Training efficiency device with intelligent positioning device

Abstract

The invention discloses a training performance device with an intelligent positioning device, and the performance device is used for executing a training method with the intelligent positioning device, and the method comprises the steps: executing data collection through an IMU, and constructing an IMU collection data set. And performing environmental influence discrimination based on the IMU acquisition data set, and generating an environmental influence discrimination result. GNSS signals and VO sensor signals are interacted, and the equipment position and state of the current node are determined. And selecting a working mode according to a signal judgment result and the position and state of the equipment, and carrying out combined positioning measurement and calculation. And carrying out data processing on a combined positioning measurement result based on robust adaptive federated filtering. And the information sharing coefficient is adaptively adjusted through the robust equivalent weight, data updating of the positioning direction and position is performed based on an adaptive adjustment result, and training is completed based on a data updating result. The technical problems of low positioning accuracy and poor positioning effect of course positioning of training equipment in a complex training environment in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a training efficiency device with an intelligent positioning device. Background Art

[0002] With the development of digital technology, traditional training methods have gradually transformed, and training equipment has gradually turned to intelligent digitalization. In recent years, the most common training mode is composed of laser model training equipment. During the training process, you only need to wear a helmet equipped with a laser receiving device, and both parties use the laser receiving device to judge the training results. However, this method has certain limitations. In the actual environment, there are certain errors in the target positioning and direction finding results, and the laser will be seriously interfered in the face of complex environments. The traditional single sensor positioning method not only has large errors in positioning accuracy, but also has many limitations in positioning scenarios. And with the development of modern digitalization, the accuracy requirements of intelligent devices are becoming more and more stringent. Faced with complex training venues, relying solely on a single sensor for positioning faces huge challenges.

[0003] Therefore, in the prior art, the heading positioning of training equipment in a complex training environment has the technical problems of low positioning accuracy and poor positioning effect. Summary of the invention

[0004] The present application solves the technical problems of low positioning accuracy and poor positioning effect in the heading positioning of training equipment in complex training environments in the prior art by providing a training efficiency device with an intelligent positioning device.

[0005] The present application provides a training efficiency device with an intelligent positioning device, the training efficiency device comprising: a data acquisition module, used to perform data acquisition through an IMU and construct an IMU acquisition data set, wherein the IMU comprises a gyroscope, an accelerometer, and a magnetometer; an environmental impact judgment module, used to perform environmental impact judgment based on the IMU acquisition data set and generate an environmental impact judgment result; a signal judgment module, used to interact with GNSS signals and VO sensor signals, perform signal judgment, and determine the device position and status of the current node; a positioning measurement module, used to select a working mode according to the signal judgment result and the device position and status, and perform combined positioning measurement based on the selected working mode, wherein the working mode includes VO-IMU positioning, GNSS-IMU positioning, and VO-GNSS-IMU positioning; a data processing module, used to perform data processing of the combined positioning measurement result based on robust adaptive federated filtering; a data update module, used to adaptively adjust the information sharing coefficient through robust equivalent weights, update the data of the positioning direction and position based on the adaptive adjustment result, and complete the training based on the data update result.

[0006] The present application also provides a training method with an intelligent positioning device, the method is applied to a training effector with an intelligent positioning device, the method includes: performing data collection through IMU, constructing an IMU collection data set, wherein the IMU includes a gyroscope, an accelerometer, and a magnetometer; performing environmental impact discrimination based on the IMU collection data set, and generating environmental impact discrimination results; interacting GNSS signals and VO sensor signals, and performing signal discrimination, and determining the device position and status of the current node; selecting a working mode according to the signal discrimination results and the device position and status, and performing combined positioning measurement based on the selected working mode. Among them, the working modes include VO-IMU positioning, GNSS-IMU positioning, and VO-GNSS-IMU positioning; performing data processing of the combined positioning measurement results based on robust adaptive federated filtering; adaptively adjusting the information sharing coefficient through robust equivalent weights, updating the positioning direction and position data based on the adaptive adjustment results, and completing the training based on the data update results.

[0007] The present application also provides an electronic device, including:

[0008] A memory for storing executable instructions;

[0009] The processor is used to implement a training method with an intelligent positioning device provided in the present application when executing the executable instructions stored in the memory.

[0010] The present application provides a computer-readable storage medium storing a computer program. When the program is executed by a processor, a training method for an intelligent positioning device provided by the present application is implemented.

[0011] A training effector with an intelligent positioning device proposed in this application is used to execute a training method with an intelligent positioning device, and the method includes: performing data collection through an IMU to construct an IMU collection data set. Performing environmental impact discrimination based on the IMU collection data set to generate environmental impact discrimination results. Interacting GNSS signals and VO sensor signals, and determining the device position and status of the current node. Selecting a working mode based on the signal discrimination results and the device position and status, and performing combined positioning calculations. Performing data processing on the combined positioning calculation results based on robust adaptive federal filtering. Adaptively adjusting the information sharing coefficient through robust equivalent weights, updating the positioning direction and position data based on the adaptive adjustment results, and completing training based on the data update results. Accurate positioning of the heading positioning of the training equipment in a complex training environment is achieved, and the positioning quality is improved. The technical problems of low positioning accuracy and poor positioning effect in the heading positioning of the training equipment in a complex training environment in the prior art are solved.

[0012] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following briefly introduces the drawings of the embodiments of the present disclosure. Obviously, the drawings described below only relate to some embodiments of the present disclosure, and are not intended to limit the present disclosure.

[0014] Figure 1 A flowchart of a training method with an intelligent positioning device provided in an embodiment of the present application;

[0015] Figure 2 A training method with an intelligent positioning device provided in an embodiment of the present application is a flowchart of VO-IMU positioning;

[0016] Figure 3 A flow chart of a training method with an intelligent positioning device provided in an embodiment of the present application, wherein the working mode is GNSS-IMU positioning;

[0017] Figure 4 A schematic diagram of the structure of a system of a training efficiency device with an intelligent positioning device provided in an embodiment of the present application.

[0018] Explanation of the accompanying drawings: data acquisition module 11, environmental impact determination module 12, signal determination module 13, positioning and calculation module 14, data processing module 15, data updating module 16. DETAILED DESCRIPTION

[0019] Embodiment 1

[0020] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0021] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application.

[0023] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, the modules are illustrative only, and different aspects of the system and method may use different modules.

[0024] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. At the same time, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0025] like Figure 4 As shown, the present application provides a training effector with an intelligent positioning device, the effector comprising:

[0026] A data acquisition module 11 is used to perform data acquisition through an IMU and construct an IMU acquisition data set, wherein the IMU includes a gyroscope, an accelerometer, and a magnetometer;

[0027] An environmental impact determination module 12 is used to perform environmental impact determination based on an IMU acquisition data set and generate an environmental impact determination result;

[0028] The signal discrimination module 13 is used to interact with the GNSS signal and the VO sensor signal, perform signal discrimination, and determine the device position and status of the current node;

[0029] The positioning and calculation module 14 is used to select a working mode according to the signal discrimination result and the position and state of the device, and perform combined positioning and calculation based on the selected working mode, wherein the working modes include VO-IMU positioning, GNSS-IMU positioning and VO-GNSS-IMU positioning;

[0030] A data processing module 15 is used for processing the data of the combined positioning measurement result based on the robust adaptive federated filtering;

[0031] The data updating module 16 is used to adaptively adjust the information sharing coefficient through the robust equivalent weight, update the data of the positioning direction and position based on the adaptive adjustment result, and complete the training based on the data updating result.

[0032] Furthermore, the environmental impact determination module 12 is also used for:

[0033] The IMU collects data sets for data analysis to obtain the initial position and initial attitude;

[0034] Extracting continuous acceleration and continuous measurement angular velocity from the IMU acquisition data set;

[0035] The continuous acceleration and the continuously measured angular velocity are respectively integrated over time to obtain the track, current position and current attitude;

[0036] Environmental impact judgment is performed through the initial position, initial attitude, track, current position and current attitude to generate an environmental impact judgment result.

[0037] Furthermore, the positioning and calculating module 14 is also used for:

[0038] When the selected working mode is VO-IMU positioning, read the VO acquisition data set;

[0039] Performing feature extraction and feature matching on the VO acquisition data set to obtain image features;

[0040] Performing device pose estimation based on the image features to generate a device pose estimation result, wherein the device pose estimation result includes a VO position result, a VO size result, and a VO direction result;

[0041] Performing IMU data integration on the continuous acceleration and the continuously measured angular velocity to obtain an IMU position result, an IMU heading result, and an IMU velocity result;

[0042] The VO position results, VO size results, VO direction results, IMU position results, IMU heading results and IMU speed results are fused to obtain the combined positioning measurement results.

[0043] Furthermore, the positioning and calculating module 14 is also used for:

[0044] When the selected working mode is GNSS-IMU positioning, obtain the GNSS acquisition data set;

[0045] Inputting continuous acceleration, continuous measurement angular velocity and GNSS acquisition data set into integrated navigation Kalamn filter for processing to generate an error correction set, wherein the error correction set includes heading angle error, attitude angle error, velocity error and position error;

[0046] Inputting the continuous acceleration and the continuously measured angular velocity into an inertial measurement unit, performing data processing through an inertial navigation solution algorithm, and obtaining a data processing result;

[0047] The data processing result is compensated for by the error correction set, and a combined positioning calculation result is generated based on the data compensation result.

[0048] Furthermore, the positioning and calculating module 14 is also used for:

[0049] When the selected working mode is VO-GNSS-IMU positioning, a multi-source fusion navigation system is constructed using VO-GNSS-IMU, and a combined positioning measurement result is obtained based on the multi-source fusion navigation system;

[0050] The IMU in the multi-source fusion navigation system is used as the main filter; IMU-VO and IMU-GNSS are used as sub-filters to update the time state to complete the robust adaptive federated filtering process.

[0051] Furthermore, the data processing module 15 is also used for:

[0052] Construct the UAV measurement update formula, the formula is as follows:

[0053] y k =E k x k +v k ;

[0054] Among them, y k is the state correction value, E k is the measurement matrix, x k is the state value, v k for speed;

[0055] Convert the GNSS acquisition data set into navigation coordinates to generate a GNSS measurement matrix;

[0056] The VO acquisition data set is used to obtain the position and attitude values ​​of the original image through pose calculation, and the VO measurement matrix is ​​generated;

[0057] The IMU acquisition data set is updated through the UAV measurement update formula, and the main filter is constructed based on the updated IMU acquisition data set;

[0058] Based on the updated IMU acquisition data set, the IMU-GNSS sub-filter and the IMU-VO sub-filter are respectively combined with the GNSS measurement matrix and the VO measurement matrix to construct the IMU-GNSS sub-filter and the IMU-VO sub-filter, and the robust adaptive federated filtering process is completed through the main filter and the MU-GNSS sub-filter and the IMU-VO sub-filter.

[0059] Furthermore, the data updating module 16 is also used for:

[0060] The observation error is represented by the residual, and the observation availability gain is changed by the robust equivalent weight coefficient. The formula is as follows:

[0061]

[0062]

[0063] in, is the residual of the i-th filter at time k in distributed filtering, is the position state of the ith filter at time k, is the measurement matrix, Measuring state covariance, is the covariance matrix, For the predicted state, is the predicted state covariance;

[0064] Normalize it and introduce the IGG3 weight function u i Robust processing is performed, and the residual gain matrix is ​​adaptively adjusted using the system normalized residual to complete the data processing of the combined positioning measurement results.

[0065] Furthermore, the data updating module 16 is also used for:

[0066] Get the precision λ of the sub-filter i (k) as follows:

[0067]

[0068] Among them, λ i (k) is the accuracy of the sub-filter, Q I (k) is the state covariance matrix, tr(·) is the matrix trace;

[0069] The accuracy of the sub-filter is updated by normalizing the residual and the current state variance;

[0070] The adaptive adjustment of the information sharing coefficient is completed through the accuracy update result.

[0071] The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other and are not used to limit the scope of protection of the present invention.

[0072] Embodiment 2

[0073] like Figure 1 As shown, the embodiment of the present application provides a training method with a smart positioning device, the method is applied to a training method with a smart positioning device, the method is applied to a training effector with a smart positioning device, the method includes:

[0074] Performing data acquisition through an IMU to construct an IMU acquisition data set, wherein the IMU includes a gyroscope, an accelerometer, and a magnetometer;

[0075] Perform environmental impact judgment based on IMU acquisition data set and generate environmental impact judgment results;

[0076] Interact GNSS signals and VO sensor signals, perform signal discrimination, and determine the device location and status of the current node;

[0077] Data collection is performed through IMU to construct an IMU collection data set, wherein the IMU includes a gyroscope, an accelerometer, and a magnetometer. Among them, IMU is a device for measuring the three-axis attitude angle and acceleration of an object. The gyroscope is used to measure the angle or angular velocity of the three axes, the accelerometer is used to measure the acceleration of the three axes, and the magnetometer provides magnetic field orientation information. Subsequently, environmental impact discrimination is performed based on the IMU collection data set to generate environmental impact discrimination results. The GNSS signal and the VO sensor signal are interacted, and signal discrimination is performed, that is, whether the GNSS has a signal, whether the VO sensor signal is normal, and the device position and status of the current node are determined. The device position of the current node is acquired using the position acquisition scheme of the GNSS and VO sensors in the prior art.

[0078] The IMU collects data sets for data analysis to obtain the initial position and initial attitude;

[0079] Extracting continuous acceleration and continuous measurement angular velocity from the IMU acquisition data set;

[0080] The continuous acceleration and the continuously measured angular velocity are respectively integrated over time to obtain the track, current position and current attitude;

[0081] Environmental impact judgment is performed through the initial position, initial attitude, track, current position and current attitude to generate an environmental impact judgment result.

[0082] The IMU collects the data set for data analysis to obtain the initial position and initial posture, that is, the position (X S ,Y S ,Z S), and the initial attitude, that is, yaw, pitch, roll (α, β, γ) in its own coordinate system. Subsequently, the fixed displacement data is used to perform the displacement of the IMU acquisition device, and the continuous acceleration and continuous measurement angular velocity in the IMU acquisition data set are extracted. The continuous acceleration and the continuous measurement angular velocity are respectively integrated over time to obtain the track, current position and current attitude. Among them, the processing of continuous acceleration is to integrate once to obtain the speed, and integrate again to obtain the displacement, and the processing of continuous measurement angular velocity is a secondary integration. Finally, the environmental impact is judged by the initial position, initial attitude, track, current position and current attitude, and the difference between the initial position, initial attitude and the current position and current attitude is obtained. The fixed displacement data is compared with the difference combined with the track to obtain the difference data, and the difference data is obtained. It is judged whether the difference data meets the preset difference data threshold, and the environmental impact judgment result is generated. The preset difference data threshold is the maximum value of the pre-set difference data. When the threshold is exceeded, the current application environment has a greater impact on the IMU, resulting in a lower accuracy of the sensor measurement result, and an early warning can be issued according to the environmental impact judgment result.

[0083] Select a working mode according to the signal discrimination result and the device position and status, and perform combined positioning calculation based on the selected working mode, where the working modes include VO-IMU positioning, GNSS-IMU positioning and VO-GNSS-IMU positioning;

[0084] Data processing of combined positioning results based on robust adaptive federated filtering;

[0085] The information sharing coefficient is adaptively adjusted through robust equivalent weights, the positioning direction and position data are updated based on the adaptive adjustment results, and the training is completed based on the data update results.

[0086] The working mode is selected according to the signal discrimination result and the position and status of the device, and the combined positioning and calculation are performed based on the selected working mode, wherein the working modes include VO-IMU positioning, GNSS-IMU positioning and VO-GNSS-IMU positioning. Since the signal discrimination result reflects whether each sensor signal exists, the working mode is selected according to the output of the sensor signal to perform combined positioning and calculation. Data processing of the combined positioning and calculation results is performed based on robust adaptive federal filtering. Finally, the information sharing coefficient is adaptively adjusted through robust equivalent weights, and the data of positioning direction and position is updated based on the adaptive adjustment results, and the training is completed based on the data update results. Thereby, accurate positioning of the heading positioning of the training equipment in a complex training environment is achieved, and the positioning effect is improved.

[0087] like Figure 2 As shown, the method provided in the embodiment of the present application also includes:

[0088] When the selected working mode is VO-IMU positioning, read the VO acquisition data set;

[0089] Performing feature extraction and feature matching on the VO acquisition data set to obtain image features;

[0090] Performing device pose estimation based on the image features to generate a device pose estimation result, wherein the device pose estimation result includes a VO position result, a VO size result, and a VO direction result;

[0091] Performing IMU data integration on the continuous acceleration and the continuously measured angular velocity to obtain an IMU position result, an IMU heading result, and an IMU velocity result;

[0092] The VO position results, VO size results, VO direction results, IMU position results, IMU heading results and IMU speed results are fused to obtain the combined positioning measurement results.

[0093] When the selected working mode is VO-IMU positioning, wherein VO-IMU positioning is a combination positioning of a visual range sensor and an inertial measurement unit IMU, the VO acquisition data set is read, and feature extraction and feature matching are performed on the VO acquisition data set to obtain image features. VO has strong autonomy, can improve high-precision position and attitude information, does not produce drift, and can directly measure rotation and translation, but is blocked by the image and disturbed by moving objects, monocular vision cannot test the scale, pure rotational motion cannot be estimated, and is easily lost during rapid motion. Based on the image features, the device posture is estimated to generate a device posture estimation result, wherein the device posture estimation result includes a VO position result, a VO size result, and a VO direction result. The continuous acceleration and the continuously measured angular velocity are integrated by IMU data to obtain the IMU position result, the IMU heading result, and the IMU velocity result. The VO position result, the VO size result, the VO direction result, the IMU position result, the IMU heading result, and the IMU velocity result are data fused to obtain a combined positioning measurement result. By combining the visual odometry sensor with the inertial measurement unit, they can compensate for each other's shortcomings and improve the output accuracy of position information.

[0094] like Figure 3 As shown, the method provided in the embodiment of the present application also includes:

[0095] When the selected working mode is GNSS-IMU positioning, obtain the GNSS acquisition data set;

[0096] Inputting continuous acceleration, continuous measurement angular velocity and GNSS acquisition data set into integrated navigation Kalamn filter for processing to generate an error correction set, wherein the error correction set includes heading angle error, attitude angle error, velocity error and position error;

[0097] Inputting the continuous acceleration and the continuously measured angular velocity into an inertial measurement unit, performing data processing through an inertial navigation solution algorithm, and obtaining a data processing result;

[0098] The data processing result is compensated for by the error correction set, and a combined positioning calculation result is generated based on the data compensation result.

[0099] When the selected working mode is GNSS-IMU positioning, a GNSS acquisition data set is obtained. The combined GNSS-IMU positioning system can form a good complementary advantage. The navigation parameters of GNSS can be used to correct the accumulated position error of IMU to a limited extent. The inertial position solution output by IMU is used as a prediction, and the output position signal of GNSS is used as an observation, and the optimal a posteriori position information is obtained by fusion correction. By inputting continuous acceleration, continuous measurement angular velocity and GNSS acquisition data set into the combined navigation Kalamn filter for processing, an error correction set is generated, and the error correction set includes heading angle error, attitude angle error, velocity error and position error. The continuous acceleration and continuous measurement angular velocity are input into the inertial measurement unit, and data processing is performed through the inertial navigation solution algorithm to obtain the data processing result. The data processing result is data compensated by the error correction set, and a combined positioning measurement result is generated based on the data compensation result.

[0100] The method provided in the embodiment of the present application also includes:

[0101] When the selected working mode is VO-GNSS-IMU positioning, a multi-source fusion navigation system is constructed using VO-GNSS-IMU, and a combined positioning measurement result is obtained based on the multi-source fusion navigation system;

[0102] The IMU in the multi-source fusion navigation system is used as the main filter; IMU-VO and IMU-GNSS are used as sub-filters to update the time state to complete the robust adaptive federated filtering process.

[0103] Specifically, when the selected working mode is VO-GNSS-IMU positioning, that is, when positioning is selected through the combination of VO sensor, GNSS sensor, and IMU sensor, VO-GNSS-IMU is used to build a multi-source fusion navigation system, and the combined positioning measurement result is obtained based on the multi-source fusion navigation system. The IMU in the multi-source fusion navigation system is used as the main filter. IMU-VO and IMU-GNSS are used as sub-filters to update the time state to complete the robust adaptive federated filtering process.

[0104] The method provided in the embodiment of the present application also includes:

[0105] Construct the UAV measurement update formula, the formula is as follows:

[0106] y k =E k x k +v k ;

[0107] Among them, y k is the state correction value, E k is the measurement matrix, x k is the state value, v k for speed;

[0108] Convert the GNSS acquisition data set into navigation coordinates to generate a GNSS measurement matrix;

[0109] The VO acquisition data set is used to obtain the position and attitude values ​​of the original image through pose calculation, and the VO measurement matrix is ​​generated;

[0110] The IMU acquisition data set is updated through the UAV measurement update formula, and the main filter is constructed based on the updated IMU acquisition data set;

[0111] Based on the updated IMU acquisition data set, the IMU-GNSS sub-filter and the IMU-VO sub-filter are respectively combined with the GNSS measurement matrix and the VO measurement matrix to construct the IMU-GNSS sub-filter and the IMU-VO sub-filter, and the robust adaptive federated filtering process is completed through the main filter and the MU-GNSS sub-filter and the IMU-VO sub-filter.

[0112] Construct the UAV measurement update formula, the formula is as follows: k =E k x k +v k Among them, y k is the state correction value, E k is the measurement matrix, x k is the state value, v k is the speed. The GNSS acquisition data set is converted into navigation coordinates to generate a GNSS measurement matrix. The VO acquisition data set is used to obtain the position and attitude values ​​of the original image through posture calculation to generate a VO measurement matrix. The IMU acquisition data set is updated through the UAV measurement update formula, and the main filter is constructed based on the updated IMU acquisition data set. Finally, based on the updated IMU acquisition data set, the GNSS measurement matrix and the VO measurement matrix are combined respectively to construct an IMU-GNSS sub-filter and an IMU-VO sub-filter, and the robust adaptive federated filtering process is completed through the main filter, the MU-GNSS sub-filter and the IMU-VO sub-filter.

[0113] The method provided in the embodiment of the present application also includes:

[0114] The observation error is represented by the residual, and the observation availability gain is changed by the robust equivalent weight coefficient. The formula is as follows:

[0115]

[0116]

[0117] in, is the residual of the i-th filter at time k in distributed filtering, is the position state of the ith filter at time k, is the measurement matrix, Measuring state covariance, is the covariance matrix, For the predicted state, is the predicted state covariance;

[0118] Normalize it and introduce the IGG3 weight function u i Robust processing is performed, and the residual gain matrix is ​​adaptively adjusted using the system normalized residual to complete the data processing of the combined positioning measurement results.

[0119] The robust adaptive federated filtering process also includes that the system state residual is determined by the model error and the observation error. When the model error is small, the observation error is represented by the residual, and the observation availability gain is changed by the robust equivalent weight coefficient, as shown in the following formula:

[0120]

[0121] in, is the residual of the i-th filter at time k in distributed filtering, is the position state of the ith filter at time k, is the measurement matrix, Measuring state covariance, is the covariance matrix, For the predicted state, is the predicted state covariance. It is normalized, and the normalization process is as follows: Where: v i 、e i 、w i are the normalized result, residual, and covariance equation of the ith subfilter residual. And the IGG3 weight function u is introduced i Robust processing is performed, and the residual gain matrix is ​​adaptively adjusted using the system normalized residual. The residual gain matrix is ​​adaptively adjusted using the system normalized residual.

[0122]

[0123] vi Obey the standard state distribution: v i ~N(0,1), robust processing is performed on observations exceeding the 95% confidence level, where k0 is set to 1 and k1 is set to 2. After the robustness processing, the measured values ​​are updated as follows:

[0124]

[0125] Where: For the predicted state, is the gain matrix, To update the status, is the average measured state covariance, is the predicted state covariance at the k+1th moment, and N is the number of sub-filters, thereby completing the data processing of the combined positioning measurement results.

[0126] The method provided in the embodiment of the present application also includes:

[0127] Get the precision λ of the sub-filter i (k) as follows:

[0128]

[0129] Among them, λ i (k) is the accuracy of the sub-filter, Q I (k) is the state covariance matrix, tr(·) is the matrix trace;

[0130] The accuracy of the sub-filter is updated by normalizing the residual and the current state variance;

[0131] The adaptive adjustment of the information sharing coefficient is completed through the accuracy update result.

[0132] In practical applications, considering the inconsistency of filtering accuracy of different sub-filters, it is necessary to adjust the proportion of information allocated to each sub-filter according to the filtering accuracy. i (k) as follows: Among them, λ i (k) is the accuracy of the sub-filter, Q I (k) is the state covariance matrix, and tr(·) is the matrix trace. The accuracy of the sub-filter is updated by the normalized residual and the current state variance. i reflects the possibility of filtered observation, so it can be combined with v i and To comprehensively consider the accuracy of the sub-filter is as follows: Assume that information sharing satisfies the following relationship:

[0133] Where: βi (k) is the information sharing coefficient. Since the main filter does not distribute information, the adaptive sharing coefficient is:

[0134] Finally, the adaptive adjustment of the information sharing coefficient is completed through the accuracy update result.

[0135] The technical solution provided by the embodiment of the present invention performs data collection through IMU to construct an IMU collection data set, wherein the IMU includes a gyroscope, an accelerometer, and a magnetometer. Environmental impact discrimination is performed based on the IMU collection data set to generate an environmental impact discrimination result. GNSS signals and VO sensor signals are exchanged, and signal discrimination is performed, and the device position and state of the current node are determined. A working mode is selected according to the signal discrimination result and the device position and state, and a combined positioning measurement is performed based on the selected working mode, wherein the working mode includes VO-IMU positioning, GNSS-IMU positioning, and VO-GNSS-IMU positioning. Data processing of the combined positioning measurement result is performed based on robust adaptive federal filtering. The information sharing coefficient is adaptively adjusted through robust equivalent weights, and the data of the positioning direction and position is updated based on the adaptive adjustment result, and the training is completed based on the data update result. Accurate positioning of the heading positioning of the training equipment in a complex training environment is achieved, and the positioning quality is improved. The technical problem of low positioning accuracy and poor positioning effect of the heading positioning of the training equipment in a complex training environment in the prior art is solved.

[0136] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A training efficiency device with an intelligent positioning device, It is characterized in that The training efficacy device comprises: A data acquisition module, used to perform data acquisition through an IMU and construct an IMU acquisition data set, wherein the IMU includes a gyroscope, an accelerometer, and a magnetometer; The environmental impact identification module is used to identify the environmental impact based on the IMU acquisition data set and generate the environmental impact identification result; The signal discrimination module is used to interact with GNSS signals and VO sensor signals, perform signal discrimination, and determine the device location and status of the current node; A positioning and calculation module is used to select a working mode according to the signal discrimination result and the device position and status, and perform combined positioning and calculation based on the selected working mode, wherein the working modes include VO-IMU positioning, GNSS-IMU positioning and VO-GNSS-IMU positioning; A data processing module, used for data processing of combined positioning measurement results based on robust adaptive federated filtering; The data updating module is used to adaptively adjust the information sharing coefficient through the robust equivalent weight, update the data of the positioning direction and position based on the adaptive adjustment result, and complete the training based on the data updating result.

2. The training efficiency device according to claim 1, It is characterized in that The environmental impact identification module is also used for: The IMU collects data sets for data analysis to obtain the initial position and initial attitude; Extracting continuous acceleration and continuous measurement angular velocity from the IMU acquisition data set; The continuous acceleration and the continuously measured angular velocity are respectively integrated over time to obtain the track, current position and current attitude; Environmental impact judgment is performed through the initial position, initial attitude, track, current position and current attitude to generate an environmental impact judgment result.

3. The training efficiency device according to claim 2, It is characterized in that The positioning and calculation module is also used for: When the selected working mode is VO-IMU positioning, read the VO acquisition data set; Performing feature extraction and feature matching on the VO acquisition data set to obtain image features; Performing device pose estimation based on the image features to generate a device pose estimation result, wherein the device pose estimation result includes a VO position result, a VO size result, and a VO direction result; Performing IMU data integration on the continuous acceleration and the continuously measured angular velocity to obtain an IMU position result, an IMU heading result, and an IMU velocity result; The VO position results, VO size results, VO direction results, IMU position results, IMU heading results and IMU speed results are fused to obtain the combined positioning measurement results.

4. The training efficiency device according to claim 2, It is characterized in that The positioning and calculation module is also used for: When the selected working mode is GNSS-IMU positioning, obtain the GNSS acquisition data set; Inputting continuous acceleration, continuous measurement angular velocity and GNSS acquisition data set into integrated navigation Kalamn filter for processing to generate an error correction set, wherein the error correction set includes heading angle error, attitude angle error, velocity error and position error; Inputting the continuous acceleration and the continuously measured angular velocity into an inertial measurement unit, performing data processing through an inertial navigation solution algorithm, and obtaining a data processing result; The data processing result is compensated for by the error correction set, and a combined positioning calculation result is generated based on the data compensation result.

5. The training efficiency device according to claim 2, It is characterized in that The positioning and calculation module is also used for: When the selected working mode is VO-GNSS-IMU positioning, a multi-source fusion navigation system is constructed using VO-GNSS-IMU, and a combined positioning measurement result is obtained based on the multi-source fusion navigation system; Taking IMU in multi-source fusion navigation system as main filter; The time state is updated with IMU-VO and IMU-GNSS as sub-filters to complete the robust adaptive federated filtering process.

6. The training efficiency device according to claim 5, It is characterized in that The data processing module is also used for: Construct the UAV measurement update formula, the formula is as follows: and k =E k x k +v k ; Among them, y k is the state correction value, E k is the measurement matrix, x k is the state value, v k for speed; Convert the GNSS acquisition data set into navigation coordinates to generate a GNSS measurement matrix; The VO acquisition data set is used to obtain the position and attitude values ​​of the original image through pose calculation, and the VO measurement matrix is ​​generated; The IMU acquisition data set is updated through the UAV measurement update formula, and the main filter is constructed based on the updated IMU acquisition data set; Based on the updated IMU acquisition data set, the IMU-GNSS sub-filter and the IMU-VO sub-filter are respectively combined with the GNSS measurement matrix and the VO measurement matrix to construct the IMU-GNSS sub-filter and the IMU-VO sub-filter, and the robust adaptive federated filtering process is completed through the main filter and the MU-GNSS sub-filter and the IMU-VO sub-filter.

7. The training efficiency device according to claim 6, It is characterized in that The data updating module is also used for: The observation error is represented by the residual, and the observation availability gain is changed by the robust equivalent weight coefficient. The formula is as follows: in, is the residual of the i-th filter at time k in distributed filtering, is the position state of the ith filter at time k, is the measurement matrix, Measuring state covariance, is the covariance matrix, For the predicted state, is the predicted state covariance; Normalize it and introduce the IGG3 weight function u i Robust processing is performed, and the residual gain matrix is ​​adaptively adjusted using the system normalized residual to complete the data processing of the combined positioning measurement results.

8. The training efficiency device according to claim 7, It is characterized in that The data updating module is also used for: Get the precision λ of the sub-filter i (k) as follows: Among them, λ i (k) is the accuracy of the sub-filter, Q I (k) is the state covariance matrix, tr(·) is the matrix trace; The accuracy of the sub-filter is updated by normalizing the residual and the current state variance; The adaptive adjustment of the information sharing coefficient is completed through the accuracy update result.

9. A training method with an intelligent positioning device, It is characterized in that The method is applied to a training efficiency device with an intelligent positioning device, and the method comprises: Performing data acquisition through an IMU to construct an IMU acquisition data set, wherein the IMU includes a gyroscope, an accelerometer, and a magnetometer; Perform environmental impact judgment based on IMU acquisition data set and generate environmental impact judgment results; Interact GNSS signals and VO sensor signals, perform signal discrimination, and determine the device location and status of the current node; Select a working mode according to the signal discrimination result and the device position and status, and perform combined positioning calculation based on the selected working mode, where the working modes include VO-IMU positioning, GNSS-IMU positioning and VO-GNSS-IMU positioning; Data processing of combined positioning results based on robust adaptive federated filtering; The information sharing coefficient is adaptively adjusted through robust equivalent weights, the positioning direction and position data are updated based on the adaptive adjustment results, and the training is completed based on the data update results.

Images