac{d}{dt} [\begin{matrix} δρ \\ δ \dot{ρ} \\ δ \overset{. .}{ρ} \\ δN \\ Δ_{ion} \\ {\dot{Δ}}_{ion} \end{matrix}] = [\begin{matrix} δ \dot{ρ} \\ δ \overset{. .}{ρ} \\ 0 \\ 0 \\ {\dot{Δ}}_{ion} \\ 0 \end{matrix}] + [\begin{matrix} 0 \\ 0 \\ w_{a} \\ 0 \\ 0 \\ w_{ion} \end{matrix}]

Observation equation is:

[\begin{matrix} δρ \\ δ \dot{ρ} \\ δ \overset{. .}{ρ} \\ δN \\ Δ_{ion} \\ {\dot{Δ}}_{ion} \end{matrix}] = [\begin{matrix} δρ \\ δ \dot{ρ} \\ δ \overset{. .}{ρ} \\ δN \\ Δ_{ion} \\ {\dot{Δ}}_{ion} \end{matrix}] + [\begin{matrix} w_{ρ} \\ w_{\dot{ρ}} \\ w_{\overset{. .}{ρ}} \\ w_{N} \\ w_{i} \\ w_{io} \end{matrix}]

The state equation of described base band prefilter is:

\frac{d}{dt} [\begin{matrix} δρ \\ δ \dot{ρ} \\ δ \overset{. .}{ρ} \\ δN \\ Δ_{ion} \\ {\dot{Δ}}_{ion} \end{matrix}] = [\begin{matrix} δ \dot{ρ} \\ δ \overset{. .}{ρ} \\ 0 \\ 0 \\ {\dot{Δ}}_{ion} \\ 0 \end{matrix}] + [\begin{matrix} 0 \\ 0 \\ w_{a} \\ 0 \\ 0 \\ w_{ion} \end{matrix}]

Observation equation is:

[\begin{matrix} Δρ \\ Δφ \end{matrix}] = [\begin{matrix} δρ + Δ_{ion} \\ δρ - Δ_{ion} + δN \end{matrix}]

Wherein: δ ρ is pseudorange residual error,

for pseudorange rates residual error,

for pseudorange acceleration residual error, δ N is carrier phase complete cycle residual error, Δ _ionfor ionosphere time delay,

for ionosphere time delay rate, w _afor pseudorange acceleration process noise, w _ionfor ionospheric noise, w _ρ,

w _n, w _i, w _iobe respectively the observation noise of pseudorange, pseudorange rates, pseudorange acceleration, ambiguity of carrier phase, ionosphere time delay, ionosphere time delay rate; Δ ρ is pseudorange residual error measured value, and Δ φ is carrier phase residual error measured value.

2. hypercompact combined satellite navigation receiver as claimed in claim 1, is characterized in that, described local signal maker, and correlator, base band prefilter, sef-adapting filter and Navigation Filter, pseudorange and pseudorange rates computing module form vector tracking loop.

3. hypercompact combined satellite navigation receiver as claimed in claim 1 or 2, it is characterized in that, described hypercompact combined satellite navigation receiver comprises at least one road navigation signal receiving cable, each navigation signal receiving cable is provided with signal preprocessor and base band prefilter, and each roadbed band prefilter is all by accessing Navigation Filter after a sef-adapting filter.

4. hypercompact combined satellite navigation receiver as claimed in claim 1, it is characterized in that, described sef-adapting filter is fuzzy logic adaptive wave filter, comprise fuzzy logic algorithm module and Kalman filter, described fuzzy logic algorithm module receives pseudorange filtering and newly ceases mean value and mean square deviation, be output as and drive noise variance weighting coefficient, for adjusting in real time the parameter value of the driving noise variance of described Kalman filter, realize auto adapted filtering.

5. hypercompact combined satellite navigation receiver as claimed in claim 4, is characterized in that, described fuzzy logic algorithm module comprises:

Degree of membership simulation unit, for determining that the pseudorange filtering of input newly ceases the membership function of mean value and mean square deviation, and the membership function of the driving noise variance weighting coefficient of output;

Fuzzy inference rule unit, for arranging fuzzy inference rule;

Noise variance weighted units, for calculating driving noise variance weighting coefficient according to described membership function and fuzzy inference rule ambiguity solution.