GNSS sensor fault diagnosis method, device, navigation system and vehicle

Abstract

The embodiment of the present invention provides a fault diagnosis method, device, navigation system and vehicle for a GNSS sensor, the method comprising: responding to a navigation information output instruction of a GNSS sensor, obtaining a historical estimation result for the GNSS sensor at the last moment; updating the prediction result of a target sensor in the GNSS sensor at the current moment according to the historical estimation result for the GNSS sensor at the last moment; obtaining a fault diagnosis threshold, and determining the fault diagnosis result of the target sensor according to the fault diagnosis threshold and the prediction result at the current moment; and fusing the prediction result at the current moment according to the fault diagnosis result to obtain the estimation result at the current moment. The fused estimation result is used as the historical estimation result when the GNSS sensor is fault diagnosed at the next moment, and the existing historical estimation result is used as the initial value of the fault diagnosis work, so as to ensure the iterative and continuous fault diagnosis of the GNSS sensor, and realize the fault diagnosis of a single sensor or multiple sensors.

Description

Technical Field

[0001] The present invention relates to the field of navigation technology, and in particular to a fault diagnosis method for a GNSS sensor, a fault diagnosis device for a GNSS sensor, a corresponding navigation system, and a corresponding vehicle. Background Art

[0002] For the autonomous driving system of vehicles / flying cars, the GNSS (Global Navigation Satellite System) sensor of the navigation equipment is the key sensor for the multi-source fusion positioning of the autonomous driving system. It is crucial to the safety and stability of the autonomous driving system. When there is a problem with the GNSS positioning result, it will have a serious impact on the final fusion positioning result. It is necessary to perform fault diagnosis and isolation (FDI) on the GNSS sensor.

[0003] At present, the GNSS sensors of vehicles / flying cars use the same type of sensor redundancy configuration scheme, which ensures the safety and stability of vehicles / flying cars by carrying multiple GNSS sensors and using hardware redundancy to detect the consistency of sensor output to diagnose GNSS sensor faults. However, this hardware redundancy method of detecting sensor output consistency has failures. For example, when there are one GNSS sensor or two GNSS sensors, the faulty sensor cannot be detected. Summary of the invention

[0004] In view of the above problems, embodiments of the present invention are proposed to provide a GNSS sensor fault diagnosis method, a GNSS sensor fault diagnosis device, a corresponding navigation system and a corresponding vehicle that overcome the above problems or at least partially solve the above problems.

[0005] An embodiment of the present invention discloses a fault diagnosis method for a GNSS sensor, the method comprising:

[0006] In response to the navigation information output instruction of the GNSS sensor, obtaining a historical estimation result for the GNSS sensor at a previous moment;

[0007] According to the historical estimation result of the GNSS sensor at the previous moment, updating the prediction result of the target sensor in the GNSS sensor at the current moment;

[0008] Acquire a fault diagnosis threshold, and determine a fault diagnosis result of the target sensor according to the fault diagnosis threshold and the prediction result at the current moment;

[0009] According to the fault diagnosis result, the prediction result at the current moment is fused to obtain the estimation result at the current moment; the estimation result at the current moment is used as the historical estimation result when the fault diagnosis is performed on the GNSS sensor at the next moment.

[0010] Optionally, updating the prediction result of the target sensor in the GNSS sensor at the current moment according to the historical estimation result for the GNSS sensor at the previous moment includes:

[0011] When the historical estimation result at the initial moment is obtained, the prediction result of the target sensor in the GNSS sensor at the current moment is updated according to the historical estimation result for the GNSS sensor at the previous moment.

[0012] Optionally, the prediction result at the current moment is determined based on a state quantity prediction value, wherein the state quantity prediction value includes a state to be estimated and a state variance matrix for each sensor in the GNSS sensor; the target sensor includes at least one target sensor;

[0013] The updating of the prediction result of the target sensor in the GNSS sensor at the current moment according to the historical estimation result for the GNSS sensor at the previous moment includes:

[0014] Adopting the state to be estimated in the historical estimation result of the previous moment and the state variance matrix in the historical estimation result, respectively updating the state to be estimated and the state variance matrix of the at least one target sensor at the current moment, and obtaining the prediction result of the at least one target sensor at the current moment; wherein, the historical estimation result for the GNSS sensor at the previous moment includes the historical estimation result at the initial moment, and / or the historical estimation result for a single sensor of the GNSS sensor at the previous moment, and / or the historical fusion data for at least two sensors of the GNSS sensor at the previous moment.

[0015] Optionally, determining the fault diagnosis result of the target sensor according to the fault diagnosis threshold and the prediction result at the current moment includes:

[0016] Obtaining a false alarm rate of the GNSS sensor, and using the false alarm rate to obtain a fault diagnosis threshold;

[0017] The fault diagnosis result of the target sensor in the GNSS sensor at the current moment is determined by using the comparison result of the prediction result at the current moment and the fault diagnosis threshold.

[0018] Optionally, obtaining the false alarm rate of the GNSS sensor includes:

[0019] Obtaining a fault diagnosis result for the GNSS sensor at the last moment and a prediction result for the GNSS sensor at the last moment, and obtaining a state quantity collected in real time for the GNSS sensor at the last moment;

[0020] The prediction result for the GNSS sensor at the last moment is compared with the state quantity collected in real time for the GNSS sensor at the last moment to determine the actual fault diagnosis result for the GNSS sensor at the last moment;

[0021] The false alarm rate of the GNSS sensor is obtained by using the ratio of the number of faulty sensors that are inconsistent between the actual fault diagnosis result and the fault diagnosis result for the GNSS sensor at the previous moment to the number of faulty sensors in the fault diagnosis result for the GNSS sensor at the previous moment.

[0022] Optionally, the using the comparison result of the prediction result at the current moment and the fault diagnosis threshold to determine the fault diagnosis result of the GNSS sensor at the current moment includes:

[0023] Using the prediction result at the current moment to construct a fault diagnosis quantity for the target sensor;

[0024] If the fault diagnosis amount of the target sensor is greater than or equal to the fault diagnosis threshold, determining that a fault exists in the target sensor in the GNSS sensor;

[0025] And / or, if the fault diagnosis amount of the target sensor is less than the fault diagnosis threshold, it is determined that the target sensor in the GNSS sensor has no fault.

[0026] Optionally, the fusing the prediction result at the current moment according to the fault diagnosis result to obtain the estimation result at the current moment includes:

[0027] Based on the fault diagnosis result, obtaining the prediction result of the normal sensor at the current moment;

[0028] Performing measurement update on the prediction result of the normal sensor at the current moment to obtain an estimation result of the normal sensor after the measurement update at the current moment;

[0029] The estimation results after the normal sensors are measured and updated at the current moment are fused to obtain the estimation results at the current moment.

[0030] The embodiment of the present invention further discloses a GNSS sensor fault diagnosis device, the device comprising:

[0031] A historical estimation result acquisition module, used to respond to the navigation information output instruction of the GNSS sensor and acquire the historical estimation result for the GNSS sensor at the last moment;

[0032] A prediction result updating module, used for updating the prediction result of the target sensor in the GNSS sensor at the current moment according to the historical estimation result for the GNSS sensor at the previous moment;

[0033] A fault diagnosis result determination module, used to obtain a fault diagnosis threshold, and determine a fault diagnosis result of the target sensor according to the fault diagnosis threshold and the prediction result at the current moment;

[0034] A prediction result fusion module is used to fuse the prediction result at the current moment according to the fault diagnosis result to obtain the estimation result at the current moment; the estimation result at the current moment is used as the historical estimation result when the fault diagnosis is performed on the GNSS sensor at the next moment.

[0035] Optionally, the prediction result updating module includes:

[0036] The prediction result updating submodule is used to update the prediction result of the target sensor in the GNSS sensor at the current moment according to the historical estimation result for the GNSS sensor at the previous moment when the historical estimation result at the initial moment is obtained.

[0037] Optionally, the prediction result at the current moment is determined based on a state quantity prediction value, wherein the state quantity prediction value includes a state to be estimated and a state variance matrix for each sensor in the GNSS sensor; the target sensor includes at least one target sensor;

[0038] The prediction result updating submodule includes:

[0039] A prediction result updating unit is used to adopt the state to be estimated in the historical estimation result of the previous moment and the state variance matrix in the historical estimation result to respectively update the state to be estimated and the state variance matrix of the at least one target sensor at the current moment, so as to obtain the prediction result of the at least one target sensor at the current moment; wherein the historical estimation result for the GNSS sensor at the previous moment includes the historical estimation result at the initial moment, and / or the historical estimation result for a single sensor of the GNSS sensor at the previous moment, and / or the historical fusion data for at least two sensors of the GNSS sensor at the previous moment.

[0040] Optionally, the fault diagnosis result determination module includes:

[0041] A fault diagnosis threshold acquisition submodule, used to acquire a false alarm rate of the GNSS sensor, and to acquire a fault diagnosis threshold using the false alarm rate;

[0042] The fault diagnosis result determination submodule is used to determine the fault diagnosis result of the target sensor in the GNSS sensor at the current moment by using the comparison result of the prediction result at the current moment and the fault diagnosis threshold.

[0043] Optionally, the fault diagnosis threshold acquisition submodule includes:

[0044] A state quantity acquisition unit, used to acquire a fault diagnosis result for the GNSS sensor at a previous moment and a prediction result for the GNSS sensor at a previous moment, and to acquire a state quantity collected in real time by the GNSS sensor at a previous moment;

[0045] an actual fault diagnosis result determination unit, configured to compare the prediction result for the GNSS sensor at the last moment with the state quantity collected in real time for the GNSS sensor at the last moment, and determine the actual fault diagnosis result for the GNSS sensor at the last moment;

[0046] The false alarm rate calculation unit is used to obtain the false alarm rate of the GNSS sensor by using the ratio of the number of faulty sensors that are inconsistent between the actual fault diagnosis result and the fault diagnosis result for the GNSS sensor at the previous moment and the number of faulty sensors in the fault diagnosis result for the GNSS sensor at the previous moment.

[0047] Optionally, the fault diagnosis result determination submodule includes:

[0048] A fault diagnosis quantity construction unit, used to construct a fault diagnosis quantity for a target sensor using the prediction result at the current moment;

[0049] a first fault diagnosis unit, configured to determine that a target sensor in the GNSS sensor has a fault when a fault diagnosis amount of the target sensor is greater than or equal to the fault diagnosis threshold;

[0050] The first fault diagnosis unit is configured to determine that a target sensor in the GNSS sensor has no fault when the fault diagnosis amount of the target sensor is less than the fault diagnosis threshold.

[0051] Optionally, the prediction result fusion module includes:

[0052] A prediction result acquisition submodule, used to acquire the prediction result of the normal sensor at the current moment based on the fault diagnosis result;

[0053] A measurement update submodule, used to perform measurement update on the prediction result of the normal sensor at the current moment, to obtain an estimation result of the normal sensor after the measurement update at the current moment;

[0054] The estimation result fusion submodule is used to fuse the estimation result after the normal sensor is measured and updated at the current moment to obtain the estimation result at the current moment.

[0055] The embodiment of the present invention further discloses a navigation system, comprising: a processor; a memory storing instructions executable by the processor; wherein the processor is coupled to the memory, and is used to read program instructions stored in the memory, and in response, performs the following operations:

[0056] In response to a navigation information output instruction of the navigation device, obtaining a historical estimation result for the navigation device at a previous moment;

[0057] According to the historical estimation result for the navigation device at the previous moment, updating the prediction result of the target sensor in the navigation device at the current moment;

[0058] Acquire a fault diagnosis threshold, and determine a fault diagnosis result of the target sensor according to the fault diagnosis threshold and the prediction result at the current moment;

[0059] According to the fault diagnosis result, the prediction result at the current moment is fused to obtain the estimation result at the current moment; the estimation result at the current moment is used as the historical estimation result when the fault diagnosis is performed on the navigation device at the next moment.

[0060] An embodiment of the present invention also discloses a vehicle, comprising: a fault diagnosis device of the navigation device, a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein when the computer program is executed by the processor, any one of the fault diagnosis methods of the navigation device is implemented.

[0061] The embodiments of the present invention include the following advantages:

[0062] In an embodiment of the present invention, the navigation information output instruction of the GNSS sensor can be responded to, and the prediction result of the target sensor in the GNSS sensor at the current moment can be updated by acquiring the historical estimation result for the GNSS sensor at the previous moment, and the fault diagnosis result of the target sensor can be determined based on the comparison between the prediction result at the current moment and the fault diagnosis threshold, so as to fuse the prediction result at the current moment to obtain the estimation result at the current moment, and use the fused estimation result at the current moment as the historical estimation result when performing fault diagnosis on the GNSS sensor at the next moment, and use the existing historical estimation result as the initial value of the fault diagnosis work, so as to ensure iterative and continuous fault diagnosis of the GNSS sensor, and realize fault diagnosis of a single sensor or multiple sensors separately. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 is a flowchart of a method for diagnosing a fault of a GNSS sensor according to an embodiment of the present invention;

[0064] Figure 2 is a flowchart of another embodiment of a method for diagnosing a fault of a GNSS sensor according to the present invention;

[0065] Figure 3 is a schematic diagram of the working process of a navigation system for fault diagnosis provided by an embodiment of the present invention;

[0066] Figure 4 It is a structural block diagram of an embodiment of a fault diagnosis device for a GNSS sensor of the present invention. DETAILED DESCRIPTION

[0067] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0068] In the related technologies for fault diagnosis of GNSS sensors, assuming that for a dual-antenna GNSS board, the GNSS navigation results usually include position, speed, and heading, the fault diagnosis of the GNSS sensor can be manifested as verifying the correctness of these quantities.

[0069] In the existing combined navigation algorithms, for a single GNSS sensor, a special fault test is generally not performed on the GNSS sensor. The correctness of the output result is usually judged only based on the output identification bit of the GNSS, such as the fixed identification of the navigation result. However, due to the defects in the principle of the GNSS test algorithm itself and the large error of the navigation result in some cases, the accuracy of the navigation result judgment method relying on the GNSS identification bit is low.

[0070] Specifically, the principle of the GNSS verification algorithm itself has defects, such as the false fixation of RTK (Real Time Kinematic, a technology that uses carrier phase observations for real-time dynamic relative positioning) positioning results. In the current GNSS algorithm, the fixation of the navigation result is determined by comparing a certain detection quantity with a threshold. For example, the common RTK fixation condition is assumed to be the integer ambiguity RATIO>3. Based on the nature of the GNSS ambiguity verification algorithm itself, this way of judging the navigation result allows a certain amount of missed detection, thereby causing the GNSS identification position to be unreliable, further reducing the accuracy of the navigation result judgment; and, in some cases, such as when the number of satellites is small, the accuracy of the GNSS navigation result cannot reach the ideal state, resulting in a large error in the navigation result, thereby causing the GNSS identification position to be unreliable, further reducing the accuracy of the navigation result judgment.

[0071] For the existing multiple sensors of the same type, a hardware redundant fault diagnosis method is usually adopted, that is, the judgment is made based on the consistency of the results output by multiple similar transmitters. Specifically, the faulty sensor is determined by mutual verification of multiple device data. However, this hardware redundant method of detecting sensor output consistency has failure situations. For example, when there are two GNSS sensors and a fault is detected based on output consistency, it is impossible to determine the specific sensor that has detected the fault. In addition, for the case of a single sensor, this fault diagnosis method of detecting sensor output consistency cannot be implemented.

[0072] One of the core ideas of the embodiments of the present invention is to use a method based on analytical redundancy to diagnose faults of GNSS sensors. Instead of using a simple hardware redundancy method, an analytical redundancy method is used. The estimation result at the current moment obtained by fusion is used as the historical estimation result when diagnosing faults of the GNSS sensor at the next moment. The existing historical estimation result is used as the initial value of the fault diagnosis work, thereby ensuring iterative and continuous fault diagnosis of the GNSS sensor, and realizing fault diagnosis of a single sensor or multiple sensors separately; further, the data of the GNSS sensor can also be checked by a fault diagnosis threshold, so that when there are two GNSS sensors, the faulty sensor can be determined in the two GNSS sensors based on the comparison between the model prediction value and the observation value.

[0073] Reference Figure 1 , shows a flowchart of a method for diagnosing a fault of a GNSS sensor according to an embodiment of the present invention, which may specifically include the following steps:

[0074] Step 101, responding to a navigation information output instruction of a GNSS sensor, obtaining a historical estimation result of the GNSS sensor at a previous moment;

[0075] In an embodiment of the present invention, after the navigation system receives navigation information such as position, speed, and heading output by the GNSS board, it can perform fault diagnosis on the GNSS sensor to determine whether the output navigation information is usable. The fault diagnosis of the GNSS sensor can mainly be manifested in using the estimation result after fusion at the previous moment, obtaining the prediction result at the current moment through time update, and then comparing the prediction result at the current moment after the time update with the data obtained by the sensor in real time to determine whether the sensor has a fault, thereby realizing fault diagnosis of the GNSS sensor.

[0076] In practical applications, the navigation information output instruction of the GNSS sensor can be responded to to obtain the estimation result after fusion at the previous moment, that is, the historical estimation result of the GNSS sensor at the previous moment, so as to update the acquired historical estimation result in time to obtain the prediction result at the current moment.

[0077] In order to ensure the fault diagnosis of a single sensor or multiple sensors, a reasonable initial value is required so that the analytical redundancy method adopted in the embodiment of the present invention can work normally for fault diagnosis. The reasonable initial value can be determined by the historical estimation results obtained. Among them, the historical estimation results used as the initial value can include the historical estimation results at the initial moment; and since there is no need to perform fusion when there is a single sensor, it can also include the historical estimation results for a single sensor in the GNSS sensor at the last moment; and, if there are at least two sensors in the GNSS sensor that are forced to open, then the historical estimation results can also include the historical fusion data for at least two sensors in the GNSS sensor at the last moment, and the embodiment of the present invention does not limit this.

[0078] Step 102, updating the prediction result of the target sensor in the GNSS sensor at the current moment according to the historical estimation result for the GNSS sensor at the previous moment;

[0079] After obtaining the estimation result after fusion at the previous moment, that is, the historical estimation result for the GNSS sensor at the previous moment, the obtained historical estimation result can be updated in time, which can be mainly manifested in updating the prediction result of the target sensor in the GNSS sensor at the current moment according to the historical estimation result for the GNSS sensor at the previous moment, such as obtaining the predicted position, speed, etc., so as to compare the prediction result at the current moment with the data obtained by the sensor in real time to perform fault diagnosis.

[0080] In a specific implementation, when the historical estimation result at the initial moment is obtained, the prediction result of the target sensor at the current moment can be updated in time. The target sensor to be updated can be determined based on the number of sensors of the GNSS sensor, that is, the target sensor can include one sensor or at least two sensors.

[0081] It should be noted that, when the historical estimation result at the initial moment is not obtained, that is, at the initial moment, since there is no historical fusion historical estimation result, the sensor fault diagnosis is not performed at this time. Then, when the prediction result of the sensor is updated, it is manifested as using the historical estimation result at the initial moment, and / or the historical estimation result of a single sensor in the GNSS sensor at the last moment, and / or the historical fusion data of at least two sensors in the GNSS sensor at the last moment to achieve the update.

[0082] In practical applications, when there is only one sensor in the GNSS sensor, it is necessary to perform time update on a single sensor by using the estimation result after fusion at the previous moment to update the single sensor; when there are at least two sensors in the GNSS sensor, the estimation result after fusion at the previous moment can be used to update the corresponding sensor.

[0083] Among them, the prediction result at the current moment can be determined based on the state quantity prediction value, and the state quantity prediction value includes the state to be estimated and the state variance matrix for each sensor in the GNSS sensor. At this time, the update of the prediction result of the sensor at the current moment is achieved by using the estimation result fused at the previous moment to update the state to be estimated and the state variance matrix at the current moment.

[0084] Step 103, obtaining a fault diagnosis threshold, and determining a fault diagnosis result of the target sensor according to the fault diagnosis threshold and the prediction result at the current moment;

[0085] When diagnosing a GNSS sensor fault, the prediction result at the current time after the time update can be compared with the data obtained by the sensor in real time to determine whether the sensor has a fault, thereby realizing the fault diagnosis of the GNSS sensor.

[0086] Specifically, the data obtained by the sensor in real time is collected by the sensor and can be expressed as a fault diagnosis threshold determined based on the real-time collected data, wherein the fault diagnosis threshold is obtained based on a given false alarm rate. Specifically, after the false alarm rate is given, the corresponding fault diagnosis threshold can be obtained by checking the card-square distribution table or based on experience.

[0087] The false alarm rate refers to the probability of detecting a fault when the device is not faulty, that is, the false alarm rate can be determined based on the ratio of the number of inconsistencies between the actual fault diagnosis results detected and the number of devices with faults in the fault diagnosis results to the number of devices with faults in the fault diagnosis results. In practical applications, since the fault diagnosis results at the current moment need to be determined by the fault diagnosis threshold, that is, the fault diagnosis results at the current moment have not been determined, the false alarm rate can be obtained by using the ratio of the number of inconsistencies between the actual fault diagnosis results and the number of devices with faults in the fault diagnosis results for the GNSS sensor at the previous moment to the number of devices with faults in the fault diagnosis results for the GNSS sensor at the previous moment.

[0088] The determination of the false alarm rate needs to be determined through real-time collected data. Therefore, when comparing the prediction result at the current moment after the time update with the data obtained by the sensor in real time, it can be expressed as a comparison result of the fault diagnosis threshold and the prediction result at the current moment to determine the fault diagnosis result of the target sensor to determine whether the sensor has a fault.

[0089] Step 104: Based on the fault diagnosis result, the prediction result at the current moment is integrated to obtain the estimation result at the current moment.

[0090] After obtaining the fault diagnosis result of the target sensor, the prediction result at the current moment can be fused based on the fault diagnosis result to obtain the estimation result at the current moment, so that the estimation result at the current moment can be used as the historical estimation result when performing fault diagnosis on the GNSS sensor at the next moment, so that the existing historical estimation result can be used as the initial value of the fault diagnosis work, thereby ensuring the iterative and continuous fault diagnosis of the GNSS sensor.

[0091] In practical applications, fault diagnosis results can be used to characterize specific faulty sensors and normal (i.e., non-faulty) sensors. When fusing the estimation results at the current moment, the estimation results of normal sensors can usually be fused. However, if the estimation results at the current moment are predicted results, then before fusion, the estimation results of normal sensors must be measured and updated.

[0092] In an embodiment of the present invention, the navigation information output instruction of the GNSS sensor can be responded to, and the prediction result of the target sensor in the GNSS sensor at the current moment can be updated by acquiring the historical estimation result for the GNSS sensor at the previous moment, and the fault diagnosis result of the target sensor can be determined based on the comparison between the prediction result at the current moment and the fault diagnosis threshold, so as to fuse the prediction result at the current moment to obtain the estimation result at the current moment, and use the fused estimation result at the current moment as the historical estimation result when performing fault diagnosis on the GNSS sensor at the next moment, and use the existing historical estimation result as the initial value of the fault diagnosis work, so as to ensure iterative and continuous fault diagnosis of the GNSS sensor, and realize fault diagnosis of a single sensor or multiple sensors separately.

[0093] In the embodiment of the present invention, the updating of the prediction results, the fusion of the prediction results and the calculation of the estimation results in the analytical redundancy method can be mainly achieved by combining the residual chi-square detection method and the federated filter. In order to facilitate those skilled in the art to understand the fault diagnosis method of the GNSS sensor proposed in the embodiment of the present invention, the basic principles of the residual chi-square detection method and the federated filter are introduced as follows:

[0094] 1. The residual chi-square detection method is a fault diagnosis method based on mathematical models. It is mainly used to replace hardware redundancy with analytical redundancy. Analytical redundancy mainly estimates the system output by constructing a system model and an observation model, and then compares the system output with the output observation to determine the fault information of the GNSS sensor, that is, to determine the faulty sensor in the GNSS sensor.

[0095] The linear system model constructed can be shown as formula 1:

[0096]

[0097] In formula 1, k refers to the current moment, k-1 refers to the previous moment; Z k is a measurement vector of dimension m×1. The measurement vector can be a variable obtained through some physical observation. k is an n×1-dimensional state vector. The state vector is a variable used to describe the dynamic characteristics of the system. It is generally an unobservable variable. Only when the state vector and the measurement vector are combined can the dynamic characteristics of the system be fully and adequately described. k / k-1 ,H k are known system structural parameters, such as detection position, speed and other parameters, which are respectively called n×n-dimensional state one-step transfer matrix and m×n-dimensional measurement matrix; W k-1 is the n×1-dimensional system noise vector, V kis a measurement noise vector of m×1 dimension. Both are zero-mean Gaussian white noise vector sequences (obeying normal distribution) and are uncorrelated with each other, that is, they satisfy the following conditions:

[0098]

[0099] Q in Formula 2 k is the process noise matrix, R k is the measurement noise variance matrix.

[0100] Then the state quantity prediction value can be obtained through the constructed linear system model P k / k-1 , as shown in Formula 3:

[0101]

[0102] in, Refers to the state to be estimated, such as position, speed, etc., P k / k-1 Refers to the state variance matrix, that is, the variance-covariance matrix of the state.

[0103] The fault diagnosis quantity λ can be further constructed as shown in Formula 4: k :

[0104]

[0105] The fault diagnosis process is shown in Formula 5:

[0106]

[0107] Among them, T D It is possible to set a fault diagnosis threshold.

[0108] 2. Federated filtering is a decentralized filtering method with good fault tolerance. Usually, federated filtering has five processes: information allocation, time update, measurement update, fault diagnosis, and information fusion. It can filter multiple GNSS sensors separately, reduce the amount of algorithm calculation, and improve the real-time performance of the algorithm.

[0109] (1) The general expression of information allocation in federated filtering is as shown in Formula 6:

[0110]

[0111] Where i = 1, 2, ... N, i is the filter subscript, N refers to the number of GNSS sensors, that is, i represents each filter in the GNSS sensor; m is the main filter subscript, β i is the information allocation coefficient, and the subscript g represents the main filter estimation value. Specifically, the main filter is the estimation result of the information fusion of each sub-filter.

[0112] The embodiment of the present invention adopts a federated filter (β m =0,β i =1 / N). Therefore, in addition to the initial i,k , Q i,k Except for information distribution, P is not distributed at other times. i,k , Q i,k The value of To update. refers to the state to be estimated, P i,k refers to the state variance matrix, Q i,k Refers to the noise matrix of the state equation, its purpose is to eliminate the recursive process of initialization.

[0113] That is, at t = t0, its effect on P i,k , Q i,k as well as The update performance is shown in Formula 7:

[0114]

[0115] At other times, only The update performance is shown in formula 8:

[0116]

[0117] (2) Time update is performed by performing corresponding time updates between the sub-filter and the main filter to obtain the predicted state, such as the predicted position, speed, etc. The constructed system model can be the same as Formula 2, and the update equation can be expressed as shown in Formula 9:

[0118]

[0119] In the embodiment of the present invention, m =0, there is no need to update the main filter in time, only the sub-filter needs to be updated in time, that is, i=1,2,...N, N refers to the number of GNSS sensors, and i does not include m.

[0120] (3) Fault diagnosis and isolation, which is manifested in the use of residual chi-square detection method to perform fault diagnosis on each GNSS sensor in the GNSS sensor, including fault diagnosis of position, speed and heading. It should be noted that for the GNSS fault diagnosis algorithm, in order to reduce the amount of calculation and reduce the algorithm time consumption and ensure the real-time performance of the detection algorithm, the position, speed and heading can be detected separately.

[0121] Specifically, the system model for the position and speed detection channel can be as follows, which may include a state equation and a measurement equation. The state equation can predict the state at the current moment based on the state at the previous moment, and the measurement equation can update the state at the current moment based on the input of the current sensor:

[0122]

[0123]

[0124]

[0125]

[0126] Among them, r k , v k , a k They represent the estimated real position, velocity, and acceleration at time k respectively; I3 represents the 3D unit matrix; Δt refers to the time interval between the updated position and velocity; They represent the position and velocity measurements output by the GNSS sensor at time k; σ *,0 is the process noise of the corresponding state quantity, which needs to be selected based on experience; * The measurement noise of the observed quantity can be obtained from the GNSS receiver output. And, Q k-1 , R k-1 The noises that characterize the state equation and the measurement equation respectively can be used to adjust the weights of the predicted value and the measured value, thereby affecting the state detection quantity.

[0127] Similarly, the system model of the heading channel can be shown as follows:

[0128]

[0129]

[0130]

[0131]

[0132] Among them, θ represents the attitude angle. For GNSS, it only includes pitch and yaw angles. is the corresponding attitude angular velocity; r k , v k , a k They represent the estimated real position, velocity, and acceleration at time k respectively; I3 represents the 3D unit matrix; Δt refers to the time interval between the updated position and velocity; They represent the position and velocity measurements output by the GNSS sensor at time k; σ*,0 is the process noise of the corresponding state quantity, which needs to be selected based on experience; * The measurement noise of the observed quantity can be obtained from the GNSS receiver output. And, Q k-1 , R k-1 The noises that characterize the state equation and the measurement equation respectively can be used to adjust the weights of the predicted value and the measured value, thereby affecting the state detection quantity.

[0133] Then the state quantity prediction value can be obtained through the system model P k / k-1 , and further by constructing the fault diagnosis quantity as shown in Formula 4, the fault diagnosis process as shown in Formula 5 is performed.

[0134] (4) Measurement update is performed for all normally operating GNSS devices, as shown in Formula 10:

[0135]

[0136] in is the estimation result of each sub-filter, and the estimation result is the physical quantity to be diagnosed, such as position, speed, heading, etc.

[0137] (5) Information fusion is performed by using formula 11 to fuse the normal sub-filter estimation results as the recursive initial value of the next step of the fault diagnosis algorithm, and continuously performing iterative detection.

[0138]

[0139] In practical applications, whether it is for a single GNSS sensor or for fault diagnosis of multiple GNSS sensors, the estimation result after fusion at the previous moment can be used to obtain the prediction result at the current moment through time update, and then the prediction result at the current moment after time update is compared with the data obtained by the sensor in real time to determine whether there is a fault in the sensor.

[0140] Reference Figure 2 , shows a flowchart of another embodiment of a fault diagnosis method for a GNSS sensor of the present invention, which may specifically include the following steps:

[0141] Step 201, when the historical estimation result at the initial moment is obtained, the prediction result of the target sensor in the GNSS sensor at the current moment is updated according to the historical estimation result for the GNSS sensor at the previous moment;

[0142] In an embodiment of the present invention, after the navigation system receives navigation information such as position, speed, and heading output by the GNSS board, it can perform fault diagnosis on the GNSS sensor to determine whether the output navigation information is usable. The fault diagnosis of the GNSS sensor can mainly be manifested in using the estimation result after fusion at the previous moment, obtaining the prediction result at the current moment through time update, and then comparing the prediction result at the current moment after the time update with the data obtained by the sensor in real time to determine whether the sensor has a fault, thereby realizing fault diagnosis of the GNSS sensor.

[0143] In practical applications, the navigation information output instruction of the GNSS sensor can be responded to to obtain the estimation result after fusion at the previous moment, that is, the historical estimation result of the GNSS sensor at the previous moment, as shown in Formula 9, and the obtained historical estimation result is updated in time to obtain the prediction result at the current moment.

[0144] It should be noted that, when the historical estimation result at the initial moment is not obtained, that is, at the initial moment, since there is no historical fusion historical estimation result, the sensor fault diagnosis is not performed at this time. Then, the historical estimation result obtained for the GNSS sensor at the last moment may include the historical estimation result at the initial moment, the historical estimation result for a single sensor in the GNSS sensor at the last moment, and the historical fusion data for at least two sensors in the GNSS sensor at the last moment. When the prediction result of the sensor is updated, it is manifested as using the historical estimation result at the initial moment, and / or the historical estimation result for a single sensor in the GNSS sensor at the last moment, and / or the historical fusion data for at least two sensors in the GNSS sensor at the last moment to achieve the update.

[0145] In practical applications, the prediction result at the current moment can be determined based on the state quantity prediction value, and the state quantity prediction value includes the state to be estimated and the state variance matrix for each sensor in the GNSS sensor. At this time, the update of the prediction result of the sensor at the current moment is achieved by using the estimation result fused at the previous moment to update the state to be estimated and the state variance matrix at the current moment.

[0146] In a specific implementation, the state to be estimated in the historical estimation results of the previous moment and the state variance matrix in the historical estimation results are used to update the state to be estimated and the state variance matrix of at least one target sensor at the current moment, respectively, to obtain the prediction result of at least one target sensor at the current moment; wherein, the historical estimation results for the GNSS sensor at the previous moment include the historical estimation results at the initial moment, and / or the historical estimation results for a single sensor in the GNSS sensor at the previous moment, and / or the historical fusion data for at least two sensors in the GNSS sensor at the previous moment.

[0147] In a preferred embodiment, the updating of prediction results, the fusion of prediction results and the calculation of estimation results in the analytical redundancy method can be mainly realized by combining the residual chi-square detection method and the federated filter. Generally, the federated filter has five processes: information allocation, time update, measurement update, fault diagnosis and information fusion. At the initial moment, information allocation can be performed on each sub-filter, such as Figure 3 As shown in the figure, each sub-filter is expressed as a sub-filter 1, sub-filter 2, and sub-filter N corresponding to GNSS device 1, GNSS device 2, and GNSS device N. At this time, the initial value of the state quantity can be selected. The initial values ​​of position and velocity can be selected according to the GNSS RTK solution. The initial value of acceleration can be assigned to 0 to obtain X0 and P0. At other times, only At the initial moment, there is no historical fusion data of the previous moment. At this time, the fault diagnosis quantity for fault diagnosis can be constructed based on the sub-filter information allocated to the initial moment, so that the prediction result of the current moment can be fused based on the fault diagnosis result to obtain the estimated result of the current moment, which can be used as the historical estimation result for fault diagnosis of the GNSS sensor at the next moment.

[0148] Step 202, obtaining a false alarm rate of the GNSS sensor, and using the false alarm rate to obtain a fault diagnosis threshold;

[0149] After the prediction result at the current moment is updated, the prediction result at the current moment after the time update can be compared with the data obtained by the sensor in real time, and the GNSS sensor fault diagnosis can be performed to determine whether the sensor has a fault, thereby realizing the fault diagnosis of the GNSS sensor.

[0150] Specifically, the data obtained by the sensor in real time is collected by the sensor and can be expressed as a fault diagnosis threshold determined based on the real-time collected data, wherein the fault diagnosis threshold is obtained based on a given false alarm rate. Specifically, after the false alarm rate is given, the corresponding fault diagnosis threshold can be obtained by checking the card-square distribution table or based on experience.

[0151] Among them, the false alarm rate refers to the probability of detecting a fault when the device is not faulty, that is, the false alarm rate can be determined based on the ratio of the number of inconsistencies between the actual fault diagnosis results detected and the devices with faults in the fault diagnosis results to the number of devices with faults in the fault diagnosis results.

[0152] In practical applications, since the fault diagnosis result at the current moment needs to be determined by the fault diagnosis threshold, that is, the fault diagnosis result at the current moment has not been determined, the fault diagnosis result for the GNSS sensor at the previous moment and the prediction result for the GNSS sensor at the previous moment can be obtained at this time, as well as the state quantity collected in real time for the GNSS sensor at the previous moment. The prediction result for the GNSS sensor at the previous moment is compared with the state quantity collected in real time for the GNSS sensor at the previous moment. Specifically, the prediction result is compared with the actually collected data to see if they are consistent, and the actual fault diagnosis result for the GNSS sensor at the previous moment is determined.

[0153] Then, the false alarm rate of the GNSS sensor can be obtained by using the ratio of the number of faulty sensors that are inconsistent with the actual fault diagnosis results and the fault diagnosis results for the GNSS sensor at the previous moment to the number of faulty sensors in the fault diagnosis results for the GNSS sensor at the previous moment.

[0154] Step 203, using the comparison result of the prediction result at the current moment and the fault diagnosis threshold, to determine the fault diagnosis result of the target sensor in the GNSS sensor at the current moment;

[0155] In the embodiment of the present invention, the prediction result at the current time after the time update can be compared with the data obtained by the sensor in real time, and the fault diagnosis of the GNSS sensor can be performed to determine whether the sensor has a fault.

[0156] Among them, the fault diagnosis threshold is determined based on the real-time collected data. At this time, as shown in Formula 4, the fault diagnosis amount for the target sensor is constructed by using the prediction result at the current moment, and then the prediction result is compared with the false alarm rate for judgment.

[0157] As shown in Formula 5, in one case, if the fault diagnosis amount of the target sensor is greater than or equal to the fault diagnosis threshold, it can be determined that there is a fault in the target sensor in the GNSS sensor; in another case, if the fault diagnosis amount of the target sensor is less than the fault diagnosis threshold, it can be determined that there is no fault in the target sensor in the GNSS sensor, that is, the target sensor is a normal sensor.

[0158] Step 204 , based on the fault diagnosis result, the prediction result at the current moment is fused to obtain the estimation result at the current moment, and the estimation result at the current moment is used as the historical estimation result when performing fault diagnosis on the GNSS sensor at the next moment.

[0159] After obtaining the fault diagnosis result of the target sensor, the prediction result at the current moment can be fused based on the fault diagnosis result to obtain the estimation result at the current moment, so that the estimation result at the current moment can be used as the historical estimation result when performing fault diagnosis on the GNSS sensor at the next moment, so that the existing historical estimation result can be used as the initial value of the fault diagnosis work, thereby ensuring the iterative and continuous fault diagnosis of the GNSS sensor.

[0160] In practical applications, when fusing the estimation results at the current moment, the estimation results of the normal sensors can usually be fused. However, the estimation results at the current moment are predicted results. Therefore, before fusion, as shown in Formula 10, the estimation results of the normal sensors need to be measured and updated. Then, it can be expressed as obtaining the prediction results of the normal sensors at the current moment, measuring and updating the prediction results of the normal sensors at the current moment, and obtaining the estimation results of the normal sensors after the measurement and update at the current moment. Then, as shown in Formula 11, the estimation results of the normal sensors after the measurement and update at the current moment are fused to obtain the estimation results at the current moment.

[0161] Exemplarily, the specific implementation process can be expressed as the information allocation, time update, measurement update, fault diagnosis, and information fusion process of the federated filtering:

[0162] First, select the initial value of the state quantity. The initial values ​​of position and velocity can be selected according to the GNSS RTK solution, and the initial value of acceleration can be set to 0 to obtain X0 and P0.

[0163] In the information allocation process, the information of each sub-filter can be allocated. For example, at the initial moment, it can be expressed as the following formula:

[0164]

[0165] i=1,2,...N

[0166] The remaining moments can be expressed as follows:

[0167]

[0168] In the time update process, the historical fusion data of the previous moment can be used to update the sub-filter data of the corresponding time, mainly updating Xn and Pn. The specific updated equation can be as follows:

[0169]

[0170]

[0171] i=1,2,...N

[0172] Among them, the state transition matrix F i,k / k-1 and process noise matrix Q k-1 for:

[0173]

[0174]

[0175] In the fault diagnosis process, each sub-filter is diagnosed. Construct the fault diagnosis quantity λ i,k :

[0176]

[0177] γ i,k =Z i,k -H i,k X i,k / k-1

[0178]

[0179] Among them, A is just an intermediate variable and has no practical significance; and I3 is just a general formula, representing a 3D unit matrix, and the specific expression of H is related to the diagnostic model, such as the detection position and speed:

[0180]

[0181] Then, we can use λ i,k With threshold T D Compare and diagnose the fault:

[0182]

[0183] It should be noted that the process is not limited to judging by a fixed threshold, and may also be judged by other methods, which is not limited by the embodiment of the present invention.

[0184] During the measurement update process, all normally functioning GNSS devices can be measured and updated, and for GNSS devices diagnosed as faulty, the diagnosis results can be output.

[0185]

[0186]

[0187]

[0188] During the information fusion process, the normal sub-filter estimation results can be fused.

[0189]

[0190]

[0191] Then, we can return to the information distribution process and continue to diagnose faults. That is, we use the fused estimation results at the current moment as the historical estimation results for diagnosing faults on the GNSS sensor at the next moment, and use the existing historical estimation results as the initial values ​​for the fault diagnosis work to ensure iterative and continuous fault diagnosis of the GNSS sensor, and realize fault diagnosis of a single sensor or multiple sensors.

[0192] It should be noted that this example takes the detection of position speed as an example, and the method for GNSS heading diagnosis can be deduced by analogy, which is not elaborated in the embodiment of the present invention.

[0193] In an embodiment of the present invention, the navigation information output instruction of the GNSS sensor can be responded to, and the prediction result of the target sensor in the GNSS sensor at the current moment can be updated by acquiring the historical estimation result for the GNSS sensor at the previous moment, and the fault diagnosis result of the target sensor can be determined based on the comparison between the prediction result at the current moment and the fault diagnosis threshold, so as to fuse the prediction result at the current moment to obtain the estimation result at the current moment, and use the fused estimation result at the current moment as the historical estimation result when performing fault diagnosis on the GNSS sensor at the next moment, and use the existing historical estimation result as the initial value of the fault diagnosis work, so as to ensure iterative and continuous fault diagnosis of the GNSS sensor, and realize fault diagnosis of a single sensor or multiple sensors separately.

[0194] It should be noted that, for the sake of simplicity, the method embodiments are described as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present invention are not limited by the order of the actions described, because according to the embodiments of the present invention, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily required by the embodiments of the present invention.

[0195] Reference Figure 4 , shows a structural block diagram of an embodiment of a fault diagnosis device for a GNSS sensor of the present invention, which may specifically include the following modules:

[0196] A historical estimation result acquisition module 401 is used to respond to the navigation information output instruction of the GNSS sensor and acquire the historical estimation result of the GNSS sensor at the last moment;

[0197] A prediction result updating module 402, configured to update a prediction result of a target sensor in the GNSS sensor at a current moment according to a historical estimation result for the GNSS sensor at a previous moment;

[0198] A fault diagnosis result determination module 403 is used to obtain a fault diagnosis threshold value, and determine the fault diagnosis result of the target sensor according to the fault diagnosis threshold value and the prediction result at the current moment;

[0199] The prediction result fusion module 403 is used to fuse the prediction result at the current moment according to the fault diagnosis result to obtain the estimation result at the current moment; the estimation result at the current moment is used as the historical estimation result when performing fault diagnosis on the GNSS sensor at the next moment.

[0200] In one embodiment of the present invention, the prediction result updating module 402 may include the following submodules:

[0201] The prediction result updating submodule is used to update the prediction result of the target sensor in the GNSS sensor at the current moment according to the historical estimation result for the GNSS sensor at the previous moment when the historical estimation result at the initial moment is obtained.

[0202] In one embodiment of the present invention, the prediction result at the current moment is determined based on a state quantity prediction value, wherein the state quantity prediction value includes a state to be estimated and a state variance matrix for each sensor in the GNSS sensor; the target sensor includes at least one target sensor;

[0203] The prediction result update submodule may include the following units:

[0204] A prediction result updating unit is used to adopt the state to be estimated in the historical estimation result of the previous moment and the state variance matrix in the historical estimation result to respectively update the state to be estimated and the state variance matrix of the at least one target sensor at the current moment, so as to obtain the prediction result of the at least one target sensor at the current moment; wherein the historical estimation result for the GNSS sensor at the previous moment includes the historical estimation result at the initial moment, and / or the historical estimation result for a single sensor of the GNSS sensor at the previous moment, and / or the historical fusion data for at least two sensors of the GNSS sensor at the previous moment.

[0205] In one embodiment of the present invention, the fault diagnosis result determination module 403 may include the following submodules:

[0206] A fault diagnosis threshold acquisition submodule, used to acquire a false alarm rate of the GNSS sensor, and to acquire a fault diagnosis threshold using the false alarm rate;

[0207] The fault diagnosis result determination submodule is used to determine the fault diagnosis result of the target sensor in the GNSS sensor at the current moment by using the comparison result of the prediction result at the current moment and the fault diagnosis threshold.

[0208] In one embodiment of the present invention, the fault diagnosis threshold acquisition submodule may include the following units:

[0209] A state quantity acquisition unit, used to acquire a fault diagnosis result for the GNSS sensor at a previous moment and a prediction result for the GNSS sensor at a previous moment, and to acquire a state quantity collected in real time by the GNSS sensor at a previous moment;

[0210] an actual fault diagnosis result determination unit, configured to compare the prediction result for the GNSS sensor at the last moment with the state quantity collected in real time for the GNSS sensor at the last moment, and determine the actual fault diagnosis result for the GNSS sensor at the last moment;

[0211] The false alarm rate calculation unit is used to obtain the false alarm rate of the GNSS sensor by using the ratio of the number of faulty sensors that are inconsistent between the actual fault diagnosis result and the fault diagnosis result for the GNSS sensor at the previous moment and the number of faulty sensors in the fault diagnosis result for the GNSS sensor at the previous moment.

[0212] In one embodiment of the present invention, the fault diagnosis result determination submodule may include the following units:

[0213] A fault diagnosis quantity construction unit, used to construct a fault diagnosis quantity for a target sensor using the prediction result at the current moment;

[0214] a first fault diagnosis unit, configured to determine that a target sensor in the GNSS sensor has a fault when a fault diagnosis amount of the target sensor is greater than or equal to the fault diagnosis threshold;

[0215] The first fault diagnosis unit is configured to determine that a target sensor in the GNSS sensor has no fault when the fault diagnosis amount of the target sensor is less than the fault diagnosis threshold.

[0216] In one embodiment of the present invention, the prediction result fusion module 403 may include the following submodules:

[0217] A prediction result acquisition submodule, used to acquire the prediction result of the normal sensor at the current moment based on the fault diagnosis result;

[0218] A measurement update submodule, used to perform measurement update on the prediction result of the normal sensor at the current moment, to obtain an estimation result of the normal sensor after the measurement update at the current moment;

[0219] The estimation result fusion submodule is used to fuse the estimation result after the normal sensor is measured and updated at the current moment to obtain the estimation result at the current moment.

[0220] The fault diagnosis device for a GNSS sensor proposed in an embodiment of the present invention, when responding to a navigation information output instruction of a GNSS sensor, obtains a historical estimation result for the GNSS sensor at a previous moment, updates a prediction result of a target sensor in the GNSS sensor at a current moment, and determines the fault diagnosis result of the target sensor based on a comparison between the prediction result at the current moment and a fault diagnosis threshold, so as to fuse the prediction result at the current moment to obtain an estimation result at the current moment, and uses the fused estimation result at the current moment as a historical estimation result when performing fault diagnosis on the GNSS sensor at the next moment, and uses the existing historical estimation result as an initial value for the fault diagnosis work, thereby ensuring iterative and continuous fault diagnosis of the GNSS sensor, and realizing fault diagnosis of a single sensor or multiple sensors separately.

[0221] As for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0222] The embodiment of the present invention also provides a navigation system, comprising: a processor; a memory storing executable instructions of the processor; wherein the processor is coupled to the memory, and is used to read the program instructions stored in the memory, and in response, performs the following operations: in response to the navigation information output instruction of the GNSS sensor, obtain the historical estimation result for the GNSS sensor at the last moment; according to the historical estimation result for the GNSS sensor at the last moment, update the prediction result of the target sensor in the GNSS sensor at the current moment; obtain the fault diagnosis threshold, and determine the fault diagnosis result of the target sensor according to the fault diagnosis threshold and the prediction result at the current moment; according to the fault diagnosis result, fuse the prediction result at the current moment to obtain the estimation result at the current moment, so as to use the fused estimation result at the current moment as the historical estimation result when the fault diagnosis is performed on the GNSS sensor at the next moment, and based on the existing historical estimation result as the initial value of the fault diagnosis work, ensure the iterative and continuous fault diagnosis of the GNSS sensor, and realize the fault diagnosis of a single sensor or multiple sensors separately. It should be noted that this navigation system can be applied to any means of transportation involving the need to perform fault diagnosis on its GNSS sensor, such as vehicles, flying cars, aircraft, etc., and this is not limited by the embodiment of the present invention.

[0223] The embodiment of the present invention further provides a means of transportation, including:

[0224] The present invention includes a fault diagnosis device for the GNSS sensor, a processor, a memory, and a computer program stored in the memory and capable of running on the processor. When the computer program is executed by the processor, each process of the fault diagnosis method embodiment of the GNSS sensor is implemented, and the same technical effect can be achieved. To avoid repetition, it is not repeated here. It should be noted that the vehicles involved include any vehicles using GNSS sensors, such as vehicles, flying cars, aircraft, etc. All vehicles using GNSS sensors need to perform fault diagnosis on their GNSS sensors, and the embodiments of the present invention are not limited to this.

[0225] An embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, each process of the above-mentioned GNSS sensor fault diagnosis method embodiment is implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0226] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.

[0227] It will be appreciated by those skilled in the art that the embodiments of the present invention may be provided as methods, devices, or computer program products. Therefore, the embodiments of the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the embodiments of the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0228] The embodiments of the present invention are described with reference to the flowcharts and / or block diagrams of the methods, terminal devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0229] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0230] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0231] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the embodiments of the present invention.

[0232] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0233] The above describes in detail a GNSS sensor fault diagnosis method, a GNSS sensor fault diagnosis device, a corresponding navigation system and a corresponding means of transportation provided by the present invention. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A GNSS sensor fault diagnosis method, characterized in that: The method comprises: In response to the navigation information output instruction of the GNSS sensor, obtaining a historical estimation result for the GNSS sensor at a previous moment; According to the historical estimation result of the GNSS sensor at the previous moment, updating the prediction result of the target sensor in the GNSS sensor at the current moment; Acquire a fault diagnosis threshold, and determine a fault diagnosis result of the target sensor according to the fault diagnosis threshold and the prediction result at the current moment; According to the fault diagnosis result, the prediction result at the current moment is fused to obtain the estimation result at the current moment, including: based on the fault diagnosis result, obtaining the prediction result of the normal sensor at the current moment; measuring and updating the prediction result of the normal sensor at the current moment to obtain the estimation result of the normal sensor after the measurement update at the current moment; fusing the estimation result of the normal sensor after the measurement update at the current moment to obtain the estimation result at the current moment; the estimation result at the current moment is used as the historical estimation result when performing fault diagnosis on the GNSS sensor at the next moment; wherein the historical estimation result is used as the initial value iteration of the fault diagnosis work to perform fault diagnosis.

2. The method according to claim 1, characterized in that The updating of the prediction result of the target sensor in the GNSS sensor at the current moment according to the historical estimation result for the GNSS sensor at the previous moment includes: When the historical estimation result at the initial moment is obtained, the prediction result of the target sensor in the GNSS sensor at the current moment is updated according to the historical estimation result for the GNSS sensor at the previous moment.

3. The method according to claim 1 or 2, characterized in that: The prediction result at the current moment is determined based on a state quantity prediction value, wherein the state quantity prediction value includes a state to be estimated and a state variance matrix for each sensor in the GNSS sensor; the target sensor includes at least one target sensor; The updating of the prediction result of the target sensor in the GNSS sensor at the current moment according to the historical estimation result for the GNSS sensor at the previous moment includes: Adopting the state to be estimated in the historical estimation result of the previous moment and the state variance matrix in the historical estimation result, respectively updating the state to be estimated and the state variance matrix of the at least one target sensor at the current moment, and obtaining the prediction result of the at least one target sensor at the current moment; wherein, the historical estimation result for the GNSS sensor at the previous moment includes the historical estimation result at the initial moment, and / or the historical estimation result for a single sensor of the GNSS sensor at the previous moment, and / or the historical fusion data for at least two sensors of the GNSS sensor at the previous moment.

4. The method according to claim 1, characterized in that: Determining the fault diagnosis result of the target sensor according to the fault diagnosis threshold and the prediction result at the current moment includes: Obtaining a false alarm rate of the GNSS sensor, and using the false alarm rate to obtain a fault diagnosis threshold; The fault diagnosis result of the target sensor in the GNSS sensor at the current moment is determined by using the comparison result of the prediction result at the current moment and the fault diagnosis threshold.

5. The method according to claim 4, characterized in that The obtaining of the false alarm rate of the GNSS sensor comprises: Obtaining a fault diagnosis result for the GNSS sensor at the last moment and a prediction result for the GNSS sensor at the last moment, and obtaining a state quantity collected in real time for the GNSS sensor at the last moment; The prediction result for the GNSS sensor at the last moment is compared with the state quantity collected in real time for the GNSS sensor at the last moment to determine the actual fault diagnosis result for the GNSS sensor at the last moment; The false alarm rate of the GNSS sensor is obtained by using the ratio of the number of faulty sensors that are inconsistent between the actual fault diagnosis result and the fault diagnosis result for the GNSS sensor at the previous moment to the number of faulty sensors in the fault diagnosis result for the GNSS sensor at the previous moment.

6. The method according to claim 4, characterized in that The method of using the comparison result of the prediction result at the current moment and the fault diagnosis threshold to determine the fault diagnosis result of the GNSS sensor at the current moment includes: Using the prediction result at the current moment to construct a fault diagnosis quantity for the target sensor; If the fault diagnosis amount of the target sensor is greater than or equal to the fault diagnosis threshold, determining that a fault exists in the target sensor in the GNSS sensor; And / or, if the fault diagnosis amount of the target sensor is less than the fault diagnosis threshold, it is determined that the target sensor in the GNSS sensor has no fault.

7. A GNSS sensor fault diagnosis device, characterized in that: The device comprises: A historical estimation result acquisition module, used to respond to the navigation information output instruction of the GNSS sensor and acquire the historical estimation result for the GNSS sensor at the last moment; A prediction result updating module, used for updating the prediction result of the target sensor in the GNSS sensor at the current moment according to the historical estimation result for the GNSS sensor at the previous moment; A fault diagnosis result determination module, used to obtain a fault diagnosis threshold value, and determine the fault diagnosis result of the target sensor according to the fault diagnosis value and the prediction result at the current moment; A prediction result fusion module is used to fuse the prediction result at the current moment according to the fault diagnosis result to obtain the estimation result at the current moment, including: a prediction result acquisition submodule, used to obtain the prediction result of the normal sensor at the current moment based on the fault diagnosis result; a measurement update submodule, used to perform measurement update on the prediction result of the normal sensor at the current moment, and obtain the estimation result of the normal sensor after the measurement update at the current moment; an estimation result fusion submodule, used to fuse the estimation result of the normal sensor after the measurement update at the current moment, and obtain the estimation result at the current moment; the estimation result at the current moment is used as the historical estimation result when performing fault diagnosis on the GNSS sensor at the next moment; wherein the historical estimation result is used as the initial value iteration of the fault diagnosis work to perform fault diagnosis.

8. A navigation system, characterized in that: include: processor; a memory storing processor executables; The processor is coupled to the memory and configured to read program instructions stored in the memory and, in response, perform the following operations: In response to the navigation information output instruction of the GNSS sensor, obtain the historical estimation result of the GNSS sensor at the last moment; According to the historical estimation result of the GNSS sensor at the previous moment, updating the prediction result of the target sensor in the GNSS sensor at the current moment; Acquire a fault diagnosis threshold, and determine a fault diagnosis result of the target sensor according to the fault diagnosis threshold and the prediction result at the current moment; According to the fault diagnosis result, the prediction result at the current moment is fused to obtain the estimation result at the current moment, including: based on the fault diagnosis result, obtaining the prediction result of the normal sensor at the current moment; measuring and updating the prediction result of the normal sensor at the current moment to obtain the estimation result of the normal sensor after the measurement update at the current moment; fusing the estimation result of the normal sensor after the measurement update at the current moment to obtain the estimation result at the current moment; the estimation result at the current moment is used as the historical estimation result when performing fault diagnosis on the GNSS sensor at the next moment; wherein the historical estimation result is used as the initial value iteration of the fault diagnosis work to perform fault diagnosis.

9. A means of transport, characterized in that: include: The fault diagnosis device of the GNSS sensor according to claim 7, the processor, the memory and the computer program stored in the memory and capable of running on the processor, wherein when the computer program is executed by the processor, the fault diagnosis method of the GNSS sensor according to any one of claims 1 to 6 is implemented.

Images