Adaptive Kalman filtering combined attitude determination method and system for star sensor and gyroscope

Abstract

The invention relates to the technical field of attitude determination of remote sensing satellites, and particularly provides an adaptive Kalman filtering combined attitude determination method and system of a star sensor and a gyro, the method comprises the following steps: constructing a discretization state equation of a spacecraft attitude determination system, the state vector of the system is composed of a quaternion vector part estimation error and a gyro drift error, the state transition matrix is dynamically updated according to the current angular velocity of the gyroscope; establishing a discretization measurement equation of the star sensor; based on the discretization state equation and the discretization measurement equation, adaptive Kalman filtering is adopted to realize optimal estimation of attitude parameters; in the adaptive Kalman filtering process, the measurement noise covariance matrix is dynamically adjusted according to the state estimation residual at the previous moment. According to the self-adaptive Kalman filtering combined attitude determination method for the star sensor and the gyroscope, in the filtering recursion process, the running state of the filter is monitored in real time or quasi-real time, and the statistical property of measured noise is adjusted online, so that the filter can be self-adaptive to a dynamically changing environment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of remote sensing satellite attitude determination, and particularly relates to a star sensor and gyro adaptive Kalman filter combined attitude determination method and system. BACKGROUND

[0002] In a high-precision spacecraft attitude control system, a star sensor and a gyroscope are often combined to achieve all-weather, high-dynamic, and high-precision attitude estimation. The star sensor has the advantages of high measurement accuracy and strong absolute pointing, but its output frequency is low and it is easily affected by light conditions, and there is a risk of data delay or loss; the gyroscope has high update rate and good short-term stability, but its measurement value contains drift error accumulated over time, and long-term use will cause attitude divergence. To fuse the advantages of the two, the existing technology generally uses a Kalman filter (KF) framework for information fusion, estimates the attitude quaternion error and gyro drift by establishing a system state equation and a measurement equation, thereby achieving high-precision attitude determination. This method has been widely used on various satellite platforms and has formed a relatively mature technical system.

[0003] However, the traditional Kalman filter method faces significant challenges in practical applications: one of its core assumptions is that the process noise covariance matrix and the measurement noise covariance matrix are known and time-invariant, but in a real space environment, the measurement noise of the star sensor will dynamically fluctuate due to factors such as severe attitude maneuver, changes in the distribution of stars in the field of view, and background light interference. At this time, if a fixed value is still used, it will lead to a mismatch in the filter gain. This uncertainty in the measurement model can lead to a decrease in the accuracy of Kalman filter estimation and even cause filter divergence, seriously threatening the safety of the spacecraft and the success of the mission. SUMMARY

[0004] To solve the above problems, the present application provides a star sensor and gyro adaptive Kalman filter combined attitude determination method and system. In the filter recursion process, the running state of the filter is monitored in real time or quasi-real time, and the statistical properties of the measurement noise are adjusted online, so that the filter can adapt to the dynamically changing environment.

[0005] To achieve the purpose of the present application, the present application provides the following technical solutions: In a first aspect, the present application provides a star sensor and gyro adaptive Kalman filter combined attitude determination method, comprising: A discrete state equation of a spacecraft attitude determination system is constructed based on a satellite attitude kinematics equation and a gyro measurement model; wherein the system state vector is composed of a quaternion vector part estimation error and a gyro drift error, and the state transition matrix is dynamically updated according to the current angular velocity of the gyro; A discrete measurement equation of the star sensor is established, and an observation is defined as a difference between a quaternion vector part of a measured attitude of the star sensor and a quaternion vector part of an estimated attitude; Based on the discrete state equation and the discrete measurement equation, optimal estimation of the attitude parameter is realized by using adaptive Kalman filtering, wherein in the adaptive Kalman filtering process, a measurement noise covariance matrix is dynamically adjusted according to a state estimation residual error of a previous time, and the measurement noise covariance matrix comprises: According to the state estimation of the previous frame, an attitude of a current frame is predicted to obtain a state prediction value; According to an estimation error covariance matrix of the previous frame, a prediction error covariance matrix is calculated; Based on the prediction error covariance matrix, a Kalman filtering gain is updated, and the updating specifically comprises: ; ; In the formula, K represents a Kalman gain matrix; represents the prediction error covariance matrix; represents a system measurement matrix; represents a transpose matrix of the system measurement matrix; represents a covariance matrix of measurement noise; is an adaptive rate, and is used for dynamically adjusting the covariance matrix of the measurement noise , the measurement noise covariance matrix is adjusted to adapt to different measurement characteristics of the star sensor; represents conversion of a 3*1 vector into a 3*3 diagonal matrix; represents a state estimation of a previous frame; represents an initial value of a covariance matrix of star sensor measurement noise; According to the measurement value of the star sensor, a system state estimation is updated.

[0006] Further improvements of the present application are that the discrete state equation of the spacecraft attitude determination system is constructed based on the satellite attitude kinematics equation and the gyro measurement model, and the discrete state equation comprises: ; In the formula, represents a discrete state transition matrix, represents a 6*6 unit matrix, represents a sampling period, , ; wherein, is a quaternion vector part estimation error, , is a real attitude quaternion vector part, represents quaternion multiplication; an estimation error of a quaternion vector part, ; a process noise vector, , represents gyro measurement noise, represents gyro drift noise.

[0007] The application further improves the measurement matrix of the discretized measurement equation is: ; the observation is defined as the quaternion product difference between the star sensor measured attitude quaternion vector part and the inverse of the current estimated attitude , specifically: ; the measurement equation of the spacecraft attitude determination system is: ; in the formula, represents a star sensor measurement noise sequence.

[0008] The application further improves the process noise vector and the star sensor measurement noise sequence are mutually independent Gaussian white noises, subject to the following statistical properties, specifically: ; ; ; wherein, represents an expectation operator, represents a covariance matrix of the process noise, represents a covariance matrix of the measurement noise.

[0009] The application further improves that when calculating the prediction error covariance matrix according to the estimation error covariance matrix of the previous frame, it is assumed that is time-invariant, specifically: ; in the formula, refers to the prediction error covariance matrix, refers to the estimation error covariance matrix of the previous frame, refers to the transpose matrix of the state transition matrix of the previous frame, .

[0010] The application further improves that the attitude prediction of the current frame is performed according to the state estimation of the previous frame to obtain a state prediction value, specifically: ; In the formula, is the state prediction value, is the state estimation of the previous frame, is the state transition matrix of the previous frame.

[0011] The further improvement of the application is that the method further comprises, before updating the system state estimation, using the gyro angular velocity after the correction of the drift error to perform attitude forward propagation compensation on the delayed star sensor measurement, to generate a star sensor measurement attitude for filtering and updating; at this time, specifically: .

[0012] The further improvement of the application is that the updating of the system state estimation according to the measurement value of the star sensor is specifically: ; In the formula, is the state prediction value; is the measurement value of the star sensor, .

[0013] The further improvement of the application is that after the state estimation is completed, the attitude correction amount and the drift increment are extracted from the updated state vector, and are respectively used for closed-loop correction of the attitude quaternion, the gyro zero offset and the angular velocity, and the specific process is: ; in the formula, is the attitude estimation error, which is obtained from the updated system state ; ; in the formula, is the attitude estimation error, which is obtained from the updated system state ; ; in the formula, is the three-axis rotation angular velocity of the gyro measurement.

[0014] In the second aspect, the application provides a star sensor and gyro adaptive Kalman filtering combined attitude determination system, which is used for implementing the star sensor and gyro adaptive Kalman filtering combined attitude determination method, and the system comprises: a sensor data acquisition unit, which is used for receiving the original output of the star sensor and the gyro; The posture calculation unit is integrated with a filtering processing module and is configured to execute an adaptive Kalman filtering combined posture determination method of the star sensor and the gyroscope, and the method comprises the following steps: based on a discretized state equation and a discretized measurement equation, optimal estimation of a posture parameter is realized by using adaptive Kalman filtering; wherein the discretized state equation is constructed based on a satellite posture kinematics equation and a gyroscope measurement model for spacecraft posture determination, wherein a system state vector is composed of a quaternion vector partial estimation error and a gyroscope drift error, and a state transition matrix is dynamically updated according to a current angular velocity of the gyroscope; in the discretized measurement equation, an observation quantity is defined as the difference between the quaternion vector part of the estimated posture and the quaternion vector part of the star sensor measured posture. In the adaptive Kalman filtering process, the measurement noise covariance matrix is dynamically adjusted according to the state estimation residual error of the last time, comprising: According to the state estimation of the last frame, the posture of the current frame is predicted to obtain a state prediction value; According to the estimation error covariance matrix of the last frame, a prediction error covariance matrix is calculated; Based on the prediction error covariance matrix, the Kalman filtering gain is updated, specifically: ; ; In the formula, K is a Kalman gain matrix; is the prediction error covariance matrix; represents a system measurement matrix; represents a transpose matrix of the system measurement matrix; represents a covariance matrix of the measurement noise; is an adaptive rate, used for dynamically adjusting the covariance matrix of the measurement noise , by adjusting to adapt to different star sensor measurement characteristics; represents converting a 3*1 vector into a 3*3 diagonal matrix; is the state estimation of the last frame; is the initial value of the covariance matrix of the star sensor measurement noise; According to the measurement value of the star sensor, the system state estimation is updated; An output interface is used to send the posture information to a satellite control system.

[0015] Compared with the prior art, the present application has the following beneficial effects: The adaptive Kalman filtering combined attitude determination method and system of the star sensor and the gyroscope provided in the application dynamically adjusts the covariance matrix of the measurement noise in the Kalman filtering gain updating process, improves the time-invariant setting of the covariance matrix of the system measurement noise, avoids the increase of the estimation result error, makes the combined attitude determination method suitable for the star sensor and the gyroscope more in line with the sensor characteristics in the actual engineering, and improves the precision of the system attitude estimation; compared with the existing adaptive Kalman filtering algorithm, the adaptive Kalman filtering combined attitude determination method of the star sensor and the gyroscope provided in the embodiment has small calculation amount, simple engineering implementation, and can meet the scene with high real-time requirement. BRIEF DESCRIPTION OF DRAWINGS

[0016] The accompanying drawings are included to provide a further understanding of the application, and constitute a part of the specification, illustrate the application together with the embodiments of the application, and are used to explain the application, and do not form a limitation on the application. Fig. 1 An optional flowchart of the adaptive Kalman filtering combined attitude determination method of the star sensor and the gyroscope provided in the embodiment of the application is shown in the figure. Fig. 2 An optional flowchart of the adaptive Kalman filtering combined attitude determination method of the star sensor and the gyroscope provided in the embodiment of the application is shown in the figure. Fig. 3 A comparison result diagram of several star sensor and gyroscope combined attitude determination methods provided in the embodiment of the application is shown in the figure. DETAILED DESCRIPTION

[0017] To make the purpose, technical scheme and advantages of the application more clear, the technical scheme in the application will be clearly and completely described below in combination with the drawings in the application. Obviously, the described embodiments are part of the embodiments of the application, rather than all the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor fall within the protection scope of the application.

[0018] The terms "first", "second" are only used for description purpose, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features limited by "first", "second" can explicitly or implicitly include one or more of the features; in the description of the application, unless otherwise specified, the meaning of "multiple" is two or more.

[0019] With the development of space missions towards high precision and long life, higher requirements are put forward for the precision and reliability of spacecraft attitude determination systems. Among many attitude sensors, star sensors have extremely high absolute attitude determination accuracy, but their output frequency is low, and they are prone to measurement abnormalities or even temporary failure under strong light, severe shaking and other disturbances. Gyroscopes can provide high-frequency angular velocity information and are the key to achieving high-bandwidth and high-real-time attitude determination. Therefore, the combined attitude determination system that combines the high precision of star sensors with the high dynamic characteristics of gyroscopes has become the mainstream solution for high-precision spacecraft attitude determination.

[0020] The Kalman filter algorithm is the core fusion technology in the combined attitude determination system. It optimally estimates the spacecraft's attitude and the gyro's error by constructing the system's dynamic model and measurement model, thereby outputting smooth, continuous and high-precision attitude information. However, the performance of the traditional Kalman filter algorithm largely depends on the prior accurate knowledge of the statistical characteristics of the sensor measurement noise. In actual operating environments, the measurement noise of star sensors may become large due to sudden changes in the spacecraft's attitude. If a fixed value is still used at this time, it will lead to a mismatch in the filter gain, and the filter result will oscillate or even diverge when the measurement quality of the star sensor is not properly reduced. Conversely, when the measurement quality is improved, the high-precision information is not fully utilized, limiting the overall performance. This uncertainty in the measurement model can lead to a decrease in the estimation accuracy of the Kalman filter and even cause filter divergence, seriously threatening the safety of the spacecraft and the success of the mission.

[0021] In existing Kalman filter technologies suitable for star sensor and gyro combined attitude determination, it is usually assumed that the covariance matrix of the combined attitude determination system measurement noise is time-invariant. Chinese Patent Publication No. CN107228672A discloses a star sensor and gyro data fusion method suitable for attitude maneuvering conditions, in which the least squares method is used to weight and fuse the data of the star sensor and the gyro. However, the measurement noise variance of the star sensor and the gyro is essentially considered as a fixed value, which can also lead to a large error in the estimation result. At the same time, the adaptive Kalman filter algorithm (such as the Sage-Husa algorithm) needs to perform a large number of calculations at each recursion to update the measurement noise covariance matrix, which can limit the real-time performance of the attitude estimation and is no longer suitable for some low-power chips or situations where the attitude update frequency is very high.

[0022] In addition, most existing methods do not effectively handle the star sensor data delay problem, and directly introducing lagged measurements can damage the stability of the filter.

[0023] To solve the above technical problems, the present application proposes the following technical solutions and corresponding embodiments.

[0024] The following will be described in conjunction with Figs. 1 to 3The embodiment shown describes the technical solutions of the application: Embodiment 1 The adaptive Kalman filtering combination attitude determination method of the star sensor and the gyroscope in the embodiment of the application refers to Fig. 1 As shown, the method comprises the following steps S101 to S104: Step S101, constructing a discretized state equation of a spacecraft attitude determination system based on a satellite attitude kinematics equation and a gyroscope measurement model; wherein a system state vector is composed of a quaternion vector part estimation error and a gyroscope drift error, and a state transition matrix is dynamically updated according to a current angular velocity of the gyroscope.

[0025] In the embodiment of the application, there are measurement noise and drift error in the gyroscope measurement, so that only an estimated attitude quaternion can be obtained when the star sensor is used to compensate the gyroscope drift, and the vector part of the estimated attitude quaternion is recorded as The embodiment defines the quaternion vector part estimation error as wherein, is a real attitude quaternion vector part, represents quaternion multiplication.

[0026] In the embodiment of the application, referring to Fig. 2 As shown, the discretized state equation of the combined attitude determination system (the star sensor and the gyroscope combined attitude determination system, i.e. the spacecraft attitude determination system) can be obtained according to the satellite attitude kinematics equation and the gyroscope measurement model, and the calculation process is as follows formula (1): Formula (1); wherein, represents a discretized state transition matrix, represents a 6*6 unit matrix, represents a sampling period, is a three-axis rotation angular velocity of the body coordinate system of the gyroscope measurement relative to the inertial coordinate system, recorded as: is a system state vector, taking the vector part of the quaternion estimation error and the gyroscope drift error as the system state vector, i.e. , represents the gyroscope measurement noise, represents the gyroscope drift noise.

[0027] Step S102, establishing a discretized measurement equation of the star sensor, and defining the observation as the difference between the quaternion vector part of the star sensor measured attitude and the estimated attitude.

[0028] ​​​In the embodiment of the present application, the observation of the combined attitude determination system is defined The difference between the output attitude quaternion of the star sensor and the estimated quaternion vector part is: The measurement equation of the system can be obtained as shown in the following formula (2): Formula (2); Wherein, represents the satellite attitude quaternion vector part measured by the star sensor in the k-th frame; represents the satellite attitude quaternion vector part measured by the star sensor in the k-th frame; is the measurement matrix of the system, ; represents the star sensor measurement noise sequence.

[0029] In the embodiment of the present application, in the state equation of formula (1) and the measurement equation of formula (2), and are mutually independent Gaussian white noises, which satisfy the following statistical properties as shown in the following formula: ; ; ; Wherein, represents the expectation operator, represents the covariance matrix of the process noise, represents the covariance matrix of the measurement noise.

[0030] Step S103, based on the discretized state equation and the discretized measurement equation, the optimal estimation of the attitude parameter is realized by using adaptive Kalman filtering; wherein, in the process of adaptive Kalman filtering, the measurement noise covariance matrix is dynamically adjusted according to the state estimation residual of the last time.

[0031] In the embodiment of the present application, based on the Kalman filtering algorithm and the preset adaptive rate, the optimal state estimation of the combined attitude determination system is performed.

[0032] In the embodiment of the present application, the content of the optimal state estimation of the combined attitude determination system includes the following steps S1031 to S1035: Step S1031, according to the state estimation of the last frame, the attitude prediction of the current frame is performed to obtain the state prediction value; Here, the state prediction value is obtained by extrapolating the prior state at k time from the state at k-1 time through the discretized state equation. As shown in the following formula (3), the state prediction value is calculated : Formula (3); In the formula, is the state estimation of the last frame, is the state transition matrix of the last frame.

[0033] Step S1032, calculating the prediction error covariance matrix according to the estimation error covariance matrix of the last frame; The prediction error covariance matrix is calculated as shown in the following formula (4) : Formula (4); In the formula, refers to the estimation error covariance matrix of the last frame, refers to the transpose matrix of the state transition matrix of the last frame , refers to the covariance matrix of the process noise. In the embodiment of the application, only relates to the measurement characteristics of the gyroscope, which is mainly affected by temperature, and temperature control measures are taken in most space projects, resulting in a small change in , so is assumed to be time-invariant, i.e.

[0034] Step S1033, updating the Kalman filter gain based on the prediction error covariance matrix obtained in step S1032; as shown in the following formula (5) and formula (6): Formula (5); Formula (6); In the formula, refers to the Kalman gain matrix; represents the system measurement matrix; refers to the transpose matrix of the system measurement matrix; represents the covariance matrix of the measurement noise; is an adaptive rate (proportionality coefficient) for dynamically adjusting the covariance matrix of the measurement noise , wherein can be adjusted to adapt to different measurement characteristics of the star sensor; represents converting a 3*1 vector into a 3*3 diagonal matrix, for realizing differentiated adjustment of the measurement reliability of different axes; is the initial value of the covariance matrix of the measurement noise of the star sensor.

[0035] In the embodiment of the application, dynamically adjusts ; in the classical Kalman filter algorithm, it is considered that is also time-invariant, but since the measurement noise of the star sensor can be affected by the dramatic change of the spacecraft attitude, the measurement noise can become large due to the dramatic change of the spacecraft attitude, which cannot be applied to the combined attitude determination of the spacecraft. By adjusting the adaptive rate The measurement noise covariance dynamic adjustment mechanism enables online evaluation of the star sensor measurement quality: when the residual of the previous state is large, it automatically increases... Decrease its weight, and conversely increase it to enhance trust, thus overcoming the limitations of traditional key-fact analysis. Gain mismatch caused by keeping it constant.

[0036] Therefore, in this embodiment, the statistical characteristics of measurement noise are no longer treated as fixed prior parameters during the filtering recursion process. Instead, the statistical characteristics of measurement noise are adjusted online by monitoring the filter's operating status in real time or near real time, enabling the filter to adapt to dynamically changing environments. This has significant theoretical and engineering value for achieving stable, reliable, and high-precision autonomous attitude determination of spacecraft throughout the entire mission cycle.

[0037] Step S1034: Update the system state estimate based on the star sensor measurement; as shown in equation (7): Equation (7); Since star sensor measurements are typically delayed, this embodiment does not directly use star sensor measurements. Instead, it first uses the gyroscope value after correcting for drift error to compensate for the delay of the star sensor, thus obtaining the star sensor's measured attitude. At this point, the measured value (observation) : Therefore, in the Kalman filter update process, using the optimally estimated velocity information to compensate for the delay of the star-sensor measurements can reduce the hysteresis error of attitude estimation and improve the accuracy of attitude information.

[0038] Step S1035: Update the estimated error covariance matrix.

[0039] Refer to the following formula (8): Equation (8); In the formula, It refers to the first The frame estimation error covariance matrix, It is the identity matrix. This refers to the Kalman gain matrix. Represents the system measurement matrix. It is the prediction error covariance matrix.

[0040] Thus, by introducing an adaptive rate mechanism based on state prediction information, the measurement noise covariance matrix is ​​dynamically adjusted. This enables online evaluation of the real-time measurement quality of star sensors and adaptive allocation of filter weights. Combined with a delay compensation strategy, it improves the stability and accuracy of the system in complex dynamic environments.

[0041] Step S104, correcting the measured attitude according to the system state estimated by the adaptive Kalman filter.

[0042] In the embodiments of the present application, after each update step is completed, the updated system state vector extracting the attitude quaternion and the gyro zero bias drift; wherein the content of correcting the measured attitude includes attitude quaternion correction, gyro drift error correction and attitude angular velocity correction; wherein the attitude quaternion correction is as follows formula (9): Formula (9); In the formula, is the attitude estimation error, which is obtained from the updated system state in formula (7) above; The gyro drift error correction is as follows formula (10): Formula (10); In the formula, is the attitude estimation error, which is obtained from the updated system state in formula (7) above; The attitude angular velocity correction is as follows formula (11): Formula (11); In the formula, is the three-axis rotation angular velocity measured by the gyro.

[0043] The adaptive Kalman filter combination attitude determination method of the star sensor and the gyro provided in the embodiments improves the time-invariant setting of the covariance matrix of the system measurement noise by dynamically adjusting the covariance matrix of the measurement noise in the Kalman filter gain update process, avoids the increase of the estimation result error, makes the combination attitude determination method suitable for the star sensor and the gyro more consistent with the sensor characteristics in actual engineering, improves the precision of the system attitude estimation, and can realize the provision of the spacecraft attitude information with real-time, high bandwidth and high precision; at the same time, compared with the existing adaptive Kalman filter algorithm, the adaptive Kalman filter combination attitude determination method of the star sensor and the gyro provided in the embodiments has small calculation amount, simple engineering implementation, and can meet the scene with high real-time requirement.

[0044] Embodiment 2 Based on the above embodiments, the present application further provides an adaptive Kalman filter combination attitude determination method of a star sensor and a gyro, which combines the measurement characteristics of the star sensor and the gyro to explain the implementation process of the present application, and assumes that the spacecraft attitude quaternion vector part has undergone a uniform speed-uniform acceleration-uniform speed maneuver process, and the initial value is . The embodiment takes a common star sensor as an example, the sampling frequency of which is 4 Hz, the time delay is 250 ms, and the influence of measurement noise is considered. The output of the star sensor is: ; wherein, represents the measurement noise, it is assumed that the standard deviation is linearly related to the speed, the greater the speed, the greater the noise.

[0045] Taking a certain type of fiber-optic gyroscope as an example, the sampling frequency is 200 Hz, the angular velocity random walk coefficient is , the angle random walk coefficient is , the scale factor nonlinearity is 50ppm, the zero bias repeatability is . The speed measured by the fiber-optic gyroscope is: ; ; wherein, represents the gyroscope drift, and represent the measurement noise, which is related to the random walk coefficient.

[0046] The adaptive Kalman filtering combination attitude determination method of the star sensor and the gyroscope of the embodiment includes the following steps S201 to S202, which includes the following contents: Step S201, according to the measured angular velocity of the gyroscope, a state transition matrix is established; using the star sensor to correct the gyroscope integration to build an initial estimate of the attitude quaternion ; according to the measured quaternion of the star sensor, an observation quantity is established ; setting the process noise covariance matrix , the measurement noise covariance matrix initial value , the state estimate initial value and the error covariance matrix initial value ; Step S202, based on the Kalman filtering algorithm and the preset adaptive rate, continuously estimating the attitude; Specifically, step S202 includes the following contents of steps S2021 to S2029: Step S2021, according to the state estimate of the last frame, the attitude of the current frame is predicted to obtain a state prediction value; ; Step S2022, according to the estimation error covariance matrix of the last frame, a prediction error covariance matrix is calculated; ; Step S2023, updating the Kalman filtering gain; ; ; wherein, is a proportional coefficient, which can be adjusted to adapt to different star sensor measurement characteristics; it should be noted that the adaptive Kalman filter combination attitude determination method of the star sensor and the gyroscope in the application can be adapted to star sensors and gyroscopes with different characteristics by designing different adaptive rates.

[0047] Step S2024, compensating for the time delay of the star sensor using the corrected gyroscope value, to obtain the star sensor measured attitude , wherein the observation ; Step S2025, updating the system state estimation according to the measured value of the star sensor: ; Step S2026, updating the estimation error covariance matrix: ; Step S2027, correcting the attitude quaternion to: ; Step S2028, correcting the gyroscope drift error to: ; Step S2029, correcting the attitude angular velocity to: .

[0048] The simulation step length is set to 0.005s in the matlab in the embodiment, and the comparative results of the star sensor corrected gyroscope integrated attitude, the classical Kalman filter estimated attitude and the adaptive Kalman filter estimated attitude of the embodiment are as shown in the attached Fig. 3 It can be seen that the precision of the attitude information obtained by the method provided in the application is obviously improved.

[0049] Therefore, any improvement and application of the method by adjusting the adaptive rate parameter to change the method all belong to the protection scope of the patent.

[0050] The application embodiment further provides a computer readable storage medium, and the computer readable storage medium stores a computer program. The computer program is executed by a processor to implement the method in any embodiment of the application. Specifically, a system or device provided with a storage medium can be provided. The storage medium stores software program code for implementing the functions of any of the above embodiments, and the computer (or CPU or MPU) of the system or device reads and executes the program code stored in the storage medium.

[0051] In particular, the processes described above with reference to the flowcharts can be implemented as a computer software program in accordance with embodiments of the present application. For example, embodiments of the present application include a computer program product comprising a computer program carried on a computer readable storage medium, the computer program comprising program code for performing the methods illustrated by the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication section, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), the above-described functions defined in the system of the present application are performed.

[0052] It should be noted that the computer readable storage medium shown in the present application can be a computer readable signal medium or a computer readable storage medium or any combination of the two. The computer readable storage medium may, for example, be an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or apparatus, or any combination of the above. More specific examples of the computer readable storage medium can include, but are not limited to, an electrical connection having one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device or apparatus. In the present application, the computer readable signal medium can include a data signal carried in a baseband or as a carrier wave in a propagated data signal, which carries computer readable program code. Such a propagated data signal can take many forms, including but not limited to, an electromagnetic signal, an optical signal, or any suitable combination of the above. The computer readable signal medium can also be any computer readable storage medium that can send, propagate or transfer a program for use by or in connection with an instruction execution system, device or apparatus. The program code contained on the computer readable storage medium can be transmitted using any suitable medium, including but not limited to wireless, wire line, optical cable, RF, or any suitable combination of the above.

[0053] The computer program product of the present application can be a computer program embodied on a storage medium. Such storage medium can be a floppy disk, a hard drive, a RAM, a ROM, a memory stick, a memory card, a DVD, a CD, a flash drive or the like. The computer program product can also be a downloadable computer program, i.e. a computer program which can be downloaded, transferred or streamed, over the Internet or another network, either from a server or from another computer which has the computer program. The computer program can be downloaded, transferred or streamed at the request of a user, or it can be downloaded, transferred or streamed automatically, without the intervention of a user, e.g. in response to an electronic mail containing the computer program, a query, or the like. The computer program can be downloaded, transferred or streamed, e.g. from a server, or from another computer, which is connected to the Internet or another network, or it can be downloaded, transferred or streamed from a storage medium, e.g. a DVD, a CD, a flash drive, a memory stick, a memory card, a ROM, a RAM, a floppy disk, or the like, which is connected to the Internet or another network. The computer program can be downloaded, transferred or streamed from a server, or from another computer, or from a storage medium, as part of a service, e.g. on demand or in response to a request.

[0054] The units described in the embodiments of the present application can be implemented by software, or by hardware, or by a combination of software and hardware. The units described can be arranged in a processor. In some cases, the names of the units do not constitute a limitation on the units themselves. The adaptive Kalman filter combination attitude determination method of the star sensor and the gyroscope according to the present application can be implemented on an embedded hardware, a chip, a PLC, a computer, and the like, and thus the implementation of the present application on any hardware platform is within the protection scope of the present patent.

[0055] It should be noted that although several modules or units of the device for action execution are mentioned in the foregoing detailed description, such a division is not mandatory. Indeed, according to an embodiment of the application, the features and functionalities of two or more modules or units described above can be embodied in one module or unit. Conversely, the features and functionalities of one module or unit described above can be further divided into several modules or units.

[0056] In several embodiments provided in the present application, it should be understood that the disclosed system, modules and methods can be implemented in other manners. For example, the above described embodiments of the modules are merely schematic. For example, the division of the units is merely a logical function division. There can be another division manner for the actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between the units can be indirect couplings or communication connections through some interfaces, and can be in electrical, mechanical or other forms.

[0057] The above examples are only used to illustrate the technical solutions of the present application, but not to limit them. The present application is not limited to the exact structure as has been described above and shown in the drawings, and the specific implementation of the present application should not be considered as limited to these descriptions. Any changes and modifications made by those skilled in the art without departing from the concept of the present application should be considered as falling within the scope of the present application.

Claims

1. An adaptive Kalman filter combination of star sensor and gyroscope for attitude determination, characterized in that, The method comprises the following steps: a discrete state equation of a spacecraft attitude determination system is constructed based on a satellite attitude kinematics equation and a gyro measurement model; wherein a system state vector is composed of a quaternion vector part estimation error and a gyro drift error, and a state transition matrix is dynamically updated according to a current angular velocity of the gyro; a discrete measurement equation of a star sensor is established, and an observation quantity is defined as a difference between a measured attitude of the star sensor and a quaternion vector part of an estimated attitude; optimal estimation of an attitude parameter is realized by using adaptive Kalman filtering based on the discrete state equation and the discrete measurement equation; wherein in the adaptive Kalman filtering process, a measurement noise covariance matrix is dynamically adjusted according to a state estimation residual error of a previous time, comprising: an attitude of a current frame is predicted according to a state estimation of a previous frame, and a state prediction value is obtained; a prediction error covariance matrix is calculated according to an estimation error covariance matrix of the previous frame; the Kalman filtering gain is updated based on the prediction error covariance matrix, specifically as follows: ； ； wherein K denotes the Kalman gain matrix; P denotes the prediction error covariance matrix; Y denotes the system measurement matrix; Y denotes the transpose of the system measurement matrix; R denotes the covariance matrix of the measurement noise; is an adaptive rate for dynamically adjusting the covariance matrix of the measurement noise by adjusting to adapt to different star sensor measurement characteristics; denotes the conversion of a 3*1 vector to a 3*3 diagonal matrix; X denotes the state estimation of the previous frame; is the initial value of the covariance matrix of the star sensor measurement noise; the system state estimation is updated according to a measurement value of the star sensor.

2. The star sensor and gyro adaptive Kalman filter combined attitude determination method according to claim 1, characterized in that, The discrete state equation of the spacecraft attitude determination system constructed based on the satellite attitude kinematics equation and the gyro measurement model is specifically as follows: ； wherein denotes the discretized state transition matrix, denotes a 6*6 identity matrix, denotes the sampling period, , ; wherein, is the quaternion vector part estimation error, , is the true attitude quaternion vector part, denotes a quaternion multiplication; is the quaternion vector part estimation error, ; is the process noise vector, , denotes the gyro measurement noise, denotes the gyro drift noise.

3. The star sensor and gyro adaptive Kalman filter combined attitude determination method according to claim 2, characterized in that, The measurement matrix of the discretized measurement equation is: ; the observation is defined as the quaternion product difference between the star sensor measured attitude quaternion vector part and the inverse of the current estimated attitude , in particular: ； the measurement equation of the spacecraft attitude determination system is obtained as follows: ； In the formula, represents the star sensor measurement noise sequence.

4. The star sensor and gyro adaptive Kalman filter combined attitude determination method according to claim 3, characterized in that, the process noise vector and the star sensor measurement noise sequence are mutually independent Gaussian white noises, subject to the following statistical properties, in particular: ； ； ； wherein represents a desired operator, represents a covariance matrix of process noise, represents a covariance matrix of measurement noise.

5. The star sensor and gyro adaptive Kalman filter combined attitude determination method according to claim 4, characterized in that, In calculating the prediction error covariance matrix from the estimated error covariance matrix of the previous frame, it is assumed that is time-invariant, in particular: ； wherein denotes the prediction error covariance matrix, denotes the estimation error covariance matrix of the previous frame, denotes the state transition matrix of the previous frame the transpose of the state transition matrix, .

6. The star sensor and gyro adaptive Kalman filter combined attitude determination method according to claim 5, characterized in that, the attitude of the current frame is predicted according to the state estimation of the previous frame, and the state prediction value is obtained, specifically as follows: ； wherein is the state prediction value for the state, is the state estimate of the previous frame, is the state transition matrix of the previous frame.

7. The star sensor and gyro adaptive Kalman filter combined attitude determination method according to claim 6, characterized in that, The method further comprises, before updating the system state estimation, using the gyro angular velocity after the correction of the drift error to compensate for the delayed star sensor measurement by attitude forward propagation, to generate a star sensor measurement attitude for filtering and updating; at this time, specifically: 。 8. The star sensor and gyro adaptive Kalman filter combined attitude determination method according to claim 7, characterized in that, the system state estimation is updated according to the measurement value of the star sensor, specifically as follows: ； wherein is the state prediction value; is the measurement value of the star sensor, .

9. The star sensor and gyro adaptive Kalman filter combined attitude determination method according to any one of claims 1-8, characterized in that, after the state estimation is completed, an attitude correction amount and a drift increment are extracted from the updated state vector, and are respectively used for closed-loop correction of an attitude quaternion, a gyro zero offset and an angular velocity, specifically as follows: ； wherein is the attitude estimation error, which is derived from the updated system state ; ； wherein is the pose estimation error, obtained from the updated system state ; ； In the formula, is the three-axis rotation angular velocity measured by the gyroscope.

10. An adaptive Kalman filter combined star sensor and gyro attitude determination system for implementing the adaptive Kalman filter combined star sensor and gyro attitude determination method of any one of claims 1-9, wherein, The system comprises: a sensor data acquisition unit configured to receive original outputs of a star sensor and a gyro; an attitude calculation unit integrated with a filtering processing module and configured to execute an adaptive Kalman filtering combined attitude determination method of the star sensor and the gyro, the method comprising: realizing optimal estimation of an attitude parameter by using adaptive Kalman filtering based on a discrete state equation and a discrete measurement equation; wherein the discrete state equation is constructed based on a satellite attitude kinematics equation and a gyro measurement model, wherein a system state vector is composed of a quaternion vector part estimation error and a gyro drift error, and a state transition matrix is dynamically updated according to a current angular velocity of the gyro; in the discrete measurement equation, an observation quantity is defined as a difference between a measured attitude of the star sensor and a quaternion vector part of an estimated attitude; in the adaptive Kalman filtering process, a measurement noise covariance matrix is dynamically adjusted according to a state estimation residual error of a previous time, comprising: an attitude of a current frame is predicted according to a state estimation of a previous frame, and a state prediction value is obtained; a prediction error covariance matrix is calculated according to an estimation error covariance matrix of the previous frame; the Kalman filtering gain is updated based on the prediction error covariance matrix, specifically as follows: ； ； wherein denotes the Kalman gain matrix; denotes the prediction error covariance matrix; denotes the system measurement matrix; denotes the transpose of the system measurement matrix; denotes the covariance matrix of the measurement noise; is an adaptation rate for dynamically adjusting the covariance matrix of the measurement noise by adapting to different star sensor measurement characteristics; denotes the conversion of a 3*1 vector to a 3*3 diagonal matrix; denotes the state estimation of the previous frame; is the initial value of the covariance matrix of the star sensor measurement noise; the system state estimation is updated according to a measurement value of the star sensor; an output interface configured to send attitude information to a satellite control system.

Images