Gun dynamic monitoring system, method and equipment based on trajectory positioning

Abstract

The present invention discloses a gun dynamic monitoring system, method and equipment based on trajectory positioning, which relates to the technical field of gun positioning and monitoring, and comprises the following steps: establishing an indoor gun positioning method, positioning the gun indoors, and performing indoor dynamic monitoring; establishing outdoor calibration points, obtaining outdoor gun position information and outdoor gun related information, and performing error correction processing to obtain corrected position information; constructing a gun motion prediction model, and performing prediction to obtain gun motion prediction information; monitoring the gun position based on the gun motion prediction information and the corrected position information; the present invention is used to solve the problem that when the existing gun positioning and monitoring technology monitors the gun through positioning, there are positioning errors caused by the limitations of the positioning device itself and the environmental interference, and the gun monitoring is performed by only relying on simple position information and limited positioning trajectories to perform position prediction, and it is impossible to accurately make early warnings and interventions.

Description

Technical Field

[0001] The present invention relates to the technical field of gun positioning and supervision, and in particular to a gun dynamic supervision system, method and equipment based on trajectory positioning. Background Art

[0002] Gun positioning and supervision technology refers to a comprehensive supervision method that uses advanced technical means, such as satellite positioning technology, wireless communication technology, and sensor technology, to monitor, track, and manage the location status of guns in real time.

[0003] Existing gun positioning and supervision technologies will have certain errors when locating guns through positioning devices, which is due to the hardware limitations of the equipment; for example, the positioning accuracy of ordinary GPS positioning devices is between several meters and more than ten meters; and the interference of the external environment on the positioning signal will increase, resulting in a decrease in positioning accuracy; however, existing gun positioning and supervision technologies often ignore the problem of insufficient accuracy of original positioning data. For example, in the patent application with publication number CN112866925A, an intelligent gun management system and intelligent gun management equipment based on UWB positioning are disclosed. In the process of locating guns through UWB, this solution does not take into account the errors existing when the UWB device locates the gun, which will cause the position calculated by the positioning device to be far from the actual position, thereby resulting in low accuracy of gun positioning and supervision; and existing gun positioning and supervision technologies often only monitor the current position of the gun, focusing on real-time tracking and post-analysis, and lack of gun positioning and supervision. Forward-looking prediction of position; this makes it impossible for the regulatory system to issue early warnings and interventions when facing potential risks, such as when a gun is about to enter a dangerous area. Measures are often taken only after a violation has occurred, and the regulatory system is not proactive enough; or when regulating by predicting the position, it may only rely on simple position information and limited historical trajectories; for example, the future position is only inferred based on the connection between several past position points, without fully considering the changes in the gun's motion state, such as speed, acceleration and direction. When the gun holder changes his motion state, such as sudden acceleration, deceleration and turning, the prediction result will deviate greatly from the actual situation. Therefore, when the existing gun positioning and regulatory technology regulates guns through positioning, there are errors in positioning due to the limitations of the positioning equipment itself and environmental interference. At the same time, gun regulation based on position prediction based on simple position information and limited positioning trajectories cannot accurately issue early warnings and interventions in advance. Summary of the invention

[0004] The present invention aims to solve at least one of the technical problems in the prior art to a certain extent, by establishing an indoor gun positioning method for positioning and performing indoor dynamic supervision; establishing outdoor calibration points, obtaining outdoor gun position information and outdoor gun-related information, and performing error correction processing to obtain corrected position information; constructing a gun motion prediction model, and performing prediction to obtain gun motion prediction information; supervising the gun position based on the gun motion prediction information and the corrected position information; so as to solve the problem that when the existing gun positioning and supervision technology supervises the gun through positioning, there are positioning errors caused by the limitations of the positioning device itself and environmental interference, and the gun supervision is carried out by only relying on simple position information and limited positioning trajectories to perform position prediction, and it is impossible to accurately make early warnings and interventions.

[0005] To achieve the above-mentioned purpose, in a first aspect, the present application provides a dynamic gun supervision system based on trajectory positioning, including an indoor supervision module, an outdoor acquisition module, a motion prediction module and a dynamic supervision module;

[0006] The indoor monitoring module includes an indoor positioning unit and an indoor management unit. The indoor positioning unit establishes an indoor gun positioning method based on ultra-wideband technology to locate indoor guns. The indoor management unit is used to dynamically monitor indoor guns.

[0007] The outdoor acquisition module is used to establish outdoor calibration points, obtain outdoor gun position information and outdoor gun related information, and perform error correction processing on the outdoor gun position information to obtain corrected position information;

[0008] The motion prediction module includes a model building unit and a position prediction unit. The model building unit is used to build a gun motion prediction model. The position prediction unit predicts the gun motion by using the corrected position information and outdoor gun related information to obtain gun motion prediction information.

[0009] The dynamic supervision module supervises the gun position based on the gun movement prediction information and the corrected position information.

[0010] Furthermore, the indoor positioning unit is configured with an indoor positioning strategy, and the indoor positioning strategy includes:

[0011] Establish an indoor gun positioning method: arrange multiple ultra-wideband base stations indoors, select one of the ultra-wideband base stations and mark it as the origin base station, establish a spatial rectangular coordinate system with the position of the origin base station as the coordinate origin, mark it as the indoor spatial coordinate system, obtain the spatial positions of all ultra-wideband base stations in the indoor spatial coordinate system, and record it as the base station position information;

[0012] The spatial distance between the ultra-wideband base station and the gun is obtained through the corresponding sensors on the ultra-wideband base station and the gun, and the four ultra-wideband base stations with the closest spatial distance are selected and marked as positioning base stations, which are denoted as C1, C2, C3 and C4 respectively, and the spatial distances between the positioning base station and the gun are denoted as d1, d2, d3 and d4 in order; then the spatial positions of the positioning base stations are obtained based on the base station position information, which are denoted as C1(x1, y1, z1), C2(x2, y2, z2), C3(x3, y3, z3) and C4(x4, y4, z4) respectively.

[0013] Furthermore, the indoor positioning unit is configured with an indoor positioning strategy, and the indoor positioning strategy further includes:

[0014] Based on the spatial positions of the positioning base stations, the spatial distance between any two positioning base stations is calculated and recorded as the standard base station distance;

[0015] For any positioning base station, the spatial distance with other positioning base stations is obtained through ultra-wideband technology, which is recorded as the measuring base station distance; then based on the standard base station distance, the error between the measuring base station distance and the corresponding standard base station distance is calculated, and the unit distance error is calculated according to the corresponding measuring base station distance, and the average is obtained, which is marked as the average unit distance error; the average unit distance error of all positioning base stations is obtained repeatedly, which is marked as the unit error information;

[0016] Based on the unit error information, d1, d2, d3 and d4 are corrected respectively. After completion, they are recorded as qd1, qd2, qd3 and qd4 in order, marked as the actual gun distance;

[0017] The gun position equations are established based on the spatial position of the positioning base station and the actual gun distance, and the gun's spatial position (x0, y0, z0) is calculated and solved. The gun position equations are as follows: ;

[0018] Based on the indoor gun positioning method, the spatial position of the indoor gun is obtained at a first time interval, marked as indoor gun position information, and the first time interval is t.

[0019] Furthermore, the indoor management unit is configured with an indoor management strategy, and the indoor management strategy includes:

[0020] Set a first indoor space range and a second indoor space range, the first indoor space range∈the second indoor space range;

[0021] Determine whether the gun is located within the first indoor space range and the second indoor space range based on the indoor gun position information;

[0022] If the gun is neither within the first indoor space nor within the second indoor space, an alarm signal is triggered;

[0023] If the gun is located in the second indoor space but not in the first indoor space, a warning signal is triggered, and the duration T1 of the gun being located in the second indoor space but not in the first indoor space is counted, and it is determined whether T1 is greater than T0. If T1 is greater than T0, an alarm signal is triggered; if T1 is greater than T0, no alarm signal is triggered, and T0 is the alarm time threshold;

[0024] If the gun is within the first indoor space, the warning signal is cancelled.

[0025] Furthermore, the outdoor data collection module is configured with a data collection correction strategy, which includes:

[0026] Establish multiple calibration points at a first distance interval on the path of the gun, and record the longitude and latitude of the calibration points as the reference longitude and latitude, and the first distance interval is LJ;

[0027] When the gun is outdoors, the latitude and longitude, speed, acceleration magnitude and direction angle of the gun are obtained at a second time interval through the satellite navigation system, and the time of acquisition is recorded, the latitude and longitude of the gun are marked as outdoor gun position information, and the speed, acceleration magnitude and direction angle of the gun are marked as outdoor gun related information, and the latitude and longitude of the two calibration points closest to the gun are obtained at the same time, recorded as actual latitude and longitude, and the second time interval is r;

[0028] Based on the actual longitude and latitude of the two calibration points and the corresponding reference longitude and longitude, the difference in longitude and latitude between the two calibration points are calculated respectively, and the average value of the difference in longitude and latitude is obtained respectively, which is marked as the average longitude and longitude error. Based on the average longitude and longitude error, the outdoor gun position information of the gun at the corresponding moment is corrected, and the corrected position information is obtained after completion.

[0029] Furthermore, the model building unit is configured with a model building strategy, and the model building strategy includes:

[0030] The first prediction model is constructed using the long short-term memory network. The first prediction model includes: an input layer, an LSTM layer, and an output layer. The input shape of the input layer is set to (A1, A2), the number of LSTM layers is A3, the number of units in the LSTM layer is A4, and the number of neurons in the output layer is A5. The loss function of the model training process is defined as follows: , where MS represents the loss value, Ui represents the actual longitude and latitude values, Vi represents the longitude and latitude values ​​predicted by the model, and n is the number of samples input to the model;

[0031] The corrected position information at the corresponding moment and the outdoor gun related information are combined to obtain the gun motion data; the gun motion data are normalized according to the data type, and all data sizes are scaled to [0, 1]; after completion, the normalized motion data is obtained; the normalized motion data is divided into a first training set and a first test set in a ratio of 8:2;

[0032] The first prediction model is trained using the first training set, and the second prediction model is obtained after completion; the second prediction model is tested using the first test set, and the mean absolute error MAE of the second prediction model is calculated. The mean absolute error calculation formula is as follows: , where MAE represents mean absolute error, Uk represents the actual longitude and latitude values, Vk represents the longitude and latitude values ​​predicted by the model, and K is the number of samples input to the model; determine whether MAE satisfies MAE≤MA0, where MA0 is the qualified threshold; if so, mark the second prediction model as a gun motion prediction model; if not, train the second prediction model again with the first training set, and then test it again with the first test set until a gun motion prediction model is obtained.

[0033] Furthermore, the location prediction unit is configured with a location prediction strategy, and the location prediction strategy includes:

[0034] Set the data sliding window size to the first time step, the sliding step to E0, and the first time step to E1;

[0035] Based on the normalized motion data, the data of the first time step is intercepted by using the data sliding window for the first time, and the first gun motion data of the future E0 moments is obtained by inputting into the gun motion prediction model, which is recorded as the first prediction data;

[0036] Slide the data sliding window again for the second time, intercept the data of the first time step again, including the first predicted data, and input the first gun motion data of the future E0 moments obtained by the gun motion prediction model, which is recorded as the second predicted data;

[0037] The sliding data sliding window is used to repeatedly intercept data and input into the gun motion prediction model until the first gun motion data of the second time step in the future is obtained, and the longitude and latitude data in the first gun motion data are extracted and recorded as gun motion prediction information. The second time step is E2.

[0038] Furthermore, the dynamic supervision module is configured with a dynamic supervision strategy, which includes:

[0039] Based on the corrected position information and the reference longitude and latitude of the two calibration points closest to the gun at the corresponding moment, calculate the vertical distance from the gun position to the line connecting the two calibration points closest to the gun, and mark it as the current gun warning distance, recorded as L0;

[0040] Based on the firearm movement prediction information and the reference longitude and latitude of the two calibration points closest to the firearm at the corresponding moment, calculate the perpendicular distance from the firearm position at all future moments to the line connecting the two calibration points closest to the firearm, which is marked as the future firearm warning distance information and denoted as LX;

[0041] Set a first warning distance and a second warning distance. The first warning distance is L1, and the second warning distance is L2, where L1 < L2;

[0042] If L0 > L2, then trigger an alarm signal;

[0043] If L1 < L0 < L2 and all LX values are less than L2, then trigger a warning signal;

[0044] If L1 < L0 < L2 and not all LX values are less than L2, then trigger an alarm signal;

[0045] If the duration of L1 < L0 < L2 is greater than T0, then trigger an alarm signal;

[0046] If L0 < L1 and not all LX values are less than L1 but all are less than L2, then trigger a warning signal;

[0047] If L0 < L1 and not all LX values are less than L2, then trigger an alarm signal.

[0048] In a second aspect, the present application provides a firearm dynamic supervision method based on trajectory positioning, including the following steps:

[0049] Establish an indoor firearm positioning method based on ultra-wideband technology to position the firearms indoors and conduct indoor dynamic supervision;

[0050] Establish calibration points outdoors, obtain outdoor firearm position information and outdoor firearm-related information, and perform error correction processing on the outdoor firearm position information to obtain corrected position information;

[0051] Construct a firearm movement prediction model, and use the corrected position information and outdoor firearm-related information to predict the firearm movement to obtain firearm movement prediction information;

[0052] Supervise the firearm position based on the firearm movement prediction information and the corrected position information.

[0053] In a third aspect, the present application provides an electronic device, including a processor and a memory. The memory stores computer-readable instructions. When the computer-readable instructions are executed by the processor, the steps in the above method are run.

[0054] Beneficial effects of the present invention: The present invention establishes an indoor gun positioning method based on ultra-wideband technology, positions indoor guns, and performs indoor dynamic supervision; establishes outdoor calibration points, obtains outdoor gun position information and outdoor gun-related information, and performs error correction processing on the outdoor gun position information to obtain corrected position information; constructs a gun movement prediction model, predicts gun movement using the corrected position information and outdoor gun-related information to obtain gun movement prediction information; supervises the gun position based on the gun movement prediction information and the corrected position information; ensures the accuracy of positioning by correcting the positioning error, and predicts the future gun position by using multiple parameters, so that gun supervision can accurately make early warnings and interventions;

[0055] The present invention performs error correction on the acquired spatial position of the gun through the position of the ultra-wideband base station itself, and performs error correction on the acquired latitude and longitude of the gun based on the calibration points. The advantages of the present invention are that the accuracy of the gun position data can be effectively improved, and the problems of insufficient accuracy and poor stability of the original positioning data in the gun positioning technology are solved, and the misjudgment caused by positioning errors is avoided, and reliable reference data is provided for accurate supervision; the future gun position is predicted by using multiple parameters such as longitude and latitude, speed, acceleration and direction, which can more comprehensively describe the movement state of the gun; the position change of the gun can be predicted more accurately, and through the accurate prediction of the future position, potential risk situations can be discovered in advance, so that there is enough time to take measures to intervene, and the occurrence of safety accidents can be effectively prevented. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 is a functional block diagram of the system of the present invention;

[0057] Figure 2 is a flow chart of the steps of the method of the present invention;

[0058] Figure 3 is a schematic diagram of error correction processing of the present invention;

[0059] Figure 4 This is a schematic diagram of the sliding window of data of the present invention;

[0060] Figure 5 It is a schematic structural diagram of the electronic device of the present invention. DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] Example 1, please refer to Figure 1 As shown, the present application provides a dynamic gun supervision system based on trajectory positioning, including an indoor supervision module, an outdoor acquisition module, a motion prediction module and a dynamic supervision module; in the actual application process, the outdoor acquisition module of the present system can be used to collect the existing escort system, obtain the positioning information of the gun through the satellite navigation system, and transmit the information to the existing intelligent escort management platform to achieve remote monitoring and the purpose of gun detachment from tracking, and can form a closed loop of supervision with the indoor supervision module to achieve the supervision of guns in the warehouse and on the way. The intelligent escort management platform can access all links involving guns, including warehouse storage, supervision during transportation and final delivery, to achieve full supervision from "in the warehouse" to "on the way". Whether it is the storage or transportation of guns, the system can provide seamless real-time monitoring and dynamic supervision to ensure the safety of guns in the warehouse and on the way, and can monitor, collaborate and remotely control in real time to improve the response speed and safety of gun management. Based on the motion prediction model, it can effectively predict the movement trajectory of the gun. Once an abnormality occurs in the gun, the system can quickly identify and start the emergency response mechanism to ensure timely disposal.

[0063] The indoor monitoring module includes an indoor positioning unit and an indoor management unit. The indoor positioning unit establishes an indoor gun positioning method based on ultra-wideband technology to locate indoor guns. The indoor management unit is used to dynamically monitor indoor guns.

[0064] The indoor positioning unit is configured with indoor positioning strategies, which include:

[0065] Establish an indoor gun positioning method: arrange multiple ultra-wideband base stations indoors, select one of the ultra-wideband base stations and mark it as the origin base station, establish a spatial rectangular coordinate system with the position of the origin base station as the coordinate origin, mark it as the indoor spatial coordinate system, obtain the spatial positions of all ultra-wideband base stations in the indoor spatial coordinate system, and record it as the base station position information; the acquisition should be as accurate as possible, because the base station position information is the reference data for subsequent error correction;

[0066] The spatial distance between the ultra-wideband base station and the gun is obtained through the ultra-wideband base station and the corresponding sensors on the gun, and the four ultra-wideband base stations with the closest spatial distance are selected. When locating the gun, it is necessary to know the specific position of the gun in the three-dimensional space including the spatial height, so at least four ultra-wideband base stations are required; if only the specific position of the gun in the two-dimensional space excluding the spatial height is required, three ultra-wideband base stations are required; marked as positioning base stations, respectively denoted as C1, C2, C3 and C4, and the spatial distance between the positioning base station and the gun is denoted as d1, d2, d3 and d4 in order; then the spatial position of the positioning base station is obtained based on the base station position information, which are denoted as C1(x1, y1, z1), C2(x2, y2, z2), C3(x3, y3, z3) and C4(x4, y4, z4) respectively;

[0067] Based on the spatial position of the positioning base station, the spatial distance between any two positioning base stations is calculated and recorded as the standard base station distance; for example, the spatial positions of C1 and C2 are C1(x1, y1, z1) and C2(x2, y2, z2), respectively, then the spatial distance from C1 to C2 is √[(x1-x2) 2 +(y1-y2) 2 +(z1-z2) 2 ];

[0068] For any positioning base station, the spatial distance with other positioning base stations is obtained through ultra-wideband technology, which is recorded as the measuring base station distance; that is, it is obtained by one base station transmitting a signal to another base station, and even if it is the same two base stations, the base stations transmitting the signals are different, and the measuring base station distances are also different. For example, C1 and C2, the distance obtained by C1 transmitting a signal to C2 is the measuring base station distance from C1 to C2, and the distance obtained by C2 transmitting a signal to C1 is the measuring base station distance from C2 to C1; the distances between the two measuring base stations are not the same, because the error of each base station transmitting the signal is different, while the standard base station distances between C1 and C2 are the same, because the coordinates of C1 and C2 are unchanged;

[0069] Based on the standard base station distance, the error between the measured base station distance and the corresponding standard base station distance is calculated, and the unit distance error is calculated according to the corresponding measured base station distance, and the average is obtained, which is marked as the average unit distance error; the average unit distance error of all positioning base stations is obtained repeatedly, which is marked as the unit error information; for example, the standard base station distances from C1 to C2, C1 to C3, and C1 to C4 are 12.5m, 15.6m, and 20.1m respectively, and the standard base station distances from C1 to C2, C1 to C3, and C1 to C4 are 12.5m, 15.6m, and 20.1m respectively. The measured base station distances from C1 to C3 and from C1 to C4 are 12.8m, 16.0m and 20.6m respectively. The unit distance errors of C1 are (12.8-12.5) / 12.5=0.024, (16.0-15.6) / 15.6=0.026, (20.6-20.1) / 20.1=0.025 respectively. The average unit distance error = (0.024+0.029+0.025) / 3=0.025.

[0070] Based on the unit error information, d1, d2, d3 and d4 are corrected respectively. After completion, they are recorded as qd1, qd2, qd3 and qd4 in order, and marked as the actual gun distance; for example, the average unit distance error of C1 is 0.025, and the distance from C1 to the gun is d1=14.2m, then qd1=d1 / (1+0.025)=13.9m;

[0071] The gun position equations are established based on the spatial position of the positioning base station and the actual gun distance, and the gun's spatial position (x0, y0, z0) is calculated and solved. The gun position equations are as follows: ;

[0072] Based on the indoor gun positioning method, the spatial position of the indoor gun is obtained at a first time interval, marked as indoor gun position information, and the first time interval is t; in this embodiment, the first time interval is 3s.

[0073] The indoor management unit is configured with indoor management strategies, which include:

[0074] Set a first indoor space range and a second indoor space range, where the first indoor space range∈the second indoor space range; for example, the entire indoor space can be set as the second indoor space range, and the indoor space 50 cm away from the indoor wall can be set as the first indoor space range. The specific setting can be based on the actual application scenario, and multiple different space ranges can also be divided;

[0075] Determine whether the gun is located within the first indoor space range and the second indoor space range based on the indoor gun position information;

[0076] If the gun is neither within the first indoor space nor within the second indoor space, an alarm signal is triggered;

[0077] If the gun is located in the second indoor space but not in the first indoor space, a warning signal is triggered, and the duration T1 of the gun being located in the second indoor space but not in the first indoor space is counted, and it is determined whether T1 is greater than T0. If T1 is greater than T0, an alarm signal is triggered; if T1 is greater than T0, no alarm signal is triggered, and T0 is the alarm time threshold; in this embodiment, T0 is 3 minutes;

[0078] If the gun is within the first indoor space, the warning signal is canceled; for example, if someone accidentally moves from the first indoor space to the second indoor space but returns in time, the warning signal can be canceled, but if the alarm signal is triggered, it cannot be canceled;

[0079] In the specific implementation process, ultra-wideband technology, also known as UWB technology, uses extremely narrow pulses in the nanosecond or microsecond level for communication and ranging. The distance can be obtained by measuring the flight time of the UWB signal from the transmitter to the receiver and then multiplying it by the speed of light. Because the UWB pulse width is extremely narrow and the time resolution is high, high-precision distance measurement can be achieved. For example, assuming that the flight time of the UWB signal is t, the distance = c*t, where c is the speed of light.

[0080] The outdoor acquisition module is used to establish outdoor calibration points, obtain outdoor gun position information and outdoor gun related information, and perform error correction processing on the outdoor gun position information to obtain corrected position information;

[0081] The outdoor acquisition module is configured with an acquisition correction strategy, which includes: establishing multiple calibration points at a first distance interval on the path of the gun, and recording the longitude and latitude of the calibration points, which are recorded as the reference longitude and latitude, and the first distance interval is LJ; the establishment of the calibration points requires the calibration points to be acquired as accurately as possible, and the satellite navigation system needs to be used for positioning; in this embodiment, the first distance interval LJ=100M;

[0082] When the gun is outdoors, the latitude and longitude, speed, acceleration magnitude and direction angle of the gun are obtained at a second time interval through the satellite navigation system, and the time of acquisition is recorded, the latitude and longitude of the gun are marked as outdoor gun position information, the speed, acceleration magnitude and direction angle of the gun are marked as outdoor gun related information, and the latitude and longitude of the two calibration points closest to the gun are obtained at the same time, which are recorded as actual longitude and latitude; in this embodiment, the second time interval r=6s;

[0083] See also Figure 3As shown, based on the actual longitude and latitude of the two calibration points and the corresponding reference longitude and longitude, the difference between the longitude and latitude of the two calibration points are calculated respectively, and the average value of the difference between the longitude and latitude is obtained respectively, which is marked as the average longitude and latitude error. Based on the average longitude and longitude error, the outdoor gun position information of the gun at the corresponding moment is corrected, and the corrected position information is obtained after completion; for example, at a certain moment, the longitude and latitude of the gun located by the satellite navigation system is (33.244653°, 117.365451), and the actual longitude and latitude of the two most recent calibration points are (33.244681°, 117.365478°) and (33.244618°, 117.365414°) in sequence, while the reference longitude and latitude of the two calibration points are (33.244697°, 117.365462°) in sequence. ) and (33.244632°, 117.365400°), then the average longitude error at this moment = [(33.244697-33.244681)+(33.244632-33.244618)] / 2=0.000015, the average latitude error = [(117.365462-117.365478)+(117.365400- 117.365414)] / 2=-0.000015; then the latitude and longitude of the gun (33.244653°, 117.365451°) is corrected, and the corrected position information at this moment is (33.244653+0.000015, 117.365451-0.000015)=((33.244668°, 117.365446°);

[0084] In the specific implementation process, the satellite navigation system itself has certain errors; this is due to the hardware limitations of the equipment, including factors such as chip accuracy and antenna performance; for example, the positioning accuracy of ordinary GPS devices is between several meters and more than ten meters. This error cannot be ignored in scenarios where the location of guns needs to be accurately monitored. At the same time, the interference of the external environment on the positioning signal will cause the positioning accuracy to decrease; and by correcting the errors of the acquired longitude and latitude based on the calibration points, the accuracy of the position data can be effectively improved; the calibration points are used as known precise reference positions, and the relationship between them and the gun positioning points is used for correction, which can reduce the deviation of the original positioning data caused by equipment errors, signal interference and other factors, and ensure that the accuracy of the positioning data is always at a high level.

[0085] The motion prediction module includes a model building unit and a position prediction unit. The model building unit is used to build a gun motion prediction model. The position prediction unit predicts the gun motion by using the corrected position information and outdoor gun related information to obtain gun motion prediction information.

[0086] The model building unit is configured with a model building strategy, which includes: using a long short-term memory network to build a first prediction model, the first prediction model includes: an input layer, an LSTM layer and an output layer, setting the input shape of the input layer to (A1, A2), the number of LSTM layers to A3, the number of units in the LSTM layer to A4, and the number of neurons in the output layer to A5; the loss function of the model training process is defined as follows: , where MS represents the loss value, Ui represents the actual longitude and latitude values, Vi represents the longitude and latitude values ​​predicted by the model, and n is the number of samples input to the model; the input shape is the time step and the number of features of the input data, A1 represents the time step, and A2 represents the number of features. For example, in this embodiment, (A1, A2) = (10, 5), because the second time interval is 6s, so A1 = 10, representing a 60s input of data at a time, A2 = 5, because it has five features: longitude, latitude, speed, acceleration, and direction angle; in this embodiment, A3 = 1, A4 = 64, A5 = 5, because the longitude, latitude speed, acceleration, and direction angle are to be output;

[0087] The corrected position information at the corresponding moment and the outdoor gun related information are combined to obtain the gun motion data; the gun motion data are normalized according to the data type, and all data sizes are scaled to [0, 1]; after completion, the normalized motion data is obtained; the normalized motion data is divided into a first training set and a first test set in a ratio of 8:2;

[0088] The first prediction model is trained using the first training set, and the second prediction model is obtained after completion; the second prediction model is tested using the first test set, and the mean absolute error MAE of the second prediction model is calculated. The mean absolute error calculation formula is as follows: , where MAE represents mean absolute error, Uk represents actual latitude and longitude values, Vk represents latitude and longitude values ​​predicted by the model, and K is the number of samples input to the model; determine whether MAE satisfies MAE≤MA0, where MA0 is a qualified threshold; if satisfied, mark the second prediction model as a gun motion prediction model; if not satisfied, train the second prediction model again with the first training set, and after completion, test it again with the first test set until a gun motion prediction model is obtained; in this embodiment, MA0=0.2, and the smaller the MAE output by the model, the higher the accuracy of the model prediction;

[0089] The location prediction unit is configured with a location prediction strategy, which includes: setting the data sliding window size to the first time step, the sliding step is E0, and the first time step is E1; in this embodiment, the sliding step E0 is 3, that is, the data of three moments are slid each time, and the data of three moments are output each time, and the first time step E1 is 5min, that is, the data of the previous 5 minutes is used to predict the data of the next 18s;

[0090] See also Figure 4 As shown, based on the normalized motion data, the data sliding window is used for the first time to intercept the data of the first time step, and the first gun motion data of the future E0 moments obtained by inputting the gun motion prediction model is recorded as the first prediction data;

[0091] Slide the data sliding window again for the second time, intercept the data of the first time step again, including the first predicted data, and input the first gun motion data of the future E0 moments obtained by the gun motion prediction model, which is recorded as the second predicted data;

[0092] The sliding window of the sliding data is used to repeatedly intercept data and input into the gun motion prediction model until the first gun motion data of the second time step in the future is obtained, and the latitude and longitude data in the first gun motion data are extracted and recorded as the gun motion prediction information. The second time step is E2; that is, the data obtained by each prediction is combined with the historical data input model to obtain new prediction data until a certain length of data is obtained. In this embodiment, the second time step E2 is 10, because the second time interval is 6s and the sliding step E0=3, that is, the gun motion prediction information of the next 3 minutes can be obtained each time;

[0093] In the specific implementation process, the LSTM layer is the core part of the model and is used to process time series data. The number of LSTM units needs to be specified, for example, 64 units. The number of units determines the complexity that the model can learn. The larger the number, the more complex time series patterns the model can capture, but it also increases the computational cost and the risk of overfitting. Multiple LSTM layers can be added as needed to further extract deep features in the time series. By re-inputting the predicted data into the model for further prediction, the model can gradually learn the dynamic change characteristics of the data and improve the prediction accuracy and stability.

[0094] The dynamic supervision module supervises the gun position based on the gun movement prediction information and the corrected position information;

[0095] The dynamic supervision module is configured with a dynamic supervision module, which includes: based on the corrected position information and the reference longitude and latitude of the two calibration points closest to the gun at the corresponding moment, calculating the vertical distance from the gun position to the line connecting the two calibration points closest to the gun, marking it as the current gun warning distance, recorded as L0;

[0096] Based on the firearm movement prediction information and the reference longitude and latitude of the two calibration points closest to the firearm at the corresponding moment, calculate the perpendicular distance from the firearm position at all future moments to the line connecting the two calibration points closest to the firearm, which is marked as the future firearm warning distance information and denoted as LX; since the firearm movement prediction information is the longitude and latitude at each moment within a certain future time period, the future firearm warning distance information is the perpendicular distance from the firearm position at each moment within a certain future time period to the line connecting the two calibration points closest to the firearm;

[0097] Set a first warning distance and a second warning distance. The first warning distance is L1, and the second warning distance is L2, where L1 < L2; in this embodiment, L1 is 50m and L2 is 100m;

[0098] If L0 > L2, then trigger an alarm signal;

[0099] If L1 < L0 < L2 and all of LX are less than L2, then trigger a warning signal; that is, the situation of L0 > L2 will not occur within a certain future time period;

[0100] If L1 < L0 < L2 and not all of LX are less than L2, then trigger an alarm signal; that is, the situation of L0 > L2 will occur within a certain future time period;

[0101] If the duration of L1 < L0 < L2 is greater than T0, then trigger an alarm signal;

[0102] If L0 < L1 and not all of LX are less than L1 but all are less than L2, then trigger a warning signal; that is, the situation of L0 > L1 will occur within a certain future time period, but the situation of L0 > L2 will not occur;

[0103] If L0 < L1 and not all of LX are less than L2, then trigger an alarm signal; that is, the situation of L0 > L2 will occur within a certain future time period;

[0104] In the specific implementation process, through the accurate prediction of future positions, potential risk situations can be discovered in advance; for example, if it is predicted that the firearm is about to enter an unauthorized area or approach an important safety facility, the supervision system can issue an alarm in advance, enabling the supervision personnel to have sufficient time to take measures for intervention and effectively preventing the occurrence of safety accidents.

[0105] Embodiment 2, please refer to Figure 2 As shown, the present application provides a firearm dynamic supervision method based on trajectory positioning, including the following steps:

[0106] Step S1, establishing an indoor gun positioning method based on ultra-wideband technology, positioning the guns indoors, and performing indoor dynamic monitoring; Step S1 includes the following sub-steps:

[0107] Step S101, establishing an indoor gun positioning method; Step S011 includes the following sub-steps:

[0108] Step S1011, multiple ultra-wideband base stations are arranged indoors, one of the ultra-wideband base stations is selected and marked as the origin base station, a spatial rectangular coordinate system is established with the position of the origin base station as the coordinate origin, and the spatial positions of all ultra-wideband base stations in the indoor spatial coordinate system are obtained and recorded as base station position information;

[0109] Step S1012, obtaining the spatial distance between the ultra-wideband base station and the gun through the ultra-wideband base station and the corresponding sensors on the gun, and selecting four ultra-wideband base stations with the closest spatial distance, marking them as positioning base stations, denoted as C1, C2, C3 and C4 respectively, and denoting the spatial distances between the positioning base station and the gun as d1, d2, d3 and d4 in order;

[0110] Step S1013, obtaining the spatial positions of the positioning base stations based on the base station position information, which are respectively recorded as C1 (x1, y1, z1), C2 (x2, y2, z2), C3 (x3, y3, z3) and C4 (x4, y4, z4);

[0111] Step S1014, calculating the spatial distance between any two positioning base stations based on the spatial positions of the positioning base stations, and recording it as the standard base station distance;

[0112] Step S1015: For any positioning base station, the spatial distance with other positioning base stations is obtained by ultra-wideband technology, and recorded as the measurement base station distance;

[0113] Step S1016, based on the standard base station distance, calculate the error between the measured base station distance and the corresponding standard base station distance, and calculate the unit distance error according to the corresponding measured base station distance, and take the average, marked as the average unit distance error; repeatedly obtain the average unit distance error of all positioning base stations, marked as unit error information;

[0114] Step S1017, based on the unit error information, d1, d2, d3 and d4 are corrected respectively, and after completion, they are recorded as qd1, qd2, qd3 and qd4 in order, marked as the actual gun distance;

[0115] Step S1018, based on the spatial position of the positioning base station and the actual gun distance, a gun position equation group is established, and the gun spatial position (x0, y0, z0) is calculated and solved. The gun position equation group is as follows: ;

[0116] Step S102, based on the indoor gun positioning method, the spatial position of the indoor gun is obtained at a first time interval, marked as indoor gun position information, and the first time interval is t;

[0117] Step S103, setting a first indoor space range and a second indoor space range, where the first indoor space range∈the second indoor space range;

[0118] Step S104, determining whether the gun is located in the first indoor space range and the second indoor space range based on the indoor gun position information;

[0119] Step S105, if the gun is not located in the first indoor space range and is not located in the second indoor space range, triggering an alarm signal;

[0120] Step S106, if the gun is located in the second indoor space range and not in the first indoor space range, a warning signal is triggered, and the duration T1 of the gun being located in the second indoor space range and not in the first indoor space range is counted, and it is determined whether T1 is greater than T0. If T1 is greater than T0, an alarm signal is triggered; if T1 is greater than T0, no alarm signal is triggered, and T0 is the alarm time threshold;

[0121] Step S107: if the gun is within the first indoor space, the warning signal is canceled.

[0122] Step S2, establishing an outdoor calibration point, obtaining outdoor gun position information and outdoor gun related information, and performing error correction processing on the outdoor gun position information to obtain corrected position information; Step S2 includes the following sub-steps:

[0123] Step S201, establishing a plurality of calibration points at a first distance interval on the path of the gun, and recording the longitude and latitude of the calibration points as the reference longitude and latitude, and the first distance interval is LJ;

[0124] Step S202, when the gun is located outdoors, the latitude and longitude, speed, acceleration magnitude and direction angle of the gun are obtained at a second time interval through the satellite navigation system, and the time of acquisition is recorded, the latitude and longitude of the gun are marked as outdoor gun position information, the speed, acceleration magnitude and direction angle of the gun are marked as outdoor gun related information, and the latitude and longitude of the two calibration points closest to the gun are obtained at the same time, recorded as actual longitude and latitude, and the second time interval is r;

[0125] Step S203, based on the actual longitude and longitude of the two calibration points and the corresponding reference longitude and longitude, respectively calculate the difference between the longitude and latitude of the two calibration points, and respectively obtain the average value of the difference between the longitude and latitude, which is marked as the average longitude and latitude error;

[0126] Step S204, based on the average longitude and latitude errors, the outdoor gun position information of the gun at the corresponding moment is corrected, and the corrected position information is obtained after completion.

[0127] Step S3, constructing a gun motion prediction model, using the corrected position information and outdoor gun related information to predict the gun motion, and obtaining gun motion prediction information; Step S3 includes the following sub-steps:

[0128] Step S301, constructing a first prediction model using a long short-term memory network, the first prediction model includes: an input layer, an LSTM layer and an output layer, setting the input shape of the input layer to (A1, A2), the number of LSTM layers to A3, the number of units in the LSTM layer to A4, and the number of neurons in the output layer to A5; defining the loss function of the model training process as follows: , where MS represents the loss value, Ui represents the actual longitude and latitude values, Vi represents the longitude and latitude values ​​predicted by the model, and n is the number of samples input to the model;

[0129] Step S302, combining the corrected position information at the corresponding moment and the outdoor gun-related information to obtain gun motion data; normalizing the gun motion data according to the data type, scaling all data sizes to [0, 1]; obtaining normalized motion data after completion; dividing the normalized motion data into a first training set and a first test set in a ratio of 8:2;

[0130] Step S303: train the first prediction model using the first training set, and obtain the second prediction model after completion; test the second prediction model using the first test set, and calculate the mean absolute error MAE of the second prediction model. The mean absolute error calculation formula is as follows: , where MAE stands for mean absolute error, Uk stands for the actual longitude and latitude values, Vk stands for the longitude and latitude values ​​predicted by the model, and K is the number of samples input to the model;

[0131] Step S304, determine whether MAE satisfies MAE≤MA0, where MA0 is a qualified threshold; if so, mark the second prediction model as a gun motion prediction model; if not, train the second prediction model again with the first training set, and after completion, test it again with the first test set until a gun motion prediction model is obtained;

[0132] Step S305, setting the data sliding window size to the first time step, the sliding step to E0, and the first time step to E1;

[0133] Step S306: Based on the normalized motion data, for the first time, use a data sliding window to intercept the data of the first time step length, and input it into the firearm motion prediction model to obtain the first firearm motion data at the next E0 moments, denoted as the first prediction data.

[0134] Step S307: Slide the data sliding window a second time, intercept the data of the first time step length again, including the first prediction data, and input it into the firearm motion prediction model to obtain the first firearm motion data at the next E0 moments, denoted as the second prediction data.

[0135] Step S308: Use the sliding data sliding window to repeatedly intercept data and input it into the firearm motion prediction model until the first firearm motion data at the next second time step length is obtained. Extract the latitude and longitude data in the first firearm motion data, denoted as firearm motion prediction information, and the second time step length is E2.

[0136] Step S4: Supervise the firearm position based on the firearm motion prediction information and the corrected position information. Step S4 includes the following sub-steps:

[0137] Step S401: Based on the corrected position information and the reference latitude and longitude of the two calibration points closest to the firearm at the corresponding moment, calculate the perpendicular distance from the firearm position to the line connecting the two calibration points closest to the firearm, marked as the current firearm warning distance, denoted as L0.

[0138] Step S402: Based on the firearm motion prediction information and the reference latitude and longitude of the two calibration points closest to the firearm at the corresponding moment, calculate the perpendicular distance from the firearm position to the line connecting the two calibration points closest to the firearm at all future moments, marked as the future firearm warning distance information, denoted as LX.

[0139] Step S403: Set the first warning distance and the second warning distance. The first warning distance is L1, and the second warning distance is L2, where L1 < L2.

[0140] Step S404: If L0 > L2, trigger an alarm signal.

[0141] Step S405: If L1 < L0 < L2 and all of LX are less than L2, trigger a warning signal.

[0142] Step S406: If L1 < L0 < L2 and not all of LX are less than L2, trigger an alarm signal.

[0143] Step S407: If the duration of L1 < L0 < L2 is greater than T0, trigger an alarm signal.

[0144] Step S408: If L0 < L1 and not all of LX are less than L1 but all are less than L2, trigger a warning signal.

[0145] In step S409, if L0 < L1 and not all of LX are less than L2, an alarm signal is triggered.

[0146] Example 3, please refer to Figure 5 as shown Figure 5 illustrates a schematic structural diagram of an electronic device, which may include: a processor, a communication interface, a memory, and a communication bus. Among them, the processor, the communication interface, and the memory complete mutual communication through the communication bus. The memory stores computer-readable instructions, and the processor can call the instructions in the memory. When the computer-readable instructions are executed by the processor, the steps in the firearm dynamic supervision method based on trajectory positioning are run to achieve the following functions: establishing an indoor firearm positioning method based on ultra-wideband technology to position the firearms indoors and conduct indoor dynamic supervision; establishing outdoor calibration points, obtaining outdoor firearm position information and outdoor firearm-related information, and performing error correction processing on the outdoor firearm position information to obtain corrected position information; constructing a firearm motion prediction model, using the corrected position information and outdoor firearm-related information to predict the firearm motion to obtain firearm motion prediction information; supervising the firearm position based on the firearm motion prediction information and the corrected position information.

[0147] In addition, when the logical instructions in the above-mentioned memory are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0148] Embodiment 4, the present application also provides a computer-readable storage medium, the present application provides a storage medium, on which a computer program is stored. When the computer program is executed by the processor, the steps in the dynamic gun supervision method based on trajectory positioning are executed to achieve the following functions: establish an indoor gun positioning method based on ultra-wideband technology, locate the indoor gun, and perform dynamic indoor supervision; establish outdoor calibration points, obtain outdoor gun position information and outdoor gun-related information, and perform error correction processing on the outdoor gun position information to obtain corrected position information; construct a gun motion prediction model, use the corrected position information and outdoor gun-related information to predict the gun motion, and obtain gun motion prediction information; supervise the gun position based on the gun motion prediction information and the corrected position information.

[0149] Through the description of the above implementation methods, the embodiments of the present invention can be provided as methods, systems or computer program products. Based on such an understanding, the above technical solutions can be essentially or partly contributed to the prior art in the form of software products, which can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and include several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0150] In the embodiments provided in the present application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of systems, modules and units can be electrical, mechanical or other forms.

[0151] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A dynamic gun monitoring system based on trajectory positioning, characterized in that: It includes indoor monitoring module, outdoor acquisition module, motion prediction module and dynamic monitoring module; The indoor supervision module includes an indoor positioning unit and an indoor management unit. The indoor positioning unit establishes an indoor gun positioning method based on ultra-wideband technology to locate the indoor guns. The indoor management unit is used to perform indoor dynamic supervision on the indoor guns. The outdoor acquisition module is used to establish outdoor calibration points, obtain outdoor gun position information and outdoor gun related information, and perform error correction processing on the outdoor gun position information to obtain corrected position information; The motion prediction module includes a model building unit and a position prediction unit, the model building unit is used to build a gun motion prediction model, and the position prediction unit predicts the gun motion by using the corrected position information and outdoor gun related information to obtain gun motion prediction information; The dynamic supervision module supervises the gun position based on the gun movement prediction information and the corrected position information; The outdoor acquisition module is configured with an acquisition correction strategy, which includes: Establish multiple calibration points at a first distance interval on the path of the gun, and record the longitude and latitude of the calibration points as the reference longitude and latitude, and the first distance interval is LJ; When the gun is outdoors, the latitude and longitude, speed, acceleration magnitude and direction angle of the gun are obtained at a second time interval through the satellite navigation system, and the time of acquisition is recorded, the latitude and longitude of the gun are marked as outdoor gun position information, and the speed, acceleration magnitude and direction angle of the gun are marked as outdoor gun related information, and the latitude and longitude of the two calibration points closest to the gun are obtained at the same time, recorded as actual latitude and longitude, and the second time interval is r; Based on the actual longitude and latitude of the two calibration points and the corresponding reference longitude and longitude, the difference between the longitude and latitude of the two calibration points is calculated respectively, and the average value of the difference between the longitude and latitude is obtained respectively, which is marked as the average longitude and latitude error. Based on the average longitude and latitude error, the outdoor gun position information of the gun at the corresponding moment is corrected, and the corrected position information is obtained after completion; The location prediction unit is configured with a location prediction strategy, which includes: Set the data sliding window size to the first time step, the sliding step to E0, and the first time step to E1; Based on the normalized motion data, the data of the first time step is intercepted by using the data sliding window for the first time, and the first gun motion data of the future E0 moments is obtained by inputting into the gun motion prediction model, which is recorded as the first prediction data; Slide the data sliding window again for the second time, intercept the data of the first time step again, including the first predicted data, and input the first gun motion data of the future E0 moments obtained by the gun motion prediction model, which is recorded as the second predicted data; The sliding data sliding window is used to repeatedly intercept data and input into the gun motion prediction model until the first gun motion data of the second time step in the future is obtained, and the longitude and latitude data in the first gun motion data are extracted and recorded as gun motion prediction information. The second time step is E2.

2. The dynamic gun monitoring system based on trajectory positioning according to claim 1 is characterized in that: The indoor positioning unit is configured with indoor positioning strategies, which include: Establish an indoor gun positioning method: arrange multiple ultra-wideband base stations indoors, select one of the ultra-wideband base stations and mark it as the origin base station, establish a spatial rectangular coordinate system with the position of the origin base station as the coordinate origin, mark it as the indoor spatial coordinate system, obtain the spatial positions of all ultra-wideband base stations in the indoor spatial coordinate system, and record it as the base station position information; The spatial distance between the ultra-wideband base station and the gun is obtained through the corresponding sensors on the ultra-wideband base station and the gun, and the four ultra-wideband base stations with the closest spatial distance are selected and marked as positioning base stations, which are denoted as C1, C2, C3 and C4 respectively, and the spatial distances between the positioning base station and the gun are denoted as d1, d2, d3 and d4 in order; then the spatial positions of the positioning base stations are obtained based on the base station position information, which are denoted as C1(x1, y1, z1), C2(x2, y2, z2), C3(x3, y3, z3) and C4(x4, y4, z4) respectively.

3. The dynamic gun monitoring system based on trajectory positioning according to claim 2 is characterized in that: The indoor positioning unit is configured with an indoor positioning strategy, which also includes: Based on the spatial positions of the positioning base stations, the spatial distance between any two positioning base stations is calculated and recorded as the standard base station distance; For any positioning base station, the spatial distance with other positioning base stations is obtained through ultra-wideband technology, which is recorded as the measuring base station distance; then based on the standard base station distance, the error between the measuring base station distance and the corresponding standard base station distance is calculated, and the unit distance error is calculated according to the corresponding measuring base station distance, and the average is obtained, which is marked as the average unit distance error; the average unit distance error of all positioning base stations is obtained repeatedly, which is marked as the unit error information; Based on the unit error information, d1, d2, d3 and d4 are corrected respectively. After completion, they are recorded as qd1, qd2, qd3 and qd4 in order, marked as the actual gun distance; The gun position equations are established based on the spatial position of the positioning base station and the actual gun distance, and the gun's spatial position (x0, y0, z0) is calculated and solved. The gun position equations are as follows: ； Based on the indoor gun positioning method, the spatial position of the indoor gun is obtained at a first time interval, marked as indoor gun position information, and the first time interval is t.

4. The dynamic gun monitoring system based on trajectory positioning according to claim 3 is characterized in that: The indoor management unit is configured with indoor management strategies, which include: Set a first indoor space range and a second indoor space range, the first indoor space range∈the second indoor space range; Determine whether the gun is located within the first indoor space range and the second indoor space range based on the indoor gun position information; If the gun is neither within the first indoor space nor within the second indoor space, an alarm signal is triggered; If the gun is located in the second indoor space but not in the first indoor space, a warning signal is triggered, and the duration T1 of the gun being located in the second indoor space but not in the first indoor space is counted, and it is determined whether T1 is greater than T0. If T1 is greater than T0, an alarm signal is triggered; if T1 is less than or equal to T0, no alarm signal is triggered, and T0 is the alarm time threshold; If the gun is within the first indoor space, the warning signal is cancelled.

5. The dynamic gun monitoring system based on trajectory positioning according to claim 4 is characterized in that: The model building unit is configured with a model building strategy, which includes: The first prediction model is constructed using the long short-term memory network. The first prediction model includes: an input layer, an LSTM layer, and an output layer. The input shape of the input layer is set to (A1, A2), where A1 represents the time step, A2 represents the number of features, the number of LSTM layers is A3, the number of units in the LSTM layer is A4, and the number of neurons in the output layer is A5. The loss function of the model training process is defined as follows: , where MS represents the loss value, Ui represents the actual longitude and latitude values, Vi represents the longitude and latitude values ​​predicted by the model, and n is the number of samples input to the model; Combine the corrected position information at the corresponding moment and the outdoor gun-related information to obtain gun movement data; perform normalization processing on the gun movement data according to the data types respectively, and scale all data sizes to [0, 1]; after completion, obtain the normalized movement data; divide the normalized movement data into a first training set and a first test set according to a ratio of 8:2; The first prediction model is trained using the first training set, and the second prediction model is obtained after completion; the second prediction model is tested using the first test set, and the mean absolute error MAE of the second prediction model is calculated. The mean absolute error calculation formula is as follows: , where MAE represents mean absolute error, Uk represents the actual longitude and latitude values, Vk represents the longitude and latitude values ​​predicted by the model, and K is the number of samples input to the model; determine whether MAE satisfies MAE≤MA0, where MA0 is the qualified threshold; if so, mark the second prediction model as a gun motion prediction model; if not, train the second prediction model again with the first training set, and then test it again with the first test set until a gun motion prediction model is obtained.

6. The dynamic gun monitoring system based on trajectory positioning according to claim 5 is characterized in that: The dynamic supervision module is configured with a dynamic supervision strategy, and the dynamic supervision strategy includes: Based on the corrected position information and the reference longitude and latitude of the two calibration points closest to the gun at the corresponding moment, calculate the vertical distance from the gun position to the line connecting the two calibration points closest to the gun, and mark it as the current gun warning distance, denoted as L0; Based on the gun movement prediction information and the reference longitude and latitude of the two calibration points closest to the gun at the corresponding moment, calculate the vertical distance from the gun position at all future moments to the line connecting the two calibration points closest to the gun, and mark it as the future gun warning distance information, denoted as LX; Set a first warning distance and a second warning distance, the first warning distance is L1, the second warning distance is L2, and L1 < L2; If L0 > L2, then trigger an alarm signal; If L1 < L0 < L2 and all LX are less than L2, then trigger a warning signal; If L1 < L0 < L2 and not all LX are less than L2, then trigger an alarm signal; If the duration of L1 < L0 < L2 is greater than T0, then trigger an alarm signal; If L0 < L1 and not all LX are less than L1 but all are less than L2, then trigger a warning signal; If L0 < L1 and not all LX are less than L2, then trigger an alarm signal.

7. A method for dynamic monitoring of firearms based on trajectory positioning, applicable to a dynamic monitoring system for firearms based on trajectory positioning as claimed in any one of claims 1 to 6, characterized in that: Including the following steps: Establish an indoor gun positioning method based on ultra-wideband technology, position the guns indoors, and perform indoor dynamic supervision; Establish calibration points outdoors, obtain outdoor gun position information and outdoor gun-related information, and perform error correction processing on the outdoor gun position information to obtain corrected position information; Construct a gun movement prediction model, and use the corrected position information and outdoor gun-related information to predict the gun movement to obtain gun movement prediction information; Supervise the gun position based on the gun movement prediction information and the corrected position information.

8. An electronic device, characterized in that: Including a processor and a memory, the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the steps in the gun dynamic supervision method based on trajectory positioning as claimed in claim 7 are run.

Images