Full-period measurement method and system for shooting aiming point track

Abstract

The invention discloses a full-period measurement method and system for a shooting aiming point track, and the method comprises the following steps: collecting a positioning light signal from a target based on a position sensor installed on a shooter, and obtaining a first on-target track of an area in the target according to the positioning light signal; acquiring an attitude signal of the shooter based on an inertial measurement unit mounted on the shooter, and obtaining a second on-target track of the region in the target and an out-target track of the region outside the target according to the attitude signal of the shooter; and according to the first on-target track, the second on-target track and the outside-target track, a full-period shooting aiming point track is obtained.

Description

Technical Field

[0001] This invention relates to the field of shooting training technology, specifically to a method and system for measuring the full-cycle trajectory of a shooting aiming point. Background Technology

[0002] Shooting is a highly specialized and technical sport. Whether in military, law enforcement, and security fields, or for legitimate civilian shooting enthusiasts, systematic and scientific shooting training is crucial. Shooting training includes basic theoretical learning, dry-fire practice, and live-fire practice. Basic theoretical learning primarily focuses on understanding the principles of firearms and aiming, and learning the essential techniques for aiming and firing. Dry-fire practice or live-fire practice involves training aiming and firing techniques under conditions of no firearms or live ammunition.

[0003] During training, the point where the aiming line intersects the target plane is the aiming point, and the change of the aiming point over time constitutes the aiming point trajectory. The aiming point trajectory can reflect the shooter's body posture stability, psychology, and shooting habits during training. By quantitatively measuring and analyzing the shooter's aiming point throughout the entire process, targeted guidance can be provided to the shooter, which is of great significance for improving their shooting level.

[0004] The inventors disclosed a shooting aiming training system in patent CN118670196A. This system utilizes a photoelectric detector to collect laser light reflected from corner reflectors on a target to determine the trajectory of the aiming point. Simultaneously, an inertial measurement unit (IMU) measures the gun's attitude and vibration signals to detect the firing action, enabling quantitative analysis of the aiming point trajectory measurement during shooting. In patent CN118532998A, the inventors used an infrared light source placed on the target to image the target within a scope, obtaining the actual movement trajectory of the aiming point for analysis and adjustment of the shooter's aiming ability.

[0005] However, existing technologies can only acquire the trajectory of the aiming point on the target during aiming and firing, lacking data support for the state during the interval between two shots. In other words, existing technologies cannot measure the entire cycle of the aiming point's movement trajectory from the start of aiming to the moment of the second shot during two consecutive shots. Consequently, they cannot assess whether the shooter can quickly establish an aiming line, whether there are unnecessary corrections or jitters, or whether the aiming point has deviated from the target during continuous shooting. Summary of the Invention

[0006] One objective of this invention is to provide a method for measuring the full-cycle trajectory of a shooting aiming point. This method can not only obtain the trajectory on the target but also the trajectory outside the target within a certain area. After splicing, the aiming point movement trajectory from the start of aiming to the second firing is obtained during two consecutive shooting processes, realizing the full-cycle shooting aiming point trajectory restoration in shooting training and providing complete data support for shooting training evaluation.

[0007] This invention is achieved through the following technical solution:

[0008] A method for measuring the trajectory of a firing aiming point throughout its entire lifecycle includes the following steps:

[0009] The positioning light signal from the target is collected by the position sensor installed on the firing device, and the first target trajectory in the area inside the target is obtained based on the positioning light signal.

[0010] The inertial measurement unit installed on the shooter collects the shooter's attitude signal, and obtains the second on-target trajectory in the inner region of the target and the outer trajectory in the outer region of the target based on the shooter's attitude signal.

[0011] Based on the trajectory on the first target, the trajectory on the second target, and the trajectory outside the target, the firing aiming point trajectory for the entire cycle is obtained.

[0012] In this technical solution, both the position sensor (PSD) and the inertial measurement unit (IMU) are mounted on the firing device. Based on the position sensor, a precise aiming point trajectory measurement method is used to accurately measure the aiming process, obtaining the first target trajectory within the target area. Based on the inertial measurement unit, a gun body motion attitude measurement method is used to measure the gun body attitude throughout the entire firing process, obtaining the second target trajectory within the target area and the off-target trajectory outside the target area. Finally, based on the first, second, and off-target trajectories, the full-cycle aiming point trajectory is constructed.

[0013] In this technical solution, a position sensor is used to collect positioning light signals from the target. The main principle of the position sensor is based on the silicon photovoltaic effect, where the magnitude of the photocurrent generated when light shines on its surface is proportional to the position of the light source. By measuring the magnitude of the photocurrent, the position of the light source can be determined, thus obtaining the trajectory of the aiming point. Since the PSD can provide position measurement with micron-level resolution, it can achieve high-precision aiming point trajectory measurement within the target area.

[0014] In some embodiments, the positioning light signal from the target can be reflected to the PSD by installing a laser emitter on the shooter and setting a corner reflector on the target. In some embodiments, the positioning light signal from the target can also be the light signal directly emitted by an infrared light source emitter set on the target.

[0015] In some implementations, the PSD can be replaced with a CCD or CMOS detector, and the position of the light spot can be obtained through a light spot recognition algorithm to obtain the trajectory of the first aiming point.

[0016] In this technical solution, an inertial measurement unit (IMU) is used to acquire the firing device's attitude signal, thereby obtaining the aiming point trajectory based on the firing device's attitude signal. Its main principle is to estimate the direction of the aiming line based on the firing device's attitude signal, and then, by combining the direction of the aiming line and the firing distance, to make the aiming line intersect with a plane at the firing distance, obtain the aiming points in the inner and outer regions of the target. The second target trajectory and the outer trajectory are obtained along a time period. Because the IMU can estimate the aiming point based on the firing device itself, such as the muzzle attitude, the aiming point position it can acquire is located in the outer region of the target, thus obtaining a larger acquisition range compared to a PSD (Pressure Point Detector). However, precisely because the aiming point is estimated using the muzzle attitude, there is a certain error between the aiming point trajectory obtained by the IMU and the actual aiming point trajectory.

[0017] In this technical solution, after obtaining the first and second target trajectories and the off-target trajectories, by fusing and splicing the target trajectories and the off-target trajectories, it is possible not only to obtain the aiming point trajectory from the first gun-raising action to firing, but also to obtain the aiming point trajectory from firing to resetting, that is, the second gun-raising and aiming action. This enables the acquisition of trajectories in two consecutive shooting processes, and thus the reconstruction of the aiming point trajectories of multiple consecutive shootings, providing full-cycle data support for shooting training and evaluation.

[0018] As a preferred trajectory fusion method in this invention, the first and second target trajectories are first fused, and then the fused target trajectory is spliced ​​with the off-target trajectory. Specifically, the first and second target trajectories are fused using an iterative nearest-point algorithm to obtain a fused target trajectory, and the fused target trajectory is spliced ​​with the off-target trajectory to obtain the full-cycle shooting aiming point trajectory.

[0019] In this technical solution, the Iterative Closest Point (ICP) algorithm is used to fuse the trajectories of two targets. ICP iteratively searches for the closest point pairs and calculates the transformation matrix until convergence, effectively handling translation and rotation. Specifically, the first coordinates (x, y, x) of each aiming point on the first target trajectory within the target area are extracted. pm , z pm ), and the second coordinates (x) of each aiming point on the trajectory of the second target.qn , z qn Next, the Euclidean distance between the first and second coordinate points is calculated, and the transformation matrix [R, T] is solved, where R is the rotation matrix and T is the translation vector, to minimize the Euclidean distance. Finally, through iteration, the process stops when the change in error before and after the iteration is less than a preset value, and the fused trajectory of the first and second targets is output.

[0020] In this technical solution, after obtaining the fused trajectory, a full-cycle aiming point trajectory is obtained by stitching together the fused trajectory within the target area and the off-target trajectory outside the target area. In some preferred embodiments, smooth spline interpolation can be used to stitch together the on-target and off-target trajectories to ensure the continuity of the entire trajectory.

[0021] As another preferred trajectory fusion method in this invention, an error correction model is first obtained based on the first and second target trajectories. Then, the off-target trajectory is corrected using the error correction model, and finally, the corrected off-target trajectory and the first target trajectory are spliced ​​together. Specifically, an error correction model is established based on the first and second target trajectories. Based on the error correction model, the off-target trajectory is corrected. The first target trajectory and the corrected off-target trajectory are then spliced ​​together to obtain the full-cycle firing aiming point trajectory.

[0022] In this technical solution, Thin Plate Spline (TPS) curves are used to obtain the error correction model. TPS aims to find a smooth mapping function that matches a set of control points as accurately as possible to another set of target points while maintaining the smoothness of the entire transformation. Specifically, the aiming points on the first target trajectory (acquired by PSD with higher precision) are used as control points, and the aiming points on the second target trajectory (acquired by IMU) are used as target points. The TPS transformation includes affine and non-affine transformations. Affine transformations are applied globally to model global deformations (translation, rotation, scaling); non-affine transformations are applied locally to model local deformations. Based on the control points and target points, the transformation parameters in the TPS model are obtained, resulting in the TPS model, i.e., the error correction model. In this technical solution, after obtaining the error correction model, it is used for the off-target trajectory obtained based on IMU data, thereby significantly improving the accuracy of the off-target trajectory.

[0023] In this technical solution, after obtaining the corrected target trajectory, for the area inside the target, a first target trajectory with higher accuracy is selected and spliced ​​with the corrected target trajectory with improved accuracy for the area outside the target to obtain a full-cycle shooting aiming point trajectory. In some preferred embodiments, smooth spline interpolation can be used to splice the target trajectory and the target trajectory to ensure the continuity of the entire trajectory.

[0024] Furthermore, in order to reduce noise such as Gaussian white noise, random walk, and zero-bias instability in the shooter attitude signal acquired by the IMU, this technical solution constructs an IMU noise model, extends the IMU noise model into an attitude domain noise model, and uses maximum likelihood estimation or Kalman filtering algorithm to optimize the parameters of the attitude domain noise model. Based on the optimized attitude domain noise model, the shooter attitude signal is denoised.

[0025] In this technical solution, an IMU noise model is first constructed, which mainly includes Gaussian white noise, random walk, and zero-bias instability. Then, the Allen variance model is used to extend the IMU noise model into an attitude domain noise model. For example, time-domain noise is passed through the attitude solution equations to obtain the attitude domain noise model.

[0026] In this technical solution, the target light signal measured by PSD is converted into a high-precision attitude angle of the shooter, which is used to correct the original attitude angle signal measured by IMU, thereby obtaining the noise model parameters of the IMU attitude angle measurement. Specifically, maximum likelihood estimation or Kalman filtering algorithms can be used to optimize the parameters of the attitude domain noise model, such as Gaussian white noise variance, random walk coefficients, and zero-bias time-varying model parameters, to minimize the root mean square of the error, ensuring that the optimized attitude domain noise model can accurately fit the IMU noise characteristics. Finally, the optimized attitude domain noise model is used to denoise the shooter attitude signal to obtain more accurate aiming points for the second target trajectory and the off-target trajectory. This not only facilitates the fusion of the target trajectory and the establishment of the TPS model, but also reduces the difficulty of correcting the off-target trajectory and further improves the accuracy of the off-target trajectory.

[0027] Furthermore, it also includes a trajectory correction process on the first target, comprising the following steps:

[0028] Establish a trajectory correction model for the first target;

[0029] Obtain the actual coordinates and test coordinates of the specified calibration point on the target;

[0030] Based on the actual coordinates and test coordinates, the coefficients in the trajectory correction model on the first target are fitted using the least squares method to obtain the fitted trajectory correction model.

[0031] The trajectory on the first target is corrected using the fitted trajectory correction model.

[0032] In this technical solution, in order to solve the problem of nonlinear error in the position label output of PSD caused by the non-uniformity of resistivity on the photosensitive surface of the position sensor, nonlinear correction is performed on the PSD.

[0033] Specifically, calibration points are set on the target and their actual coordinates are obtained. Then, the firing device is used to aim at each calibration point and obtain the test coordinates of each calibration point.

[0034] A trajectory correction model on the first target is selected, which may include two quadratic polynomials. The actual coordinates and test coordinates of each calibration point are substituted into the two quadratic polynomials, and the coefficients in the polynomials are fitted using the least squares method. When the goodness of fit is greater than or equal to a preset value, the fitted coefficients are used as the coefficients of the polynomials to obtain the fitted trajectory correction model.

[0035] During testing, the original positioning optical signal data is input into the fitted trajectory correction model to obtain the positioning optical signal after PSD nonlinear correction. Subsequently, the trajectory on the first target is obtained based on the corrected positioning optical signal to further improve the accuracy of the trajectory on the first target.

[0036] Furthermore, before fusing the first target trajectory with the second target trajectory and / or the off-target trajectory, the coordinate systems of the two data sources are aligned. Specifically, this includes the following steps:

[0037] Based on the first coordinates of each trajectory point in the first target trajectory and the second coordinates of each trajectory point in the second target trajectory, an initial rigidity transformation is set, which includes a rotation matrix R and a translation vector T.

[0038] Adjust the rotation matrix R and translation vector T so that the error E = To reach the minimum value, in the formula, A i Let B be the first coordinate. i Using the second coordinate, obtain the rotation matrix and translation vector corresponding to the minimum error;

[0039] Based on the rotation matrix and translation vector corresponding to the minimum error, the second coordinate is transformed to the coordinate system corresponding to the first coordinate.

[0040] Furthermore, in order to accurately evaluate the firing aiming point trajectory, the entire firing aiming point trajectory is divided into three segments. The first segment is from the first time the gun is raised to aiming, the second segment is from aiming to firing, and the third segment is from firing to the second time the gun is raised.

[0041] Obtain the value of the first parameter of the first trajectory segment, the value of the second parameter of the second trajectory segment, and the value of the third parameter of the third trajectory segment. Based on the values ​​of the first parameter, the second parameter, and the third parameter, evaluate the quality of shooting training.

[0042] Furthermore, the sum of the areas of the inner and outer regions of the target corresponding to the shooting device attitude signal is 2 to 10 times the area of ​​the inner region of the target corresponding to the positioning light signal.

[0043] When calculating the aiming point based on the shooter's attitude signal measured by the inertial measurement unit (IMU), if the angle of attack of the aiming line, i.e., the pitch angle, exceeds a certain angle, the aiming point intersects the target plane at a very far distance, resulting in a trajectory that is too small and meaningless. Therefore, trajectories outside a certain range within the target area are no longer calculated. In this technical solution, preferably, the area measured by the IMU is 2 to 10 times the area measured by the PSD.

[0044] Another object of the present invention is to provide a full-cycle measurement system for the trajectory of a firing aiming point, the system comprising:

[0045] A position sensor, mounted on the firing mechanism, is used to collect positioning light signals from the target;

[0046] An inertial measurement unit, mounted on the firing mechanism, is used to acquire the firing mechanism's attitude signals;

[0047] The processing unit is used to transmit the positioning optical signal and the shooter attitude signal to the analysis unit;

[0048] The analysis unit obtains a first target trajectory within the target area based on the positioning light signal, and a second target trajectory within the target area and an external target trajectory outside the target area based on the shooter attitude signal. It then uses an iterative nearest-point algorithm to fuse the first and second target trajectories to obtain a fused target trajectory. Finally, it stitches the fused target trajectory and the external target trajectory together to obtain the full-cycle aiming point trajectory.

[0049] Based on the first target trajectory and the second target trajectory, an error correction model is established. Based on the error correction model, the off-target trajectory is corrected. The first target trajectory and the corrected off-target trajectory are then spliced ​​together to obtain the full-cycle shooting aiming point trajectory.

[0050] Furthermore, it also includes an evaluation unit, which divides the entire shooting aiming trajectory into three segments: the first segment is from the first time the gun is raised to aiming, the second segment is from aiming to firing, and the third segment is from firing to the second time the gun is raised; it obtains the value of the first parameter of the first segment, the value of the second parameter of the second segment, and the value of the third parameter of the third segment, and evaluates the shooting training quality based on the values ​​of the first parameter, the second parameter, and the third parameter.

[0051] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0052] 1. After obtaining the first and second target trajectories and the target-outside trajectory, this invention can not only obtain the aiming point trajectory from the first gun-raising action to firing, but also the aiming point trajectory from firing to resetting, that is, the second gun-raising and aiming, by fusing and splicing the target-on and target-outside trajectories, but also obtain the aiming point trajectory from firing to resetting, that is, the second gun-raising and aiming. This enables the acquisition of the trajectory in two consecutive shooting processes, and thus can reconstruct the aiming point trajectory of multiple consecutive shootings, providing full-cycle data support for shooting training and evaluation.

[0053] 2. This invention uses PSD data and IMU data to obtain the first and second target trajectories, obtains an error correction model based on the first and second target trajectories, then uses the error correction model to correct the target-outside trajectory obtained from the IMU data, and finally splices the first target-onside trajectory and the target-outside trajectory together, which can significantly improve the accuracy of the target-outside trajectory.

[0054] 3. This invention utilizes the error between the high-precision attitude angle calculated by vision and the original attitude angle calculated by the IMU to optimize the parameters of the attitude domain noise model, and uses the optimized attitude domain noise model to denoise the shooter attitude signal to obtain more accurate aiming points for the second target trajectory and the off-target trajectory. This not only facilitates the fusion of the target trajectory and the establishment of the TPS model, but also reduces the difficulty of correcting the off-target trajectory and further improves the accuracy of the off-target trajectory.

[0055] 4. By performing nonlinear correction on the PSD data, this invention can effectively improve the accuracy of the trajectory on the first target, which is beneficial for further improving the accuracy of the trajectory outside the target through the TPS model.

[0056] 5. By dividing the entire shooting training cycle into three segments and assigning weights to the key parameters of each segment, this invention can establish a scientific and reasonable shooting training evaluation system, significantly improving the accuracy of shooting training. Attached Figure Description

[0057] The accompanying drawings, which are included to provide a further understanding of embodiments of the invention and form part of this application, do not constitute a limitation thereof. In the drawings:

[0058] Figure 1 This is a flowchart of a full-cycle measurement method according to a specific embodiment of the present invention;

[0059] Figure 2 The projection of data measured by the inertial measurement unit in a specific embodiment of the present invention onto a two-dimensional plane is shown;

[0060] Figure 3 This is a flowchart of another full-cycle measurement method in a specific embodiment of the present invention;

[0061] Figure 4 This is a schematic diagram of the structure of a measurement system according to a specific embodiment of the present invention;

[0062] Figure 5 This is a schematic diagram of the experimental platform used in a specific embodiment of the present invention;

[0063] Figure 6 This is a schematic diagram of the full-cycle shooting aiming point trajectory spliced ​​in a specific embodiment of the present invention. Detailed Implementation

[0064] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0065] In the description of this invention, it should be understood that the terms "front", "rear", "left", "right", "up", "down", "vertical", "horizontal", "high", "low", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limiting the scope of protection of this invention.

[0066]

Example 1

[0067] like Figure 1 The method for measuring the trajectory of a firing aiming point throughout its entire lifecycle, as shown, includes the following steps:

[0068] The positioning light signal from the target is collected by the position sensor installed on the firing device, and the first target trajectory in the area inside the target is obtained based on the positioning light signal.

[0069] The inertial measurement unit installed on the shooter collects the shooter's attitude signal, and obtains the second on-target trajectory in the inner region of the target and the outer trajectory in the outer region of the target based on the shooter's attitude signal.

[0070] Based on the first target trajectory, the second target trajectory, and the off-target trajectory, the full-cycle shooting aiming point trajectory is obtained. The first target trajectory and the second target trajectory are fused using an iterative nearest point algorithm to obtain a fused target trajectory. The fused target trajectory and the off-target trajectory are then spliced ​​together to obtain the full-cycle shooting aiming point trajectory.

[0071] In some embodiments, the positioning light signal from the target can be a light signal reflected by a corner reflector on the target. Specifically, an infrared light source emitter is set on the shooter, which projects a conical laser beam onto the target, illuminating a corner reflector placed on the target surface. The corner reflector reflects the laser beam back to the PSD. As long as the corner reflector is within the illumination range of the conical laser beam, the photodetector will acquire an aiming point. The photodetector measures the position change of the aiming point by measuring the reflected light spot, and the PSD sensor calculates the coordinates of the light spot center based on the photocurrent distribution. Finally, a first target trajectory is generated based on the position of each aiming point.

[0072] For example, if the position sensor uses a four-quadrant PSD sensor, the center coordinates (x, y) of the light spot can be obtained using the position calculation formula. and Solve for I. x1 I x2 I y1 I y2 This is the current signal output by the four-quadrant PSD sensor.

[0073] In some embodiments, the positioning light signal from the target can be a light signal directly emitted by an infrared light source emitter installed on the target. Specifically, when the shooter aims at the infrared light source emitter on the target, the infrared light source emits light through the light outlet, transmits it to the lens, and is converged onto the PSD. The PSD outputs the position of the infrared light source in the lens's field of view. When the aiming point moves, the measurement position also moves synchronously. After determining the relative positional relationship between the acquired infrared light source and the aiming point, the position of each aiming point can be obtained, generating the first target trajectory.

[0074] In one or more embodiments, to reduce ambient light interference, the wavelength of the infrared light source should avoid the bands that cause ambient light interference, and a narrowband filter of the corresponding wavelength should be provided at the front end of the PSD to reduce ambient light interference. In some embodiments, the wavelength of the infrared light source is 900~1500 nm. In one or more embodiments, when the wavelength of the infrared light source is 900~1000 nm, the PSD is a silicon-based device. In one or more embodiments, when the wavelength of the infrared light source is 1000~1500 nm, the PSD is an InGaS device.

[0075] In one or more embodiments, other light source emitters, such as lasers, may also be used.

[0076] In one or more embodiments, a filter and a lens are also provided between the position sensor and the muzzle.

[0077] In some embodiments, such as Figure 2As shown, by acquiring data such as the x, y, and z-axis rotation angles in space output by the inertial sensor, the shooter's attitude signal, such as the muzzle motion attitude angle, is calculated. This muzzle motion attitude angle is used as the pointing angle of the muzzle, and then, based on the shooting distance, trigonometric functions are used to finally obtain the estimated trajectory of the aiming point. The coordinates of n are calculated based on the angle β between the z-axis and the XOY plane, and the coordinates of m are calculated based on the angle γ between the y-axis and the XOZ plane. The three-dimensional vector is mapped to a two-dimensional vector to obtain the aiming point coordinates. The trajectory coordinates are displayed in real time based on β and γ. Since α has no effect on the attitude trajectory change, β and γ are used to obtain the trajectory on the second target and the trajectory outside the target.

[0078] In one or more embodiments, the shooting distance can be manually determined and used as a calculation parameter input, or a laser rangefinder can be installed on the shooter to measure and record the distance during the initialization phase.

[0079] In some preferred embodiments, after the position sensor measures the position of the positioning light signal spot, the signal is sampled by a processing unit, such as an MCU. Simultaneously, the MCU also samples the shooter attitude signal acquired by the inertial measurement unit (IMU). Synchronization is performed during sampling to ensure data alignment between the positioning light signal and the shooter attitude signal. When their sampling frequencies are inconsistent, the MCU interpolates the lower sampling rate signal to obtain aligned data. In one or more embodiments, the IMU's sampling frequency is not less than 100 Hz, and the PSD's sampling frequency is not less than 2 kHz.

[0080] In some preferred embodiments, the sum of the areas of the inner and outer regions of the target corresponding to the shooter attitude signal is 2 to 10 times the area of ​​the inner region of the target corresponding to the positioning light signal.

[0081] In some embodiments, the collected data is synchronized in time and space. In time synchronization, the MCU generates a 1PPS trigger signal to synchronize the IMU and PSD data, and linear interpolation is used to complete the time points with different sampling times. In spatial calibration, a world coordinate system is established with the center of the live target as the origin (X-axis horizontal to the right, Y-axis aiming direction, Z-axis vertical upward), and the IMU coordinate system is aligned with the coordinate system of the shooter (Y-axis in the same aiming direction).

[0082] In some preferred embodiments, for live-fire training, the bullet's descent time can be calculated based on the initial bullet velocity and the aiming distance. (t = distance / initial velocity). Taking a distance of 20 meters as an example, the Type 95 rifle (initial velocity 800 m / s) drops 2.27 mm at 20 meters, while the pistol (initial velocity 350 m / s) drops approximately 15 mm at 20 meters. When aiming, if the laser point is above the bullseye by a distance d, it indicates that the target has been aimed at the bullseye. Before firing, aim at the target. When the laser line coincides with the bullseye, the aiming direction is the center of the bullseye. Record the position (x0, y0) output by the PSD detector and the angle (a0, b0, c0) of the IMU. Calculate the aiming point position (x1, y1) on the target plane based on the firing distance. Record (x0, y0) and (x1, y1) for later trajectory point alignment and correction.

[0083]

Example 2

[0084] Based on Example 1, such as Figure 3 The method for measuring the full-cycle firing aiming point trajectory shown in the figure establishes an error correction model based on the first and second target trajectories and the off-target trajectory when fusing the first target trajectory and the second target trajectory. Based on the error correction model, the off-target trajectory is corrected, and the first target trajectory and the corrected off-target trajectory are spliced ​​together to obtain the full-cycle firing aiming point trajectory.

[0085] In this embodiment, after obtaining the error correction model, the error correction model is used to obtain the off-target trajectory based on IMU data, thereby significantly improving the accuracy of the off-target trajectory.

[0086] Specifically, in one or more embodiments, the trajectory on the first target is set as the control point: ,in (In the two-dimensional trajectory, d=2). The trajectory on the second target is the target point: ,in The TPS transformation includes affine and non-affine transformations. Affine transformations are applied globally to model global deformations (translation, rotation, scaling); non-affine transformations are applied locally to model local deformations, implemented through radial basis functions (RBFs). The transformation formula is:

[0087]

[0088] In the formula: A is a The affine transformation matrix, b is The translation vector; It is a weight vector (each control point corresponds to a weight); These are radial basis functions; for two-dimensional TPS, they are typically used. (when At that time, and The goal of TPS is to minimize the following energy function:

[0089]

[0090] In the formula: the first term is the fitting error, the second term is the bending energy (smoothness constraint), and λ is the regularization parameter (controlling smoothness; the larger λ is, the smoother the transformation, but the fitting accuracy may decrease). The transformation parameters (affine and non-affine weights) can be obtained by solving the linear equation system:

[0091]

[0092] In the formula, K is a The matrix, elements , It is A matrix, each row is ), It is a weight matrix , It is a combination of affine transformation matrix and translation vector. , The target point coordinates .

[0093] Based on the control points and target points, the transformation parameters in the TPS model can be obtained, thus deriving the TPS model. After obtaining the error correction model, it is applied to the off-target trajectory obtained based on IMU data, thereby significantly improving the accuracy of the off-target trajectory. Finally, after obtaining the corrected off-target trajectory, for the area inside the target, the first on-target trajectory with higher accuracy is selected and stitched with the corrected off-target model with improved accuracy for the area outside the target to obtain a full-cycle shooting aiming point trajectory. In some preferred embodiments, smooth spline interpolation can be used to stitch the on-target and off-target trajectories to ensure the continuity of the entire trajectory.

[0094]

Example 3

[0095] Based on the above embodiments, in order to reduce noise such as Gaussian white noise, random walk, and zero bias instability in the shooter attitude signal acquired by the IMU, the following steps are also included: constructing an IMU noise model, extending the IMU noise model into an attitude domain noise model, using maximum likelihood estimation or Kalman filtering algorithm to optimize the parameters of the attitude domain noise model, and denoising the shooter attitude signal based on the optimized attitude domain noise model.

[0096] In this embodiment, the noise in the Mems IMU mainly includes Gaussian white noise, random walk, and zero-bias instability. The classic Allan variance model is used to extend it into an attitude domain noise model. During optimization, the error between the high-precision attitude angle calculated by the PSD measurement system and the original attitude angle signal measured by the IMU is used to optimize the noise model parameters. Specifically, the PSD measurement attitude and IMU data can be synchronized using timestamps, and linear interpolation is used to interpolate the PSD measurement attitude to the IMU sampling time. Subsequently, an error vector is constructed for error calculation. Next, the parameter identification method of maximum likelihood estimation (MLE) or extended Kalman filtering (EKF) is used to optimize key parameters in the noise model, such as Gaussian white noise variance, random walk coefficients, and zero-bias time-varying model parameters. Finally, the root mean square of the error vector is minimized to ensure that the optimized model can accurately simulate the IMU noise characteristics.

[0097]

Example 4

[0098] Based on the above embodiments, a first target trajectory correction process is also included, comprising the following steps:

[0099] Establish a trajectory correction model for the first target;

[0100] Obtain the actual coordinates and test coordinates of the specified calibration point on the target;

[0101] Based on the actual coordinates and test coordinates, the coefficients in the trajectory correction model on the first target are fitted using the least squares method to obtain the fitted trajectory correction model.

[0102] The trajectory on the first target is corrected using the fitted trajectory correction model.

[0103] In one or more embodiments, nine calibration points are set on the target, and the nine calibration points are distributed in three rows and three columns with equal spacing of 10 cm.

[0104] In one or more embodiments, the first target trajectory correction model may consist of two quadratic polynomials:

[0105]

[0106]

[0107] Where a_0~a_5 and b_0~b_5 are the fitting coefficients, and the coefficients in the polynomial are fitted using the least squares method. The goodness of fit R0 is... 2 ≥0.998. During testing, the acquired raw positioning optical signal data is fed into the fitted trajectory correction model to obtain the positioning optical signal after PSD nonlinear correction.

[0108] In one or more embodiments, nonlinear correction of the PSD can also be performed using an improved neural network conjugate gradient algorithm, which uses the gradient of the objective function to gradually generate conjugate directions and uses the derivative method of the function to perform fast search and interpolation operations, so that it can maintain good linearity under high-precision measurement.

[0109] In one or more embodiments, the distortion of the PSD can also be corrected by a lookup table interpolation method to obtain the corrected spot position to determine the aiming point, thereby further improving the measurement accuracy of the PSD.

[0110]

Example 5

[0111] Based on the above embodiments, the following steps are also included:

[0112] Based on the first coordinates of each trajectory point in the first target trajectory and the second coordinates of each trajectory point in the second target trajectory, an initial rigidity transformation is set, which includes a rotation matrix R and a translation vector T.

[0113] Adjust the rotation matrix R and translation vector T so that the error E = To reach the minimum value, in the formula, A i Let B be the first coordinate. i Using the second coordinate, obtain the rotation matrix and translation vector corresponding to the minimum error;

[0114] Based on the rotation matrix and translation vector corresponding to the minimum error, the second coordinate is transformed to the coordinate system corresponding to the first coordinate.

[0115] In this embodiment, because the PSD and IMU upload the same number of aiming point trajectory coordinates per unit time, based on the one-to-one correspondence between corresponding points on the two trajectories on the target, an initial rigid transformation from the PSD coordinate system to the IMU coordinate system is estimated. A rotation matrix R and a translation vector T are found such that the second coordinate, after being transformed to the first coordinate system, is as close as possible to the first coordinate. That is, the error E is minimized. .

[0116] Calculate matrix H:

[0117]

[0118] Perform singular value decomposition (SVD) on matrix H:

[0119]

[0120] Calculate the rotation matrix R:

[0121]

[0122] In the formula, It is a determinant, used to ensure that the rotation matrix is ​​a rigid body rotation, i.e., without reflection.

[0123] Calculate the translation vector T:

[0124]

[0125] Finally, using the estimated R and T, all trajectory points measured by the IMU are transformed into the coordinate system of the PSD, and then a suitable nonlinear transformation model is established on this basis to finely correct the coordinates.

[0126]

Example 6

[0127] Based on the above embodiments, the shooting aiming point trajectory of the entire cycle is divided into three segments. The first segment is from the first time the gun is raised to aiming, the second segment is from aiming to firing, and the third segment is from firing to the second time the gun is raised.

[0128] Obtain the value of the first parameter of the first trajectory segment, the value of the second parameter of the second trajectory segment, and the value of the third parameter of the third trajectory segment. Based on the values ​​of the first parameter, the second parameter, and the third parameter, evaluate the quality of shooting training.

[0129] In one or more embodiments, the first parameter may be at least one of peak angular velocity, acceleration variance, and mean motion velocity; the second parameter may be trajectory length and / or trajectory divergence; and the third parameter may be at least one of angular change rate, trajectory length, and motion velocity.

[0130] In one or more embodiments, the quality of a continuous firing process is evaluated by constructing a weighted evaluation model, wherein a first weighting coefficient is assigned to the value of a first parameter, a second weighting coefficient is assigned to the value of a second parameter, and a third weighting coefficient is assigned to the value of a third parameter. In some preferred embodiments, the second weighting coefficient is greater than the third weighting coefficient, and the third weighting coefficient is greater than the first weighting coefficient.

[0131]

Example 7

[0132] A full-cycle measurement system for the trajectory of a firing aiming point, characterized in that it comprises:

[0133] A position sensor, mounted on the firing mechanism, is used to collect positioning light signals from the target;

[0134] An inertial measurement unit, mounted on the firing mechanism, is used to acquire the firing mechanism's attitude signals;

[0135] The processing unit is used to transmit the positioning optical signal and the shooter attitude signal to the analysis unit;

[0136] The analysis unit obtains a first target trajectory within the target area based on the positioning light signal, and a second target trajectory within the target area and an external target trajectory outside the target area based on the shooter attitude signal. It then uses an iterative nearest-point algorithm to fuse the first and second target trajectories to obtain a fused target trajectory. Finally, it stitches the fused target trajectory and the external target trajectory together to obtain the full-cycle aiming point trajectory.

[0137] Based on the first target trajectory and the second target trajectory, an error correction model is established. Based on the error correction model, the off-target trajectory is corrected. The first target trajectory and the corrected off-target trajectory are then spliced ​​together to obtain the full-cycle shooting aiming point trajectory.

[0138] In one or more embodiments, the inertial measurement unit is a three-axis gyroscope and a six-axis sensor with a three-axis accelerometer.

[0139] In one or more embodiments, the processing unit is disposed on the shooter to acquire the shooter attitude signal and positioning light signal in real time and synchronously.

[0140] In one or more embodiments, the processing unit (MCU), position sensor (PSD), and inertial measurement unit (IMU) are integrated together to form a single circuit board as the aiming point trajectory measurement module. In some embodiments, the aiming point trajectory measurement module can be mounted on the barrel of the firearm via a Picatinny rail or similar mounting surface.

[0141] In some preferred embodiments, the inertial measurement unit is an MPU6050, with a triaxial angular velocity range of ±2000° / s (accuracy ±0.01° / s), a triaxial accelerometer range of ±16g (accuracy ±0.001g), and a sampling frequency of 1 kHz. The position sensor has an infrared response wavelength of 850nm, an effective detection area of ​​9mm×9mm, a resolution of 10μm, and a sampling frequency of 1000Hz. More preferably, it also includes a 50mm focal length convex lens (aperture F / 2.0, distortion rate < 1%); a narrowband filter with a wavelength of 850nm and a bandwidth of ±10nm. The processing unit uses an STM32H743 MCU, supporting 1PPS hardware synchronous triggering and Bluetooth 5.0 transmission.

[0142] In some embodiments, a live-fire target and an infrared light source target can be set up. The live-fire target is used for live-fire firing by the gunner, and the infrared light source target is set up next to the live-fire target to send positioning light signals to the PSD. By establishing a translational mapping relationship between the coordinates of the infrared light source target and the live-fire target, the positioning light signal is translated to obtain the first target trajectory on the live-fire target.

[0143] In some preferred embodiments, the live target is a standard 10-ring target with a diameter of 100 cm. The infrared light source target has a built-in infrared light source and is covered with a 10 cm thick rubber adsorption layer with a Shore hardness of 60±5. The distance between the target and the live target is 100 cm.

[0144] In some preferred embodiments, an evaluation unit is also included, which divides the entire shooting aiming point trajectory into three segments: the first segment is from the first time the gun is raised to aiming, the second segment is from aiming to firing, and the third segment is from firing to the second time the gun is raised; the evaluation unit obtains the value of the first parameter of the first segment, the value of the second parameter of the second segment, and the value of the third parameter of the third segment; and evaluates the shooting training quality based on the values ​​of the first parameter, the second parameter, and the third parameter.

[0145]

Example 8

[0146] To verify the effectiveness of the aforementioned measurement method and system in extracting the aiming point trajectory during shooting training, a system was built as follows: Figure 5 The experimental platform shown mainly consists of an industrial camera, a hardware platform, an infrared light pen, a light curtain, and a host computer. Its purpose is to analyze the accuracy of the aiming point trajectory measured by this hardware platform in the upper and lower parts of the target. The hardware platform is zero-point calibrated on the optical platform (the origin of the trajectory acquired by the PSD sensor and IMU sensor coincides with the center point of the actual target's 10 rings). The full-path target model is defined as 200% of the maximum trajectory range that the PSD can acquire, representing the full-path range of the aiming point trajectory.

[0147] 1) Target trajectory accuracy test. Trajectory accuracy was tested at a typical firing distance of 25m. The hardware platform was fixed to a displacement platform, and the platform was precisely moved 100mm in both the horizontal and vertical directions. The coordinate movement measured by the system was recorded. Statistically significant measurement data were obtained through 20 repeated tests. The results of the 20 tests at typical distances are shown in Table 1.

[0148] Table 1:

[0149]

[0150] As shown in the table, the average SD (standard deviation) of the visual measurements (experimental true value) at a distance of 25m is 1.05mm, and the average SD of the system measurements is 2.14mm. Based on the timestamps, the same number of system measurement points and visual measurement points N are taken, and their RMSE (root mean square error) is calculated. Therefore, each set of data yields an RMSE between the measured value and the experimental true value. The RMSE is the square root of the mean of the sum of the squares of the Euclidean distances between all corresponding points. Finally, the average RMSE value is calculated as the measurement error, which is 2.85mm. It can be seen that the target trajectory measurement has excellent measurement accuracy and can meet the accuracy requirements of aiming point trajectory measurement in shooting training.

[0151] 2) Off-target trajectory accuracy test. In actual testing, the greater the deviation of the aiming point from the target, the greater the non-linear increase in measurement error during off-target trajectory measurement. Therefore, the off-target portion is divided into 125%, 150%, 175%, and 200% of the experimental target radius. The range greater than 200% of the target radius is called the far-target portion, to facilitate evaluation of the correlation between the measured trajectory and the true trajectory of the visual measurement system. Off-target trajectory accuracy tests were conducted at a typical shooting distance of 25m. The hardware platform was fixed to a displacement platform, and the platform was moved horizontally / vertically 330mm from the edge of the target surface (within the range of 100%-210% of the target radius). The coordinate movement measured by the system was recorded. Statistically significant measurement data were obtained through 20 repeated tests at typical distances. The results of the 20 off-target tests at a typical distance of 25m are shown in Table 2.

[0152] Table 2:

[0153]

[0154] Using the same error calculation method as on the target, the mean SD (standard deviation) of the visual measurement (experimental true value) was calculated to be 1.11 mm, while the mean SD of the system measurement was 4.42 mm. The mean RMSE of the system in a typical off-target area was 5.86 mm. This demonstrates that the corrected off-target system measurement path has a good correlation with the visual measurement aiming point path, meeting the needs of shooting training for reconstructing and predicting the trajectory of aiming points that have deviated from the target.

[0155] 3) Off-target trajectory correction results and correlation analysis. To better observe the experimental results, the trajectory correction effect was evaluated using two key regression indicators based on the Euclidean distance between the system-measured trajectory data and the visually measured trajectory: RMSE (root mean square error of the Euclidean distance between trajectories) and PCC (Pearson correlation coefficient). Tables 3 and 4 show the RMSE, SD, and PCC indices of the off-target measured trajectory before and after TPS algorithm correction compared to the true visual measurement.

[0156] Table 3: Off-target trajectory indices before correction

[0157]

[0158] Table 4: Corrected Off-Target Trajectory Indicators

[0159]

[0160] Test results show that, overall, the average root mean square error (RMSE) decreased significantly from 15.14 mm to 4.42 mm, an improvement of 70.8%; simultaneously, the correlation coefficient (PCC) increased from 0.681 to 0.825, indicating a significant enhancement in data correlation. Looking at different regions, the improvement was more significant with increasing distance: the RMSE decreased by 39.2% in the closest region (100%-125%), while it decreased by a staggering 78.5% in the farthest region (175%-200%). This demonstrates that the algorithm not only significantly reduced systematic bias but also greatly improved the consistency and stability of measurements.

[0161] 4) Trajectory stitching. Finally, the corrected off-target trajectory is stitched with the trajectory on the first target to form the aiming point trajectory for the entire firing cycle, such as... Figure 6 As shown, through accuracy testing at typical distances and evaluation of off-target trajectory fitting, the system demonstrates good accuracy and stability in target trajectory measurement, and also exhibits excellent effectiveness in reconstructing and predicting the off-target aiming point trajectory.

[0162] The terms "first," "second," etc., used in this invention (e.g., first target trajectory, second target trajectory, etc.) are merely for clarity of description and are not intended to limit any order or emphasize importance. Furthermore, the term "connection" used in this invention, unless otherwise specified, can refer to a direct connection or an indirect connection via other components.

[0163] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for measuring the trajectory of a firing aiming point throughout its entire lifecycle, characterized in that, Includes the following steps: The positioning light signal from the target is collected by the position sensor installed on the firing device, and the first target trajectory in the area inside the target is obtained based on the positioning light signal. The inertial measurement unit installed on the shooter collects the shooter's attitude signal, and obtains the second on-target trajectory in the inner region of the target and the outer trajectory in the outer region of the target based on the shooter's attitude signal. Based on the trajectory on the first target, the trajectory on the second target, and the trajectory outside the target, the firing aiming point trajectory for the entire cycle is obtained.

2. The method for measuring the trajectory of a firing aiming point throughout its entire lifecycle according to claim 1, characterized in that, The first target trajectory and the second target trajectory are fused using the iterative nearest point algorithm to obtain the fused target trajectory. The fused target trajectory and the off-target trajectory are then spliced ​​together to obtain the full-cycle shooting aiming point trajectory.

3. The method for measuring the trajectory of a firing aiming point throughout its entire lifecycle according to claim 1, characterized in that, Based on the first target trajectory and the second target trajectory, an error correction model is established. Based on the error correction model, the off-target trajectory is corrected. The first target trajectory and the corrected off-target trajectory are then spliced ​​together to obtain the full-cycle shooting aiming point trajectory.

4. A method for measuring the full-cycle trajectory of a firing aiming point according to any one of claims 1 to 3, characterized in that, It also includes the following steps: An IMU noise model is constructed and extended to an attitude domain noise model. The parameters of the attitude domain noise model are optimized using maximum likelihood estimation or Kalman filtering algorithm. The shotgun attitude signal is then denoised based on the optimized attitude domain noise model.

5. The method for measuring the trajectory of a firing aiming point throughout its entire lifecycle according to claim 4, characterized in that, It also includes a trajectory correction process on the first target, comprising the following steps: Establish a trajectory correction model for the first target; Obtain the actual coordinates and test coordinates of the specified calibration point on the target; Based on the actual coordinates and test coordinates, the coefficients in the trajectory correction model on the first target are fitted using the least squares method to obtain the fitted trajectory correction model. The trajectory on the first target is corrected using the fitted trajectory correction model.

6. The method for measuring the trajectory of a firing aiming point throughout its entire lifecycle according to claim 4, characterized in that, It also includes the following steps: Based on the first coordinates of each trajectory point in the first target trajectory and the second coordinates of each trajectory point in the second target trajectory, an initial rigidity transformation is set, which includes a rotation matrix R and a translation vector T. Adjust the rotation matrix R and translation vector T so that the error E = To reach the minimum value, in the formula, A i Let B be the first coordinate. i Using the second coordinate, obtain the rotation matrix and translation vector corresponding to the minimum error; Based on the rotation matrix and translation vector corresponding to the minimum error, the second coordinate is transformed to the coordinate system corresponding to the first coordinate.

7. The method for measuring the trajectory of a firing aiming point throughout its entire lifecycle according to claim 1, characterized in that, The entire firing aiming trajectory is divided into three segments: the first segment is from the first time the gun is raised to aiming, the second segment is from aiming to firing, and the third segment is from firing to the second time the gun is raised. Obtain the value of the first parameter of the first trajectory segment, the value of the second parameter of the second trajectory segment, and the value of the third parameter of the third trajectory segment. Based on the values ​​of the first parameter, the second parameter, and the third parameter, evaluate the quality of shooting training.

8. The method for measuring the trajectory of a firing aiming point throughout its entire lifecycle according to claim 1, characterized in that, The sum of the areas of the inner and outer regions of the target corresponding to the shooting device attitude signal is 2 to 10 times the area of ​​the inner region of the target corresponding to the positioning light signal.

9. A full-cycle measurement system for the trajectory of a firing aiming point, characterized in that, include: A position sensor, mounted on the firing mechanism, is used to collect positioning light signals from the target; An inertial measurement unit, mounted on the firing mechanism, is used to acquire the firing mechanism's attitude signals; The processing unit is used to transmit the positioning optical signal and the shooter attitude signal to the analysis unit; The analysis unit obtains a first target trajectory within the target area based on the positioning light signal, and a second target trajectory within the target area and an external target trajectory outside the target area based on the shooter attitude signal. It then uses an iterative nearest-point algorithm to fuse the first and second target trajectories to obtain a fused target trajectory. Finally, it stitches the fused target trajectory and the external target trajectory together to obtain the full-cycle aiming point trajectory. Based on the first target trajectory and the second target trajectory, an error correction model is established. Based on the error correction model, the off-target trajectory is corrected. The first target trajectory and the corrected off-target trajectory are then spliced ​​together to obtain the full-cycle shooting aiming point trajectory.

10. A full-cycle measurement system for the trajectory of a firing aiming point according to claim 9, characterized in that, It also includes an evaluation unit, which divides the entire shooting aiming trajectory into three segments: the first segment is from the first time the gun is raised to aiming, the second segment is from aiming to firing, and the third segment is from firing to the second time the gun is raised; it obtains the value of the first parameter of the first segment, the value of the second parameter of the second segment, and the value of the third parameter of the third segment, and evaluates the quality of shooting training based on the values ​​of the first parameter, the second parameter, and the third parameter.

Images