Estimation Method, Device, Electronic Device and Storage Medium for Vehicle Positioning Status

Abstract

The present application relates to a method, device, electronic device and storage medium for estimating the positioning state of a vehicle. Among them, the method includes: when the positioning system is a non-linear model, establishing a state equation and a measurement equation of the non-linear system to calculate the predicted value of the state error covariance and the cross-correlation error covariance at time k; based on a preset statistical linear model, using the predicted value of the state error covariance and the cross-correlation error covariance to construct a pseudo-measurement equation; determining a mixed Agnesi criterion kernel function, and combining the preset fixed-point iteration condition and the pseudo-measurement equation to perform fixed-point iteration to obtain the optimal state error covariance and the state estimate value at time k, so as to analyze the positioning state of the target vehicle. Thus, the problems that the existing integrated positioning technology does not consider the situation where both the system and the measurement information have non-high noise, cannot suppress the interference of process non-Gaussian noise and measurement non-Gaussian noise at the same time, and the positioning accuracy and stability are relatively low are solved.

Description

Technical Field

[0001] The present application relates to the technical field of vehicle positioning, and particularly relates to a method, device, electronic device and storage medium for estimating the positioning state of a vehicle. Background Art

[0002] Accurate and robust positioning is very important for vehicle and intelligent transportation system applications. Currently, many vehicle positioning technologies for urban environments use various sensors, such as radar, lidar, ultrasonic, high-definition maps, inertial sensors, and cameras. Among them, the integrated navigation system (INS / GNSS) of the inertial navigation system (INS) and the global navigation satellite system (GNSS) is the most widely used and most promising vehicle positioning method at present.

[0003] However, there are system uncertainties in the actual vehicle positioning system application, which are mainly reflected in two aspects: inaccurate system mathematical models and unknown noise interference. Due to the complex and changeable external application environment, there are usually many uncertain factors that make it impossible to accurately establish a system mathematical model. When the system equipment ages or is affected by other external interferences, the system noise and measurement noise have non-Gaussian noise interference, resulting in a decrease or even divergence in the estimation accuracy of the algorithm. Therefore, in-depth research on the ability of non-linear state estimation algorithms to suppress non-Gaussian noise interference is one of the important ways to improve the estimation accuracy of the algorithm.

[0004] In actual engineering applications, the performance of traditional INS / GNSS integrated navigation systems may be poor when GNSS is interrupted, and they are less robust to measurement noise in changing urban environments. When the accuracy of positioning information is affected by data noise, they cannot meet the high-precision positioning requirements of autonomous driving. The data output by GNSS is used as the measurement information of this integrated navigation system, but the satellite signal is subject to non-Gaussian noise interference during transmission, propagation, and reception. In addition, due to sudden maneuvers of the vehicle carrier or abnormal operation of the GNSS receiver device, the GNSS data may have non-Gaussian noise with large outliers.

[0005] Currently, the current vehicle-mounted integrated positioning state estimation methods, such as the extended Kalman filter (EKF), unscented Kalman filter (UKF), and cubature Kalman filter (CKF), are all proposed based on the assumption of Gaussian noise background. In the non-Gaussian noise background, the estimation performance of the above filters will be greatly impacted, and even the filtering may diverge and the navigation accuracy may decrease.

[0006] To improve the robust performance of state estimation algorithms in non-Gaussian noise environments, scholars at home and abroad have proposed a series of improved robust state estimation algorithms based on optimal estimation in recent years. Among them, the robust state estimation algorithm based on the Maximum Correntropy Criterion (MCC) has become a current research hotspot due to its excellent ability to suppress non-Gaussian noise. However, these algorithms have shown limitations in dealing with complex noise environments containing large outliers of non-Gaussian noise and the simultaneous presence of process non-Gaussian noise and measurement non-Gaussian noise.

[0007] In summary, non-Gaussian noise is an inevitable problem in the vehicle integrated positioning process. Most of the existing integrated positioning methods propose solutions for measurement non-Gaussian noise problems and do not consider the situation where non-Gaussian noise exists simultaneously in system and measurement information. They cannot suppress the interference of process non-Gaussian noise and measurement non-Gaussian noise simultaneously, resulting in low positioning accuracy and stability, which urgently need to be solved. Summary of the Invention

[0008] This application provides a method, device, electronic device, and storage medium for estimating the vehicle positioning state to solve the problems that existing integrated positioning technologies do not consider the situation where non-Gaussian noise exists simultaneously in system and measurement information, cannot suppress the interference of process non-Gaussian noise and measurement non-Gaussian noise simultaneously, and have low positioning accuracy and stability.

[0009] The first aspect of the embodiments of this application provides a method for estimating the vehicle positioning state, including the following steps: detecting the type of the positioning system model of the constructed integrated navigation system, and establishing the state equation and measurement equation of the integrated navigation system when it is detected that the type is a non-linear model type; calculating the state error covariance and cross-correlation error covariance of the non-linear model at time k according to the state equation and the measurement equation, and constructing a pseudo-measurement equation of the non-linear model by using the state error covariance and the cross-correlation error covariance, where k is a positive integer; determining a mixed Agnesi criterion kernel function, and performing a fixed-point iteration operation through a preset fixed-point iteration condition, the pseudo-measurement equation, and the mixed Agnesi criterion kernel function to obtain the Kalman gain of the non-linear model at time k; calculating the optimal state error covariance and the optimal state estimate value of the non-linear model according to the Kalman gain, so that the integrated navigation system estimates the positioning state of the target vehicle according to the optimal state error covariance and the optimal state estimate value.

[0010] Optionally, in an embodiment of the present application, it further includes: when it is detected that the type is a linear model type, establishing a linear state equation and a linear measurement equation of the integrated navigation system; performing a fixed-point iteration operation based on the linear state equation and the linear measurement equation to obtain the optimal state error covariance and the optimal state estimate of the linear model; estimating the positioning state of the target vehicle according to the optimal state error covariance and the optimal state estimate of the linear model.

[0011] Optionally, in an embodiment of the present application, the calculating the state error covariance and the cross-correlation error covariance of the nonlinear model at the k-th moment according to the state equation and the measurement equation respectively includes: calculating the first cubature points at the (k-1)-th moment, and propagating the first cubature points through the state equation to obtain the first updated cubature points; based on the first updated cubature points, calculating the state estimate and the state error covariance of the nonlinear model at the k-th moment; calculating the second cubature points at the (k-1)-th moment, and propagating the second cubature points through the measurement equation to obtain the second updated cubature points; based on the second updated cubature points, calculating the cross-correlation error covariance of the nonlinear model at the k-th moment.

[0012] Optionally, in an embodiment of the present application, the performing a fixed-point iteration operation through the preset fixed-point iteration condition, the pseudo-measurement equation and the mixed Agnesi criterion kernel function to obtain the Kalman gain of the nonlinear model at the k-th moment includes: determining the initial value of the preset fixed-point iteration condition, and based on the initial value, calculating the Kalman gain of the nonlinear model at the i-th fixed-point iteration at the k-th moment, where i is a positive integer.

[0013] Optionally, in an embodiment of the present application, the calculating the optimal state error covariance and the optimal state estimate of the nonlinear model according to the Kalman gain includes: based on the Kalman gain, calculating the state estimate and the state error covariance of the nonlinear model at the i-th fixed-point iteration at the k-th moment; based on the fixed-point iteration condition, determining whether the fixed-point iteration ends, where if it ends, the state estimate and the state error covariance at the end of the fixed-point iteration are used as the optimal state error covariance and the optimal state estimate of the nonlinear model.

[0014] The second aspect of the present application provides an estimation device for the positioning state of a vehicle, including: a detection module, configured to detect the type of the positioning system model of the constructed integrated navigation system, and establish the state equation and measurement equation of the integrated navigation system when detecting that the type is a non-linear model type; a first construction module, configured to calculate the state error covariance and cross-correlation error covariance of the non-linear model at the k-th moment according to the state equation and the measurement equation respectively, and construct the pseudo-measurement equation of the non-linear model by using the state error covariance and the cross-correlation error covariance, where k is a positive integer; a first iteration module, configured to determine a mixed Agnesi criterion kernel function, and perform a fixed-point iteration operation through a preset fixed-point iteration condition, the pseudo-measurement equation and the mixed Agnesi criterion kernel function to obtain the Kalman gain of the non-linear model at the k-th moment; a first estimation module, configured to calculate the optimal state error covariance and the optimal state estimate value of the non-linear model according to the Kalman gain, so that the integrated navigation system estimates the positioning state of the target vehicle according to the optimal state error covariance and the optimal state estimate value.

[0015] Optionally, in an embodiment of the present application, it further includes: a first construction module, configured to establish the linear state equation and linear measurement equation of the integrated navigation system when detecting that the type is a linear model type; a second iteration module, configured to perform a fixed-point iteration operation based on the linear state equation and the linear measurement equation to obtain the optimal state error covariance and the optimal state estimate value of the linear model; a second estimation module, configured to estimate the positioning state of the target vehicle according to the optimal state error covariance and the optimal state estimate value of the linear model.

[0016] Optionally, in an embodiment of the present application, the first construction module includes: a first calculation unit, configured to calculate the first cubature point at the (k - 1)-th moment, and propagate the first cubature point through the state equation to obtain a first updated cubature point; a second calculation unit, configured to calculate the state estimate value and state error covariance of the non-linear model at the k-th moment based on the first updated cubature point; a third calculation unit, configured to calculate the second cubature point at the (k - 1)-th moment, and propagate the second cubature point through the measurement equation to obtain a second updated cubature point; a fourth calculation unit, configured to calculate the cross-correlation error covariance of the non-linear model at the k-th moment based on the second updated cubature point.

[0017] Optionally, in an embodiment of the present application, the first iteration module includes: an operation unit, configured to determine the initial value of the preset fixed-point iteration condition, and calculate the Kalman gain of the non-linear model at the i-th fixed-point iteration at the k-th moment based on the initial value, where i is a positive integer.

[0018] Optionally, in an embodiment of the present application, the first estimation module includes: a fifth calculation unit configured to calculate a state estimation value and a state error covariance of the nonlinear model under the i-th fixed-point iteration at the k-th moment based on the Kalman gain; a determination unit configured to determine whether the fixed-point iteration ends based on the fixed-point iteration condition, where if it ends, the state estimation value and the state error covariance at the end of the fixed-point iteration are used as the optimal state error covariance and the optimal state estimation value of the nonlinear model.

[0019] An embodiment of the third aspect of the present application provides an electronic device, including: a memory, a processor, and a computer program stored on the memory and executable on the processor, where the processor executes the program to implement the method for estimating the vehicle positioning state as described in the above embodiment.

[0020] An embodiment of the fourth aspect of the present application provides a computer-readable storage medium storing a computer program, and when the program is executed by a processor, it implements the method for estimating the vehicle positioning state as described above.

[0021] Therefore, the embodiments of the present application have the following beneficial effects:

[0022] The embodiments of the present application can, when the positioning system is a nonlinear model, establish a state equation and a measurement equation of the nonlinear system to calculate a predicted value of the state error covariance and a cross-correlation error covariance at the k-th moment; based on a preset statistical linear model, construct a pseudo-measurement equation by using the predicted value of the state error covariance and the cross-correlation error covariance; determine a mixed Agnesi criterion kernel function, and perform fixed-point iteration in combination with the preset fixed-point iteration condition and the pseudo-measurement equation to obtain the optimal state error covariance and the state estimation value at the k-th moment for analyzing the positioning state of the target vehicle. The present application constructs a mixed Agnesi kernel function and combines the CKF with the maximum mixed Agnesi criterion, so that the CKF algorithm can simultaneously suppress the interference of process non-Gaussian noise and measurement non-Gaussian noise during the vehicle integrated positioning process, greatly improving the positioning accuracy and stability. Thus, the problems that the existing integrated positioning technology does not consider the situation where both the system and the measurement information have non-Gaussian noise, cannot simultaneously suppress the interference of process non-Gaussian noise and measurement non-Gaussian noise, and the positioning accuracy and stability are relatively low are solved.

[0023] Additional aspects and advantages of the present application will be given in part in the following description, become apparent in part from the following description, or be learned through the practice of the present application. Description of the Drawings

[0024] The above and / or additional aspects and advantages of the present application will become apparent and be readily understood from the following description of the embodiments in conjunction with the drawings, where:

[0025] Figure 1 Flow chart of a method for estimating the vehicle positioning status provided according to an embodiment of the present application;

[0026] Figure 2 Schematic diagram of a process based on a hybrid Agnesi function provided by an embodiment of the present application;

[0027] Figure 3 Schematic diagram of RMSE comparison of the position in Case 1 of a target tracking embodiment provided by an embodiment of the present application;

[0028] Figure 4 Schematic diagram of RMSE comparison of the speed in Case 1 of a target tracking embodiment provided by an embodiment of the present application;

[0029] Figure 5 Schematic diagram of RMSE comparison of the turning rate in Case 1 of a target tracking embodiment provided by an embodiment of the present application;

[0030] Figure 6 Schematic diagram of RMSE comparison of the position in Case 2 of a target tracking embodiment provided by an embodiment of the present application;

[0031] Figure 7 Schematic diagram of RMSE comparison of the speed in Case 2 of a target tracking embodiment provided by an embodiment of the present application;

[0032] Figure 8 Schematic diagram of RMSE comparison of the turning rate in Case 2 of a target tracking embodiment provided by an embodiment of the present application;

[0033] Figure 9 Schematic diagram of RMSE comparison of the roll angle in a vehicle positioning embodiment provided by an embodiment of the present application;

[0034] Figure 10 Schematic diagram of RMSE comparison of the pitch angle in a vehicle positioning embodiment provided by an embodiment of the present application;

[0035] Figure 11 Schematic diagram of RMSE comparison of the heading angle in a vehicle positioning embodiment provided by an embodiment of the present application;

[0036] Figure 12 Schematic diagram of RMSE comparison of the longitude in a vehicle positioning embodiment provided by an embodiment of the present application;

[0037] Figure 13 Schematic diagram of RMSE comparison of the latitude in a vehicle positioning embodiment provided by an embodiment of the present application;

[0038] Figure 14 An exemplary diagram of an estimation device for a vehicle positioning state according to an embodiment of the present application;

[0039] Figure 15 A schematic structural diagram of an electronic device provided by an embodiment of the present application.

[0040] Wherein, 10 - an estimation device for a vehicle positioning state, 100 - a detection module, 200 - a first construction module, 300 - an iteration module, 400 - an estimation module, 1501 - a memory, 1502 - a processor, 1503 - a communication interface. Specific embodiments

[0041] Embodiments of the present application will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present application and should not be construed as limiting the present application.

[0042] The vehicle positioning state estimation method, device, electronic device, and storage medium according to embodiments of the present application will be described below with reference to the accompanying drawings. To address the problems mentioned in the above background art, the present application provides a vehicle positioning state estimation method. In this method, when the positioning system is a non-linear model, the state equation and measurement equation of the non-linear system are established to calculate the predicted value of the state error covariance and the cross-correlation error covariance at time k; based on a preset statistical linear model, a pseudo-measurement equation is constructed using the predicted value of the state error covariance and the cross-correlation error covariance; a mixed Agnesi criterion kernel function is determined, and fixed-point iteration is performed in combination with the preset fixed-point iteration condition and the pseudo-measurement equation to obtain the optimal state error covariance and state estimation value at time k for analyzing the positioning state of the target vehicle. By constructing a mixed Agnesi kernel function and combining the CKF with the maximum mixed Agnesi criterion, the CKF algorithm can simultaneously suppress the process non-Gaussian noise and measurement non-Gaussian noise interference during the vehicle integrated positioning process, greatly improving the positioning accuracy and stability. Thus, the problems that the existing integrated positioning technology does not consider the situation of simultaneous non-Gaussian noise in the system and measurement information, cannot simultaneously suppress the process non-Gaussian noise and measurement non-Gaussian noise interference, and has low positioning accuracy and stability are solved.

[0043] Specifically, Figure 1 A flowchart of a vehicle positioning state estimation method provided by an embodiment of the present application.

[0044] As Figure 1 shown, the vehicle positioning state estimation method includes the following steps:

[0045] In step S101, detect the type of the positioning system model of the constructed integrated navigation system, and establish the state equation and measurement equation of the integrated navigation system when it is detected that the type is a non-linear model type.

[0046] In the embodiment of the present application, first, the type of the positioning system model of the integrated navigation system can be analyzed and judged. When it is a non-linear model, the state equation and measurement equation of the non-linear model of the integrated navigation system can be established. The expression of the discrete-time non-linear system model with additive noise in the embodiment of the present application is as follows:

[0047]

[0048] Wherein, and are the state vector and measurement vector at time k respectively, and k is a positive integer; f(·) and h(·) are the non-linear state equation and non-linear measurement equation respectively; and are the uncorrelated system noise and measurement noise respectively, both of which follow a Gaussian distribution, and their means are as shown in Equation (2), and their covariances are as shown in Equations (3) and (4). The initial values of the state estimate and state error covariance are as shown in Equations (5) and (6):

[0049]

[0050]

[0051]

[0052]

[0053]

[0054] Thus, the embodiment of the present application provides a reliable theoretical guidance and basis for the implementation of subsequent non-linear time update and other operations by establishing the state equation and measurement equation of the system.

[0055] In step S102, calculate the state error covariance and cross-correlation error covariance of the non-linear model at time k according to the state equation and measurement equation respectively, and construct the pseudo-measurement equation of the non-linear model by using the state error covariance and cross-correlation error covariance, where k is a positive integer.

[0056] After establishing the state equation and measurement equation of the integrated navigation system, further, the embodiments of the present application can also obtain the state prediction value and the state error covariance prediction value at a certain time point by calculating the cubature points at the previous moment of that time point, and then perform the measurement update operation to realize the construction of the pseudo-measurement equation and the calculation of the pseudo-measurement noise covariance.

[0057] Optionally, in an embodiment of the present application, calculating the state error covariance and the cross-correlation error covariance of the nonlinear model at the kth moment according to the state equation and the measurement equation includes calculating the first cubature points at the (k - 1)th moment, and propagating the first cubature points through the state equation to obtain the first updated cubature points; based on the first updated cubature points, calculating the state estimation value and the state error covariance of the nonlinear model at the kth moment; calculating the second cubature points at the (k - 1)th moment, and propagating the second cubature points through the measurement equation to obtain the second updated cubature points; based on the second updated cubature points, calculating the cross-correlation error covariance of the nonlinear model at the kth moment.

[0058] It should be noted that the specific steps for the nonlinear time update in the embodiments of the present application are as follows:

[0059] (1) Calculate the cubature points χ at the (k - 1)th moment i,k-1|k-1

[0060]

[0061]

[0062] where S k-1|k-1 is obtained through the Cholesky decomposition of P x,k-1|k-1

[0063] (2) Propagate the cubature points x using the state equation i,k|k-1

[0064]

[0065] (3) Calculate the state prediction value at the kth moment

[0066]

[0067] (4) Calculate the state error covariance prediction value P at the kth moment x,k|k-1

[0068]

[0069] Furthermore, the specific steps for the measurement update in the embodiments of the present application are as follows:

[0070] (1) Calculate the cubature points χ for measurement update i,k|k-1 ​

[0071]

[0072]

[0073] (2) Update the propagated volume points z using the measurement equation i,k|k-1

[0074]

[0075] (3) Calculate the predicted measurement value at time k

[0076]

[0077] (4) Calculate the autocorrelation error covariance P at time k z,k|k-1

[0078]

[0079] (5) Calculate the cross-correlation error covariance P at time k xz,k|k-1

[0080]

[0081] (6) Calculate the pseudo-measurement equation and the pseudo-measurement noise covariance

[0082] First, introduce the idea of linear regression, and transform the non-linear measurement function into a pseudo-linear measurement function in linear form through statistical linear model technology; based on the CKF algorithm, define the pseudo-measurement function as follows:

[0083]

[0084] The relationship between the measurement information and the linearized pseudo-measurement function can be expressed as:

[0085]

[0086] where, obeys distribution, and there is:

[0087]

[0088] Thus, the embodiments of the present application effectively ensure the numerical stability of the filtering algorithm during the process of minimizing the loss function by calculating the pseudo-measurement equation and the pseudo-measurement noise covariance.

[0089] In step S103, a mixed Agnesi criterion kernel function is determined, and fixed-point iteration operation is performed through preset fixed-point iteration conditions, a pseudo-measurement equation, and the mixed Agnesi criterion kernel function to obtain the Kalman gain of the nonlinear model at time k.

[0090] After constructing the pseudo-measurement equation of the nonlinear model, further, an embodiment of the present application can also combine the mixed Agnesi criterion kernel function with fixed-point iteration operation to obtain the Kalman gain of the nonlinear model at time k, as Figure 2 shown.

[0091] Thus, the embodiment of the present application designs a mixed Agnesi kernel loss function, and on the basis of the cubature Kalman algorithm, introduces the maximum Agnesi function to replace the MMSE criterion in the traditional CKF, and then constructs a new cost function by combining the weighted least squares method and the mixed Agnesi kernel loss function, and derives a numerically stable and more robust robust nonlinear state estimation algorithm; at the same time, the embodiment of the present application obtains the state estimate of the system by maximizing the sum of the mixed Agnesi functions of the prediction error and the residual, so as to realize the simultaneous processing of system non-Gaussian noise and measurement non-Gaussian noise.

[0092] Optionally, in an embodiment of the present application, fixed-point iteration operation is performed through preset fixed-point iteration conditions, a pseudo-measurement equation, and the mixed Agnesi criterion kernel function to obtain the Kalman gain of the nonlinear model at time k, including: determining the initial value of the preset fixed-point iteration condition, and based on the initial value, calculating the Kalman gain of the nonlinear model at the i-th fixed-point iteration at time k, where i is a positive integer.

[0093] It should be noted that the specific process of calculating the Kalman gain of the nonlinear model at the i-th fixed-point iteration at time k in the embodiment of the present application is as follows:

[0094] (1) Design a mixed Agnesi criterion kernel function

[0095]

[0096] where a1 and a2 are the radii of two different Agnesi functions respectively, and a1 < a2, e is the independent variable of the Agnesi kernel function, and 1 ≥ α ≥ 0 is the weight of the Agnesi function;

[0097] (2) Set the initial value of the fixed-point iteration condition and start fixed-point iteration

[0098]

[0099] (3) Calculate the Kalman gain at the i-th (i = 1, 2, K) fixed-point iteration at time k

[0100]

[0101] Among them

[0102]

[0103]

[0104]

[0105] Therefore, the embodiments of the present application consider the situation where both system noise and measurement noise have non-Gaussian noise distributions in actual engineering applications, construct a volume Kalman filtering algorithm with a hybrid Agnesi function and a hybrid Agnesi kernel function, and combine the CKF with the maximum hybrid Agnesi criterion, so that in the process of vehicle integrated positioning, the CKF algorithm can simultaneously suppress process non-Gaussian noise and measurement non-Gaussian noise interference, enabling the nonlinear system to more accurately estimate the system state, effectively improving the robustness of the state estimation algorithm in the complex case where the measurement noise has a multi-modal distribution, and strongly guaranteeing the realization of the accurate positioning of the target vehicle.

[0106] In step S104, the optimal state error covariance and the optimal state estimate of the nonlinear model are calculated according to the Kalman gain, so that the integrated navigation system estimates the positioning state of the target vehicle based on the optimal state error covariance and the optimal state estimate.

[0107] After obtaining the Kalman gain of the nonlinear model at time k, further, the embodiments of the present application can also calculate the optimal state error covariance and state estimate of the nonlinear model according to the Kalman gain to achieve accurate estimation of the positioning state of the target vehicle.

[0108] Optionally, in an embodiment of the present application, calculating the optimal state error covariance and the optimal state estimate of the nonlinear model according to the Kalman gain includes: calculating the state estimate and state error covariance of the nonlinear model at the i-th fixed-point iteration at time k based on the Kalman gain; judging whether the fixed-point iteration ends based on the fixed-point iteration condition, where if it ends, the state estimate and state error covariance at the end of the fixed-point iteration are used as the optimal state error covariance and the optimal state estimate of the nonlinear model.

[0109] It should be noted that the specific steps for calculating the optimal state error covariance and state estimate of the nonlinear model according to the Kalman gain in the embodiments of the present application are as follows:

[0110] (1) Calculate the state estimate at the i-th fixed-point iteration at time k

[0111]

[0112] And let

[0113] (2) Estimate the state error covariance P at the i-th fixed-point iteration at time k x i ,k|k

[0114]

[0115] (3) When the iteration ends, output the state estimate value and the state error covariance.

[0116] It should be noted that the embodiments of the present application can use the fixed-point iteration algorithm to obtain the optimal solution and the estimated error covariance P x,k|k , since is the only available information before the iterative measurement update, so in the fixed-point iteration algorithm, can be selected as the initial value, that is

[0117] It should be noted that in the processing process, the non-Gaussian part only includes information. At the first fixed-point iteration Therefore, the first fixed-point iteration has no inhibitory effect on the process non-Gaussian noise. Therefore, the embodiments of the present application need to consider more than one fixed-point iteration process.

[0118] In the embodiments of the present application, the conditions for the end of the iteration generally include the following two types:

[0119] 1. Use the accuracy as the cut-off condition;

[0120] 2. Limit the number of iterations.

[0121] However, the judgment process of the accuracy cut-off condition will increase the additional computational burden. Therefore, the embodiments of the present application adopt the method of setting a fixed number of iterations as the judgment condition for the end of the iteration.

[0122] In summary, the embodiments of the present application use the above method based on the maximum mixture Agnesi criterion to complete the suppression of the process non-Gaussian noise and the measurement non-Gaussian noise, that is, the maximum mixture Agnesi cubature Kalman filter algorithm (MMA-CKF).

[0123] Accordingly, the embodiments of the present application utilize the MMA-CKF algorithm to calculate the optimal state error covariance and state estimate of the nonlinear model through the Kalman gain, which can effectively reduce the computational complexity, thereby effectively improving the positioning accuracy in a complex environment where both the nonlinear system process and measurement data have non-Gaussian noise. In addition, the embodiments of the present application can be used in systems simultaneously affected by process non-Gaussian noise and measurement non-Gaussian noise, such as integrated navigation and target tracking systems, etc., and it is an algorithm with universality and scalability under the Kalman filter framework.

[0124] Optionally, in an embodiment of the present application, it further includes: when detecting that the type is a linear model type, establishing a linear state equation and a linear measurement equation of the integrated navigation system; performing a fixed-point iteration operation based on the linear state equation and the linear measurement equation to obtain the optimal state error covariance and the optimal state estimate of the linear model; estimating the positioning state of the target vehicle according to the optimal state error covariance and the optimal state estimate of the linear model.

[0125] In addition, when the measurement equation in the nonlinear system is a linear equation, the embodiments of the present application can establish a linear state equation and a linear measurement equation to perform a fixed-point iteration operation to obtain the optimal state error covariance and the optimal state estimate of the linear model, thereby realizing the estimation of the positioning state of the target vehicle.

[0126] Accordingly, in the case of a linear model, the embodiments of the present application do not need to execute the calculation steps of formulas (18) to (19), and can directly use the linear measurement equation and the measurement noise covariance in subsequent calculations, thereby effectively solving problems such as the coexistence of non-Gaussian noise in the process and measurement data of the nonlinear system, as well as complex non-ideal noise interference in the actual operation of the integrated navigation system, and considering both linear / nonlinear measurement equation cases, providing a flexible method for suppressing non-Gaussian noise interference for different nonlinear systems.

[0127] The following of the present application will introduce in detail the execution logic of the method for estimating the vehicle positioning state proposed in the present application through a specific target tracking embodiment and in combination with the drawings.

[0128] In a specific embodiment of the present application, the expression of its target tracking system is as follows:

[0129]

[0130] Wherein, (ξ k , η k ) respectively represent the position components in the x-axis and y-axis directions at time k, represents the velocity components in the x-axis and y-axis directions at time k, T = 1 is the sampling time, Ω is the turning rate of the target, wk is white Gaussian noise at time k with zero mean and covariance matrix Q; the measurement model of this model is:

[0131]

[0132] where r k is the measurement noise at time k; the relevant parameters of this simulation model are set as shown in Table 1.

[0133] Table 1

[0134]

[0135] In this specific embodiment, the root mean square error (RMSE, Root Mean Squre Error) and the average root mean square error (ARMSE, Average Root Mean Squre Error) of the estimation result can be used as performance indicators to evaluate the algorithm. The RMSE pos and ARMSE pos are defined as follows:

[0136]

[0137]

[0138] The RMSE vel and ARMSE vel of the speed information are defined as follows:

[0139]

[0140]

[0141] The simulation time step is K = 500, and the number of Monte Carlo runs is M = 200.

[0142] To better compare the algorithm performance, this specific embodiment can consider the following two cases:

[0143] Case 1: The process noise does not satisfy the Gaussian distribution characteristic and is simulated by the expression in Equation (35):

[0144] w k ~0.95N(0,Q)+0.05N(0,50Q) (35)

[0145] The measurement noise follows a complex multi-model distribution and is simulated by the form of Equation (36):

[0146] r k~0.6N(0,R)+0.2N([10m,-0.3rad] T ,100R)+0.2N([-10m,0.3rad] T ,100R)(36)

[0147] Case 2: The process noise does not satisfy the Gaussian distribution characteristics and is simulated in the form of formula (37):

[0148] w k ~0.95N(0,Q)+0.05N(0,50Q) (37)

[0149] The measurement noise follows a complex multi-model distribution with large outliers and is simulated using Equation (38):

[0150] r k ~0.48N([-10m,-0.3rad] T ,R)+0.48N([10m,0.3rad] T ,R)+0.04N(0,1000R)(38)

[0151] This simulation compares the existing CKF, the existing maximum entropy mixed Gaussian kernel-based CKF robust algorithm (MCC-CKFQR) (α=0.5, σ1=10, σ2=1), and the MMA-CKF1 (α=0.2, a1=10, a2=1), MMA-CKF2 (α=0.5, a1=10, a2=1) and MMA-CKF3 (α=0.7, a1=10, a2=1) with different parameter settings proposed in this application.

[0152] To ensure the fairness of the results, the number of fixed point iterations used in both the MCC-CKFQR and MMA-CKF algorithms is set to 5.

[0153] Figure 3 、 Figure 4 and Figure 5 The RMSE of position, speed and turning rate of different algorithms in the time period of 200s to 300s are shown respectively. Figure 3 、 Figure 4 and Figure 5 It can be seen that MMA-CKF and MCC-CKFQR perform significantly better than CKF under non-Gaussian noise background.

[0154] In order to more comprehensively compare the performance of the above algorithms, Table 2 shows the ARMSE of different algorithms.

[0155] Table 2

[0156]

[0157] As can be seen from Table 2, the MMA-CKF algorithm of this application performs best under different parameter allocations.

[0158] Figure 6 、 Figure 7 and Figure 8 The RMSE of position and velocity of the existing method and the method of the present application in the noise environment described in Case 2 are respectively shown in the time period of 200 seconds to 300 seconds.

[0159] Depend on Figure 6 、 Figure 7 and Figure 8 It can be seen that the gap between CKF and other algorithms is large, while the performance of other algorithms is relatively close to it, even in Figure 7 and Figure 8 The gap between the two algorithms cannot be seen intuitively, mainly because CKF is greatly affected by multi-model distribution noise accompanied by large outliers, while MCC-CKFQR, MCC-CKF and MMA-CKF have the ability to suppress process and measurement non-Gaussian noise.

[0160] Similarly, in order to comprehensively compare the performance of the above algorithms, Table 3 shows the ARMSE of different algorithms in case 2.

[0161] Table 3

[0162] Algorithm Position Speed Turning Rate MCC-CKFQR1 85.3097 5.6213 0.3373 MMA-CKF1 51.2856 1.8377 0.0469 MMA-CKF2 28.0667 1.7095 0.6014 MMA-CKF3 45.7383 1.8915 0.3334 MCK-CKF 71.1746 4.9811 0.3342 CKF 385.8054 347.526 2.2709

[0163] The following description of the present application will use another specific vehicle positioning embodiment and combine with the accompanying drawings to illustrate the vehicle positioning state estimation method of the present application.

[0164] This specific embodiment simulates the situation where both process noise and measurement noise are non-Gaussian in a vehicle-mounted INS / GNSS integrated navigation and positioning system, and compares the performance of the CKF, MCC-CKFQR, and MMA-CKF algorithms.

[0165] This specific embodiment uses the system equation of the loosely combined navigation system based on the direct method of process nonlinearity / measurement linearity to set the process noise to a mixed Gaussian distribution and the measurement noise to a bimodal multi-model distribution:

[0166] w k ~0.95N(0,Q)+0.05N(0,9Q) (39)

[0167] r 1,2 (k)~0.96N(10 -4 ,3 / R e )+0.04N(-10 -4 ,((3000 / R e ))2 ) (40)

[0168] r3(k) ~ 0.96N(10 -4 ,1 2 ) + 0.04N(-10 -4 ,1000 2 )

[0169] The initial filtering parameters of the algorithm are set as shown in Table 3.

[0170] Table 3

[0171] Algorithm Parameter MMA-CKF <![CDATA[α = 0.5, a1 = 15, a2 = 1, N = 5]]> MCC-CKFQR <![CDATA[α = 0.5, σ1 = 15, σ2 = 1, N = 5]]>

[0172] Table 3 intuitively shows the ARMSEs of different algorithms. As can be seen from Table 3, MMA-CKF always performs better than other algorithms with the change of α, and performs optimally when α = 0.5; MCK-CKF and MCC-CKFQR perform similarly. The main reason is that MCK-CKF is designed for large outlier noise, and the advantage of this algorithm will gradually become prominent when the probability of large outliers in the measurement increases. However, because it does not consider complex non-Gaussian noise problems, its performance is still inferior to MMA-CKF in complex environments.

[0173] Figure 9 and Figure 10 respectively show the RMSE of the attitude angle and position information obtained after 100 Monte Carlo simulations, and the ARMSE of the attitude angle and position information is shown in Table 4.

[0174] Table 4 ARMSE of different algorithms

[0175] Parameter CKF MCC-CKFQR MMA-CKF Roll Angle (deg) 1.6360 <![CDATA[1.1070×10 -3 > <![CDATA[1.1058×10 -3 > Pitch Angle (deg) 1.6336 <![CDATA[1.1071×10 -3 ]]> <![CDATA[1.1056×10 -3 > Yaw Angle (deg) 38.9978 <![CDATA[8.4501×10 -2 > <![CDATA[8.3956×10 -2 > Longitude (deg) 0.0012 <![CDATA[1.4556×10 -6 > <![CDATA[1.4528×10 -6 > Latitude (deg) 0.0012 <![CDATA[1.4819×10 -6 > <![CDATA[1.4692×10 -6 >

[0176] As can be seen from Table 4, the performance of MMA-CKF is significantly better than other algorithms; from Figure 11 、 Figure 12 and Figure 13 it can be seen that due to the non-Gaussian of the process and measurement affecting the estimation accuracy of CKF, MCC-CKFQR and MMA-CKF can effectively suppress non-Gaussian noise, and the performance of MMA-CKF of this application is better than the existing MCC-CKFQR.

[0177] The vehicle positioning state estimation method proposed according to the embodiments of the present application, when the positioning system is a non-linear model, establishes the state equation and measurement equation of the non-linear system to calculate the predicted value of the state error covariance and the cross-correlation error covariance at time k; based on a preset statistical linear model, uses the predicted value of the state error covariance and the cross-correlation error covariance to construct a pseudo-measurement equation; determines the mixed Agnesi criterion kernel function, and combines the preset fixed-point iteration condition and the pseudo-measurement equation to perform fixed-point iteration to obtain the optimal state error covariance and state estimation value at time k, so as to analyze the positioning state of the target vehicle. The present application constructs a mixed Agnesi kernel function and combines the CKF with the maximum mixed Agnesi criterion, so that the CKF algorithm can simultaneously suppress the interference of process non-Gaussian noise and measurement non-Gaussian noise during the vehicle integrated positioning process, greatly improving the positioning accuracy and stability.

[0178] Secondly, the vehicle positioning state estimation device proposed according to the embodiments of the present application will be described with reference to the accompanying drawings.

[0179] Figure 14 It is a block diagram of the vehicle positioning state estimation device according to the embodiments of the present application.

[0180] As Figure 14 shown, the vehicle positioning state estimation device 10 includes: a detection module 100, a first construction module 200, a first iteration module 300, and a first estimation module 400.

[0181] Among them, the detection module 100 is used to detect the type of the positioning system model of the constructed integrated navigation system, and establish the state equation and measurement equation of the integrated navigation system when the detected type is a non-linear model type.

[0182] The first construction module 200 is used to calculate the state error covariance and cross-correlation error covariance of the non-linear model at time k according to the state equation and measurement equation, and use the state error covariance and cross-correlation error covariance to construct a pseudo-measurement equation of the non-linear model, where k is a positive integer.

[0183] The first iteration module 300 is used to determine the mixed Agnesi criterion kernel function, and perform fixed-point iteration operations through the preset fixed-point iteration condition, the pseudo-measurement equation, and the mixed Agnesi criterion kernel function to obtain the Kalman gain of the non-linear model at time k.

[0184] The first estimation module 400 is used to calculate the optimal state error covariance and optimal state estimation value of the non-linear model according to the Kalman gain, so that the integrated navigation system estimates the positioning state of the target vehicle according to the optimal state error covariance and optimal state estimation value.

[0185] Optionally, in an embodiment of the present application, the estimation device 10 for the vehicle positioning state of the embodiment of the present application further includes: a first construction module, a second iteration module, and a second estimation module.

[0186] Among them, the first construction module is used to establish a linear state equation and a linear measurement equation of the integrated navigation system when a linear model type is detected.

[0187] The second iteration module is used to perform a fixed-point iteration operation based on the linear state equation and the linear measurement equation to obtain the optimal state error covariance and the optimal state estimate of the linear model.

[0188] The second estimation module is used to estimate the positioning state of the target vehicle according to the optimal state error covariance and the optimal state estimate of the linear model.

[0189] Optionally, in an embodiment of the present application, the first construction module 200 includes: a first calculation unit, a second calculation unit, a third calculation unit, and a fourth calculation unit.

[0190] Among them, the first calculation unit is used to calculate the first cubature point at time k-1 and propagate the first cubature point through the state equation to obtain the first updated cubature point.

[0191] The second calculation unit is used to calculate the state estimate and the state error covariance of the nonlinear model at time k based on the first updated cubature point.

[0192] The third calculation unit is used to calculate the second cubature point at time k-1 and propagate the second cubature point through the measurement equation to obtain the second updated cubature point.

[0193] The fourth calculation unit is used to calculate the cross-correlation error covariance of the nonlinear model at time k based on the second updated cubature point.

[0194] Optionally, in an embodiment of the present application, the first iteration module 300 includes: an operation unit, which is used to determine the initial value of the preset fixed-point iteration condition and calculate the Kalman gain of the nonlinear model at the i-th fixed-point iteration at time k based on the initial value, where i is a positive integer.

[0195] Optionally, in an embodiment of the present application, the first estimation module 400 includes: a fifth calculation unit and a judgment unit.

[0196] Among them, the fifth calculation unit is used to calculate the state estimate and the state error covariance of the nonlinear model at the i-th fixed-point iteration at time k based on the Kalman gain.

[0197] A judgment unit, configured to judge whether the fixed-point iteration ends based on the fixed-point iteration condition. If it ends, the state estimation value and state error covariance at the end of the fixed-point iteration are used as the optimal state error covariance and optimal state estimation value of the nonlinear model.

[0198] It should be noted that the foregoing explanations of the embodiments of the vehicle positioning state estimation method are also applicable to the vehicle positioning state estimation device of this embodiment, and will not be elaborated here.

[0199] According to the vehicle positioning state estimation device provided by the embodiments of the present application, when the positioning system is a nonlinear model, the state equation and measurement equation of the nonlinear system are established to calculate the predicted value of the state error covariance and the cross-correlation error covariance at time k. Based on the preset statistical linear model, a pseudo-measurement equation is constructed by using the predicted value of the state error covariance and the cross-correlation error covariance. The mixed Agnesi criterion kernel function is determined, and fixed-point iteration is performed in combination with the preset fixed-point iteration condition and the pseudo-measurement equation to obtain the optimal state error covariance and state estimation value at time k, so as to analyze the target vehicle positioning state. The present application constructs a mixed Agnesi kernel function and combines the CKF with the maximum mixed Agnesi criterion, so that the CKF algorithm can simultaneously suppress the interference of process non-Gaussian noise and measurement non-Gaussian noise during the vehicle integrated positioning process, greatly improving the positioning accuracy and stability.

[0200] Figure 15 The following is a schematic structural diagram of an electronic device provided by an embodiment of the present application. The electronic device may include:

[0201] A memory 1501, a processor 1502, and a computer program stored on the memory 1501 and executable on the processor 1502.

[0202] When the processor 1502 executes the program, it implements the vehicle positioning state estimation method provided in the above embodiments.

[0203] Further, the electronic device further includes:

[0204] A communication interface 1503, configured for communication between the memory 1501 and the processor 1502.

[0205] The memory 1501 is used to store a computer program executable on the processor 1502.

[0206] The memory 1501 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory.

[0207] If the memory 1501, the processor 1502, and the communication interface 1503 are implemented independently, the communication interface 1503, the memory 1501, and the processor 1502 can be interconnected through a bus and communicate with each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended Industry Standard Architecture (EISA) bus, or the like. The bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 15 only a thick line is used in Figure 15 , but it does not mean that there is only one bus or one type of bus.

[0208] Optionally, in a specific implementation, if the memory 1501, the processor 1502, and the communication interface 1503 are integrated on a chip, the memory 1501, the processor 1502, and the communication interface 1503 can communicate with each other through an internal interface.

[0209] The processor 1502 may be a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application.

[0210] The embodiments of the present application further provide a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned method for estimating the vehicle positioning state is implemented.

[0211] In the description of this specification, the descriptions with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples", etc. mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or N embodiments or examples. In addition, without conflict, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0212] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the technical features indicated. Thus, features defined with "first" and "second" may explicitly or implicitly include at least one such feature. In the description of the present application, the meaning of "N" is at least two, such as two, three, etc., unless otherwise specifically defined.

[0213] Any process or method description represented in a flowchart or described otherwise herein can be understood to represent a module, segment, or portion of code including one or N executable instructions for implementing a customized logical function or process. The scope of the preferred embodiments of the present application includes additional implementations, where functions may be executed in a substantially simultaneous manner or in a reverse order according to the functions involved, rather than in the order shown or discussed, which should be understood by those skilled in the art to which the embodiments of the present application pertain.

[0214] The logic and / or steps represented in a flowchart or described otherwise herein, for example, can be considered as a sequenced list of executable instructions for implementing a logical function and can be specifically implemented in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection portion with one or N wirings (electronic device), a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, as the program can be obtained electronically by optically scanning the paper or other media, followed by editing, interpretation, or other appropriate processing as necessary, and then stored in a computer memory.

[0215] It should be understood that various parts of the present application can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, the N steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits with logic gate circuits for implementing logic functions on data signals, application specific integrated circuits with appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0216] Those of ordinary skill in the art can understand that all or part of the steps carried by the method of the above embodiments can be completed by instructing relevant hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiments.

[0217] In addition, in each embodiment of the present application, each functional unit can be integrated in a processing module, can also exist physically separately for each unit, or two or more units can be integrated in a module. The above integrated module can be implemented in the form of hardware or in the form of a software functional module. When the above integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0218] The above-mentioned storage medium can be a read-only memory, a magnetic disk, an optical disk, etc. Although the embodiments of the present application have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present application. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A method for estimating the positioning state of a vehicle, characterized in that The following steps are involved: Detecting the type of the positioning system model of the constructed integrated navigation system, and establishing a state equation and a measurement equation of the integrated navigation system when detecting that the type is a nonlinear model type; Calculate the state equation and the measurement equation respectively k The state error covariance and the cross-correlation error covariance of the nonlinear model at the moment are obtained, and the pseudo measurement equation of the nonlinear model is constructed by using the state error covariance and the cross-correlation error covariance, wherein, k is a positive integer; Determine the mixed Agnesi criterion kernel function, and perform fixed-point iteration operations through preset fixed-point iteration conditions, the pseudo-measurement equation, and the mixed Agnesi criterion kernel function to obtain the k Kalman gain of the nonlinear model at the moment; Calculating an optimal state error covariance and an optimal state estimation value of the nonlinear model according to the Kalman gain, so that the integrated navigation system estimates the positioning state of the target vehicle according to the optimal state error covariance and the optimal state estimation value; Wherein, calculating respectively according to the state equation and the measurement equation k the state error covariance and the cross-correlation error covariance of the nonlinear model at the moment, including: calculate k a first volume point at time -1, and propagating the first volume point through the state equation to obtain a first updated volume point; Based on the first updated volume point, calculate the k State estimates and state error covariances of nonlinear models at each moment; Calculate the k second volume point at time -1, and propagate the second volume point through the measurement equation to obtain a second updated volume point; Based on the second updated volume point, calculate the k Cross-correlated error covariance of the moment-to-moment nonlinear model.

2. The method according to claim 1, wherein Also includes: In the case where it is detected that the model type is a linear model type, establishing a linear state equation and a linear measurement equation of the integrated navigation system; Performing a fixed point iteration operation based on the linear state equation and the linear measurement equation to obtain an optimal state error covariance and an optimal state estimate of the linear model; The positioning state of the target vehicle is estimated according to the optimal state error covariance and the optimal state estimation value of the linear model.

3. The method according to claim 1, characterized in that Performing a fixed-point iteration operation through the preset fixed-point iteration condition, the pseudo-measurement equation, and the hybrid Agnesi criterion kernel function to obtain the k Kalman gain of the nonlinear model at the moment, including: Determine the initial value of the preset fixed point iteration condition, and calculate the k Moment i The Kalman gain of the nonlinear model under the fixed point iteration, where i Is a positive integer.

4. The method according to claim 3, wherein Calculating the optimal state error covariance and the optimal state estimate of the nonlinear model according to the Kalman gain includes: Based on the Kalman gain, the k Moment i The state estimate and state error covariance of the nonlinear model under the fixed point iteration; Based on the fixed point iteration condition, determine whether the fixed point iteration is completed, wherein, if it is completed, the state estimation value and the state error covariance at the end of the fixed point iteration are used as the optimal state error covariance and optimal state estimation value of the nonlinear model.

5. An estimation device for a vehicle positioning state, characterized in that, include: a detection module, configured to detect the type of the positioning system model of the constructed integrated navigation system, and, if it is detected that the type is a nonlinear model type, establish a state equation and a measurement equation of the integrated navigation system; A first construction module, configured to calculate respectively k the state error covariance and the cross-correlation error covariance of the nonlinear model at a moment according to the state equation and the measurement equation, and construct a pseudo-measurement equation of the nonlinear model by using the state error covariance and the cross-correlation error covariance, where k is a positive integer; The first iteration module is used to determine the mixed Agnesi criterion kernel function, and perform fixed-point iteration operations through preset fixed-point iteration conditions, the pseudo-measurement equation, and the mixed Agnesi criterion kernel function to obtain the k Kalman gain of the time nonlinear model; a first estimation module, configured to calculate an optimal state error covariance and an optimal state estimate of the nonlinear model according to the Kalman gain, so that the integrated navigation system estimates a positioning state of the target vehicle according to the optimal state error covariance and the optimal state estimate; The first building block includes: The first computing unit is used to calculate k a first volume point at time -1, and propagating the first volume point through the state equation to obtain a first updated volume point; A second calculation unit, configured to calculate the k state estimation value and state error covariance of the time nonlinear model based on the first updated volume point; The third calculation unit is used to calculate the k a second volume point at time -1, and propagating the second volume point through the measurement equation to obtain a second updated volume point; A fourth calculation unit for calculating the k cross-correlation error covariance of the time nonlinear model based on the second updated volume point.

6. The device according to claim 5, characterized in that, Also includes: A first building module is configured to establish a linear state equation and a linear measurement equation of the integrated navigation system when detecting that the model type is a linear model type; A second iteration module is configured to perform a fixed point iteration operation based on the linear state equation and the linear measurement equation to obtain an optimal state error covariance and an optimal state estimate of the linear model; The second estimation module is used to estimate the positioning state of the target vehicle according to the optimal state error covariance and the optimal state estimation value of the linear model.

7. The device according to claim 5, characterized in that The first iteration module includes: An arithmetic unit is configured to determine an initial value of the preset fixed-point iteration condition and, based on the initial value, calculate the k Kalman gain of the nonlinear model at the i -th fixed-point iteration at the i -th moment, where i is a positive integer.

8. The device according to claim 7, characterized in that, The first estimation module includes: A fifth calculation unit, configured to calculate the state estimate value and the state error covariance of the nonlinear model under the k th i fixed-point iteration at the moment; A judgment unit is used to judge whether the fixed point iteration is completed based on the fixed point iteration condition, wherein, if it is completed, the state estimation value and the state error covariance at the end of the fixed point iteration are used as the optimal state error covariance and optimal state estimation value of the nonlinear model.

9. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the program to implement the method for estimating the vehicle positioning state according to any one of claims 1 to 4.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by a processor to implement the method for estimating the vehicle positioning state according to any one of claims 1-4.

Images