Regional navigation positioning method, system and equipment based on sequential information and medium

Abstract

The invention provides a regional navigation positioning method, system and device based on sequential information, and a medium, which can effectively improve the positioning precision and reduce the calculation complexity when processing the sequential information, and have better precision and robustness. Under a satellite navigation denial condition, aiming at distance measurement information at different moments obtained by land-based radio area navigation, a factor graph model is established to represent an information transmission process between different moment states of a positioning target. Compared with a traditional Kalman filtering framework, the factor graph algorithm is better in positioning precision, higher in flexibility, lower in calculation complexity and better in robustness when sequential information is processed. According to the method, state estimation is converted into a nonlinear least square optimization problem, a Gaussian Newton method is adopted for iterative solution, and compared with extended Kalman filtering, the matrix operand is reduced by more than 40%.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of navigation high-precision positioning technology, and particularly relates to a regional navigation positioning method, system, device and medium based on sequential information. BACKGROUND

[0002] Navigation positioning has been an important part of civil and military fields since ancient times, and plays an irreplaceable role in activities such as rescue search and precision strike. Current high dynamic high precision navigation positioning technology greatly depends on satellite navigation systems. However, due to the limitations of space environment and electromagnetic environment, when subjected to space occlusion and electromagnetic interference, the performance of satellite navigation systems will decrease significantly or even be unavailable. In the case of satellite navigation denial, in order to ensure the reliable operation of the navigation positioning system, as an effective supplement to satellite navigation, land-based radio regional navigation positioning technology has become an important research direction, and various navigation high-precision positioning technologies based on non-simultaneous ranging information have developed rapidly.

[0003] In the mobile single-user positioning scenario of regional navigation, land-based radio positioning technology is used to realize the communication transmission between the user and the ground by using multiple fixed base stations. The user sends signals to each base station at the same time at a certain moment, and receives the return signals of each base station at different times. By using the observation information at different times for positioning calculation, a high-precision positioning result is obtained. However, the traditional Kalman filtering method needs to marginalize the past state quantity in data processing, and only uses the recent state information for estimation, so that the non-linear coefficients representing the measurement information of these navigation sensors cannot be well linearized during data calculation, resulting in discontinuity of the process probability. SUMMARY

[0004] Therefore, the present application provides a regional navigation positioning method, system, device and medium based on sequential information, which can effectively improve the positioning accuracy and reduce the computational complexity when processing sequential information, and has good accuracy and robustness.

[0005] In order to solve the above problems, the technical scheme of the present application is as follows:

[0006] A regional navigation positioning method based on sequential information, comprising the following steps:

[0007] Step one, using land-based radio to realize regional navigation ranging: at each observation epoch, the mobile target receives the response signals of N ground stations with known positions, calculates the signal propagation delay, and obtains N sequential observation information containing distance, distance rate, speed, speed rate, measurement distance error and measurement speed error;

[0008] Step two, dividing the sequential information time slot: a single processing time slot contains N time points, and it is ensured that the mobile target receives the response signals of all ground stations in sequence within the time slot;

[0009] Step three, constructing factor graph model: based on the sequential observation information of the current time slot, a model containing the variable node of the state to be estimated of the maneuvering target, and the prior information factor node, the state transition factor node and the observation information factor node is established;

[0010] Step four, constructing global target function: according to the factor graph model, the state transition cost function and the observation information cost function are defined, and the state estimation is converted into a joint optimization problem in the form of maximum a posteriori estimation;

[0011] Step five, solving the target function: the Gauss-Newton method is used to iteratively optimize the global target function, and the optimal state variable that minimizes the function is obtained as the positioning result of the current time slot;

[0012] Step six, global positioning update: steps three to five are repeated in the next time slot to solve the next epoch state based on the current epoch observation information, and the global positioning result is iteratively output.

[0013] In the step two, the time slot division needs to meet: let the total number of time points T in the time slot and the number of ground stations N meet the mapping relationship T=kN (k≥1), k is the kth time point, and the response signal of each ground station is processed only once in the time slot.

[0014] In the step three, the specific steps for constructing the factor graph model according to the sequential observation information are:

[0015] Step 31, setting the variable node, which should contain the coordinates of the state to be estimated of the maneuvering target;

[0016] Step 32, setting the state transition factor node, which uses the velocity measurement model to constrain the transition process of the adjacent time points of the state to be estimated of the maneuvering target, and the velocity measurement error is regarded as a Gaussian random distribution;

[0017] Step 33, setting the observation information factor node, regarding the ranging error as a Gaussian random distribution, and establishing the covariance matrix by the elevation angle model and the signal-to-noise ratio.

[0018] In the step 2, the covariance matrix is adjusted by the nominal velocity measurement accuracy of the receiver or multiple tests.

[0019] In the step four, the specific steps for constructing the global target function according to the factor graph model are:

[0020] Step 41, designing the state transition cost function, when the velocity measurement error satisfies the Gaussian distribution, the velocity measurement error function, i.e. the state transition cost function, is obtained;

[0021] Step 2, design observation information cost function, when the ranging error satisfies the Gaussian distribution, the expression of the ranging error function, that is, the observation information cost function at a given time at a given base station is obtained.

[0022] Step 3, design global target function, when the error satisfies the Gaussian distribution, the negative logarithm of the probability distribution is proportional to the error function, and the maximum posterior estimation can be converted into a nonlinear least square problem, combined with the global target function described in step three, the least square solution of the factor graph model of the above node and edge is obtained.

[0023] In step five, the specific steps of solving the global target function by using the Gauss-Newton method are as follows:

[0024] Step 51, for the nonlinear least square form function, the error term is approximated by first-order Taylor expansion, which is changed into a linear least square form function;

[0025] Step 52, in the process of continuous iteration, the regression coefficient is optimized and adjusted for many times, so that the difference between the regression coefficient and the optimal regression coefficient is continuously reduced, and finally the smallest residual sum of squares is obtained by the algorithm, the solution of the global target function is realized, and the state to be estimated of the maneuvering target in the current time slot is obtained.

[0026] The application also provides a sequential information regional navigation positioning system, comprising:

[0027] State node factor unit: configured to set a variable node containing a state coordinate to be estimated according to a maneuvering target motion scene;

[0028] State transition factor unit: configured to construct a state transition model by using velocity information to constrain the state relationship of adjacent time;

[0029] Observation information factor unit: configured to fuse prior information and sequential ranging data to construct an observation model;

[0030] Global target function unit: configured to combine three types of nodes to generate a global target function in the form of nonlinear least square;

[0031] Factor graph solving unit: configured to solve the target function by using the Gauss-Newton method, and output the positioning result;

[0032] The state transition factor unit executes the state transition factor node construction logic, and the observation information factor unit executes the observation information factor node construction logic.

[0033] The application also provides an electronic device, characterized by comprising:

[0034] A processor 61;

[0035] A memory 62 stores a computer program;

[0036] A transmission device 63 is configured to communicate with an external base station.

[0037] When the processor executes the computer program, the positioning method of the application is implemented.

[0038] The transmission device 63 is configured to establish a communication link with N ground stations through a ground-based radio in a satellite denial environment.

[0039] The application also provides a computer readable storage medium, which stores computer instructions, and the instructions are executed by a processor to implement the steps of the method of the application.

[0040] Advantages:

[0041] 1. In the method of the application, the ranging information obtained by the land-based radio regional navigation at different times is used to establish a factor graph model to represent the information transmission process between the states of the positioning target at different times. The variable node represents the state to be estimated of the positioning target, the factor node represents the prior information, state transition and measurement process of the positioning target, the state estimation of the positioning target is converted into a maximum a posteriori estimation problem of a global cost function in the form of a joint probability density function based on the Bayes formula, a target cost function in the form of a non-least square is designed, the optimal state variable that minimizes the global target function is found, and the state to be estimated of the positioning target is obtained by solving. Compared with the traditional Kalman filter framework, the factor graph algorithm has better positioning accuracy, higher flexibility and lower computational complexity when processing sequential information, and has good robustness.

[0042] 2. In the method of the application, the state estimation is converted into a nonlinear least square optimization problem, and the Gauss-Newton method is used for iterative solution, which reduces the matrix operation amount by more than 40% compared with the extended Kalman filter. Compared with the Kalman filter and the corresponding improved Kalman filter method, the high-precision regional navigation positioning method based on sequential information in the application has high flexibility, low computational complexity, better accuracy and robustness when processing sequential information and non-periodic data.

[0043] 3. In the application, the joint optimization model is constructed by fusing the sequential ranging information (distance / speed / error) of multiple base stations, and the experimental verification shows that the positioning error radius is reduced by 62% compared with the traditional method.

[0044] 4. In the application, the observation data is adaptively integrated by dynamically dividing time slots, the asynchronous arrival of base station response signals is supported, the positioning availability is still maintained at 90% under electromagnetic interference, and the robustness of non-periodic data processing is enhanced.

[0045] 5、In the application, a double cost function model is constructed: the state transition cost function restricts the speed error and the observation cost function restricts the ranging error, and the cumulative error is reduced by 53% through dynamic adjustment of the covariance matrix, realizing the collaborative suppression of multi-source errors.

[0046] 6、The positioning system hardware architecture, medium storage and solving unit (factor graph solving unit) are designed in the application, so that the method can be executed in real time in an embedded device, the processing delay is less than 10 ms, and a full-link technical implementation scheme is provided. BRIEF DESCRIPTION OF DRAWINGS

[0047] Figure 1 The figure is a general execution flowchart of the method of the application.

[0048] Figure 2 The figure is a factor graph algorithm model diagram of the method of the application.

[0049] Figure 3 The figure is a positioning solving effect diagram of unordered sequential information least square positioning in the verification experiment of the method of the application.

[0050] Figure 4 The figure is a positioning solving effect diagram of sequential information traditional extended Kalman filter positioning in the verification experiment of the method of the application.

[0051] Figure 5 The figure is a positioning solving effect diagram of sequential information factor graph algorithm positioning used in the application.

[0052] Figure 6 The figure is a structural schematic diagram of an electronic device provided by the embodiment of the application. DETAILED DESCRIPTION

[0053] The application will be described in detail below with reference to the drawings and embodiments.

[0054] The application adopts the factor graph algorithm, which is a kind of probabilistic graph model and can intuitively represent the information transmission relationship between system states, and all historical observation information and system state information are used to constrain the estimation of system state. The application adopts high-precision regional navigation positioning based on sequential information for the case of satellite navigation denial. The application provides a regional navigation positioning method based on sequential information, and the general execution flowchart is as shown in Figure 1 The method comprises the following steps:

[0055] Step one, under the condition of satellite navigation denial, regional navigation ranging is realized by using land-based radio. At each observation epoch t k , the mobile target receives the response signals of N ground stations with known positions, calculates the propagation delay of the signals, and obtains N sequential observation information, wherein the observation information includes distance, distance rate, speed, speed rate, measurement distance error and measurement speed error;

[0056] Step two, dividing the sequential information time slot. Set a sequential information processing time slot, ensure that in each time slot, the mobile target receives all ground station response signals in turn;

[0057] Step three, in the current time slot, constructing a factor graph model according to the sequential observation information. The model includes variable nodes of the mobile target to be estimated state and factor nodes of prior information, state transition and observation information;

[0058] Step four, combining the factor graph model of step three to construct a global target function, including state transition cost function and observation information cost function, converting the state estimation of the mobile target into the joint optimization of the global target function in the form of maximum posterior estimation;

[0059] Step five, solving the global target function by using Gauss-Newton method to obtain the optimal state variable that minimizes the global target function, that is, the state to be estimated of the mobile target in the current time slot;

[0060] Step six, in the next time slot, repeating steps three to five, solving the state to be estimated at the next epoch t k k+1 Updating iteration to obtain the global positioning result of the mobile target;

[0061] Specifically, the factor graph algorithm model constructed by the method of the present application is as shown in Figure 2 In step three, the specific steps of constructing a factor graph model according to the sequential observation information are as follows:

[0062] Step 31, setting a variable node, which should contain the coordinates of the state to be estimated of the mobile target, and the specific formula is as follows:

[0063] Q=[x0,x1,x2,…,x N ]

[0064] x N ∈R 3 =(X N ,Y N ,Z N ) T

[0065] Wherein, Q represents the state set of each time when the mobile target receives the response signal of each ground station;

[0066] Step 32, setting a state transition factor node, using a velocity measurement model to constrain the transition process of the state to be estimated of the mobile target between adjacent time, regarding the velocity measurement error as a Gaussian random distribution, and the state transition factor between t time and t+1 time is simplified as:

[0067] ​

[0068] h v (x t ,x t+1 )=(x t+1 -x t ) / Δt

[0069] Wherein, the covariance matrix can be adjusted by the nominal velocity measurement accuracy of the receiver or multiple tests;

[0070] Step 33, set the observation information factor node, and simplify the pseudorange observation factor of the ground station j at time t by regarding the ranging error as a Gaussian random distribution:

[0071]

[0072] Wherein, the covariance matrix is established by the elevation angle model, signal-to-noise ratio and the like;

[0073] In the step four, the specific steps for constructing the global target function according to the factor graph model are:

[0074] Step 41, design the state transition cost function, and when the velocity measurement error satisfies the Gaussian distribution, the velocity measurement error function is the state transition cost function At time t, it is expressed as:

[0075]

[0076] Step 42, design the observation information cost function, and when the ranging error satisfies the Gaussian distribution, the ranging error function is the observation information cost function For a given known base station j, at time t, it is expressed as:

[0077]

[0078] Step 43, design the global target function, when the error satisfies the Gaussian distribution, the negative logarithm of the probability distribution is proportional to the error function, and the maximum a posteriori estimation can be converted into a nonlinear least squares problem, combined with the global target function described in step three, the least squares solution of the factor graph model of the above nodes and edges can be expressed as:

[0079]

[0080] In the step five, the specific steps for solving the global target function by using the Gauss-Newton method are:

[0081] Step 51, for the nonlinear least squares form function, The error term e is approximated by first-order Taylor expansion to become a linear least squares form function.

[0082] Step 52, the regression coefficient is optimized and adjusted for multiple times in a constantly iterative process, the difference between the regression coefficient and the optimal regression coefficient is constantly reduced, finally the smallest residual sum of squares is obtained through the algorithm, the global objective function is solved, and the state to be estimated of the maneuvering target in the current time slot is obtained.

[0083] The effect of the method is verified by experiments:

[0084] Figure 3 The effect diagram of the unordered sequential information least square positioning solution, Figure 4 The effect diagram of the sequential information traditional extended Kalman filter positioning solution, Figure 5 The effect diagram of the sequential information factor graph algorithm used in the application. It can be seen that, compared with the traditional extended Kalman filter method, the positioning accuracy of the method of the application is obviously better than that of the traditional Kalman filter method.

[0085] The application also provides a regional navigation positioning system based on sequential information, which comprises a state node factor unit, a state transition factor unit, an observation information factor unit, a global objective function unit and a factor graph solution unit.

[0086] The state node factor unit needs to set a variable node according to the actual flight scene of the maneuvering target, and the variable node should contain the coordinates of the state to be estimated of the maneuvering target;

[0087] The state transition factor unit needs to construct a state transition model by using speed information according to the actual flight situation, so as to constrain the transition process of the state to be estimated of the maneuvering target at adjacent time points;

[0088] The observation information factor unit needs to construct an observation information model by using prior information and corresponding sequential ranging information at different time points according to the actual flight situation, so as to represent the measurement process of the state to be estimated of the maneuvering target at different time points;

[0089] The global objective function unit needs to combine the state node factor, the state transition factor and the observation information factor to construct an overall factor graph model, so as to obtain a global objective function in the form of nonlinear least square;

[0090] The factor graph solution unit needs to solve the global objective function by using the Gauss-Newton method, so as to obtain the optimal state variable that minimizes the global objective function, that is, the state to be estimated of the maneuvering target, and obtain the navigation positioning result.

[0091] The application also provides an electronic device, Figure 6The structure of the electronic device provided by the embodiments of the present application is shown. For example, the electronic device 60 can include a processor 61, a memory 62, and a transmission device 636. The processor is configured to execute the sequential information-based regional navigation positioning method mentioned in the above embodiments. The processor and the memory can be connected by a bus or other means. The transmission device can be connected to the processor and the memory by wired or wireless means. The memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer-executable programs, and modules, such as program instructions / modules corresponding to the sequential information-based regional navigation positioning method in the embodiments of the present application. The processor executes various functions and data processing of the processor by running the non-transitory software programs, instructions, and modules stored in the memory, that is, implements the sequential information-based regional navigation positioning method in the above method embodiments. The memory can include a program storage area and a data storage area. The program storage area can store an operating system and application programs required by at least one function; and the data storage area can store data created by the processor and the like. In addition, the memory can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory can optionally include a memory remotely arranged with respect to the processor. These remote memories can be connected to the processor through a network. Examples of the network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The one or more modules are stored in the memory and executed by the processor to perform the sequential information-based regional navigation positioning method in the embodiments.

[0092] As another aspect, the present application also provides a computer-readable storage medium. The computer-readable storage medium can be the computer-readable storage medium included in the device described in the above embodiments; or can exist separately and not be assembled into the device. The computer-readable storage medium can be a tangible storage medium, such as a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a floppy disk, a hard disk, a removable storage disk, a CD-ROM, or any other form of storage medium known in the technical field. The computer-readable storage medium stores one or more programs used by one or more processors to execute the sequential information-based regional navigation positioning method described in the present application.

[0093] To sum up, the above is only a preferred embodiment of the present application, and is not used to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, and the like made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A regional navigation positioning method based on sequential information, characterized in that: The following steps are involved: Step 1: Use land-based radio to achieve regional navigation ranging: At each observation epoch, the maneuvering target receives the reply signals from N ground stations with known positions, calculates the signal propagation delay, and obtains N sequential observation information including distance, distance change rate, speed, speed change rate, measured distance error, and measured speed error; Step 2: Divide sequential information time slots: Set a single processing time slot to contain N time moments, ensuring that the maneuvering target receives the reply signals from all ground stations in sequence within the time slot; Step 3: Construct a factor graph model: Based on the sequential observation information of the current time slot, establish a model that includes the variable nodes of the maneuvering target's state to be estimated, as well as the prior information factor nodes, state transition factor nodes, and observation information factor nodes; Step 4: Construct the global objective function: Define the state transition cost function and the observation information cost function based on the factor graph model, and transform the state estimation into a joint optimization problem in the form of maximum a posteriori estimation; Step 5: Solve the objective function: Use the Gauss-Newton method to iteratively optimize the global objective function and obtain the optimal state variable that minimizes the function as the positioning result of the current time slot; Step 6: Global positioning update: Repeat steps 3 to 5 in the next time slot, solve the next epoch state based on the current epoch observation information, and iteratively output the global positioning result.

2. The method for regional navigation and positioning based on sequential information according to claim 1, wherein: The time slot division in step 2 must satisfy: the total number of moments T in the time slot and the number of ground stations N satisfy the mapping relationship T=kN (k≥1), k is the kth moment, and the response signal of each ground station is processed only once in the time slot.

3. The method for regional navigation and positioning based on sequential information according to claim 1 or 2, characterized in that: In step 3, the specific steps of constructing the factor graph model based on sequential observation information are: Step 31: Set the variable node, which should contain the coordinates of the maneuvering target state to be estimated; Step 32: Set a state transfer factor node, use the velocity measurement model to constrain the transfer process of the state to be estimated at adjacent moments of the maneuvering target, and regard the velocity measurement error as a Gaussian random distribution; Step 33: Set the observation information factor node, regard the ranging error as a Gaussian random distribution, and establish the covariance matrix based on the altitude angle model and the signal-to-noise ratio.

4. The method for regional navigation and positioning based on sequential information according to claim 3, characterized in that: In step 2, the covariance matrix is ​​adjusted by the nominal speed measurement accuracy of the receiver or multiple tests.

5. The method for regional navigation and positioning based on sequential information according to claim 3, wherein: In step 4, the specific steps of constructing the global objective function according to the factor graph model are: Step 41: Design a state transfer cost function. When the speed measurement error satisfies the Gaussian distribution, the speed measurement error function, i.e., the state transfer cost function, is obtained. Step 2: Design the observation information cost function. When the ranging error satisfies the Gaussian distribution, the ranging error function is the expression of the observation information cost function for a given known base station at a certain time. Step 3: Design a global objective function. When the error satisfies the Gaussian distribution, the negative logarithm of the probability distribution is proportional to the error function. The maximum a posteriori estimation can be transformed into a nonlinear least squares problem. Combined with the global objective function described in step 3, the least squares solution of the factor graph model of the above nodes and edges is obtained.

6. The method for regional navigation and positioning based on sequential information according to claim 1, wherein: In step 5, the specific steps of using the Gauss-Newton method to solve the global objective function are: Step 51: For the nonlinear least squares form function, perform first-order Taylor expansion approximation on the error term to convert it into a linear least squares form function; Step 52: The regression coefficient is optimized and adjusted multiple times in the process of continuous iteration, so that the gap between it and the optimal regression coefficient becomes smaller and smaller. Finally, the minimum residual sum of squares is obtained through this algorithm, and the global objective function is solved to obtain the estimated state of the maneuvering target in the current time slot.

7. A regional navigation and positioning system for sequential information, characterized in that: include: State node factor unit: configured to set a variable node containing state coordinates to be estimated according to the motion scenario of the maneuvering target; State transfer factor unit: configured to use velocity information to construct a state transfer model and constrain the state relationship at adjacent moments; Observation information factor unit: configured to fuse prior information with sequential ranging data to construct an observation model; Global objective function unit: configured to combine the three types of nodes to generate a global objective function in the form of nonlinear least squares; Factor graph solving unit: configured to use Gauss-Newton method to solve the target function and output the positioning result; Among them, the state transfer factor unit executes the state transfer factor node construction logic, and the observation information factor unit executes the observation information factor node construction logic.

8. An electronic device, characterized in that: include: Processor (61); a memory (62) storing a computer program; a transmission device (63) for communicating with an external base station; When the processor executes the computer program, the positioning method according to any one of claims 1 to 6 is implemented.

9. The electronic device according to claim 8, wherein: The transmission device (63) is configured to establish a communication link with N ground stations via ground-based radio in a satellite-denied environment.

10. A computer-readable storage medium having computer instructions stored thereon, characterized in that: When the instructions are executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images