A method and apparatus for track optimization

Abstract

The application relates to a track optimization method and device, and the method comprises the following steps: obtaining initial point track data of each scanning period of a radar and pre-processing the initial point track data to obtain first point track data; obtaining point-point correlation initial wave gates; calculating the spatial relative displacement amount between the first point track data of adjacent scanning periods, and determining whether the point track of the adjacent scanning periods is successfully correlated according to the spatial relative displacement amount and the point-point correlation initial wave gates; when the correlation is successful, obtaining an initial prediction value of point-point correlation according to the spatial relative displacement amount, setting the weight of the point-point correlation wave gate based on the r-order convergence thought, and setting the optimized point-point correlation wave gate based on the initial prediction value of the point-point correlation to obtain first point-point correlation wave gates; when the target point track falls into the first point-point correlation wave gates, filtering the target point track based on an alpha-beta filtering algorithm to obtain track data and perform track optimization, and obtaining optimal track data of the target point track. The application reduces the system resource occupation amount of the target track.

Description

Technical Field

[0001] This invention relates to the field of radar tracking technology, and in particular to a trajectory optimization method. Background Technology

[0002] Track quality is an important indicator for evaluating the performance of a radar system.

[0003] Because ground-based early warning radars are often deployed in environments such as coastlines, urban edges, deserts, and forests, they face complex and ever-changing clutter effects. Even after traditional clutter map processing, a large amount of clutter and interference still exists, consuming system resources and leading to a reduction in track quality and system detection performance. Summary of the Invention

[0004] In view of this, it is necessary to provide a trajectory optimization method to address the goal of reducing the resource consumption of the target table trajectory system.

[0005] To achieve the above objectives, the present invention provides a trajectory optimization method, comprising:

[0006] Acquire the initial point data for each radar scan cycle;

[0007] The initial dot data is preprocessed to obtain the first dot data;

[0008] Obtain the initial gate associated with each point;

[0009] Calculate the spatial relative displacement between the first point data in adjacent scanning cycles of the radar in each scanning cycle, and determine whether the point data in adjacent scanning cycles are successfully associated based on the spatial relative displacement and the point association initial gate.

[0010] If the point association between adjacent scanning cycles is successfully determined, the initial prediction value of the point association is obtained based on the spatial relative displacement, the point navigation association gate is obtained, the weight of the point navigation association gate is set based on the r-order convergence idea to obtain the optimized point navigation association gate, and the first point navigation association gate is obtained by setting the optimized point navigation association gate based on the initial prediction value of the point association.

[0011] If it is determined that the target point track falls within the first point flight path gate, the target point track is filtered based on the α-β filtering algorithm to obtain the target point track data;

[0012] The trajectory data of the target point is optimized to obtain the optimal trajectory data of the target point.

[0013] In some possible implementations, the point-to-point association initial gate includes a range initial gate, an azimuth initial gate, and a pitch initial gate, and the expression for the range initial gate is:

[0014] R_thr0=W0*tgtV_max*Frame_updateTime

[0015] The initial gate expression for the azimuth angle is:

[0016] Az_thr0=W1*(3*Az_index+Az_max(i))

[0017] The initial gate expression for the pitch angle is:

[0018]

[0019] In the formula, R_thr0 represents the initial range gate; tgtV_max represents the maximum target detection velocity; Frame_updateTime represents the inter-frame update time; Az_th0 represents the initial azimuth gate; Az_max(i) represents the azimuth change in each range segment; and Az_index represents the system azimuth measurement accuracy.

[0020] El_thr0 represents the initial gate for the pitch angle; W0, W1, and W2 represent weighting values; El_BeamWidth represents the pitch beamwidth; and K is a constant.

[0021] In some possible implementations, the spatial relative displacement between the initial point data of adjacent scan cycles in each scan cycle of the radar includes: the relative displacement of the distance between the initial point data of adjacent scan cycles, the relative displacement of the azimuth angle between the initial point data of adjacent scan cycles, and the relative displacement of the elevation angle between the initial point data of adjacent scan cycles.

[0022] In some possible implementations, determining whether the traces of adjacent scan cycles are successfully associated based on the spatial relative displacement and the point-to-point associated initial gate includes:

[0023] When the spatial relative displacement is less than the initial gate of the point-to-point association, the point-to-point association of the adjacent scanning cycles is determined to be successful.

[0024] In some possible implementations, when the target point trace length is greater than a preset value, the expression for the weight of the point navigation associated gate is:

[0025]

[0026] In the formula, W(p0) represents the weighting value currently used for the target point trace; W(p0+1) represents the weighting value of the next point navigation gate; r represents the convergence coefficient; and L represents the length of the target point trace.

[0027] In some possible implementations, the filtering of the target point traces based on the α-β filtering algorithm to obtain the target point trace track data includes:

[0028] The α-β filtering algorithm is used to obtain the filtered value of the current point's trace data based on the current point's trace data of the target point trace;

[0029] Based on the filtered value of the current point's trace data, the predicted distance, azimuth, and pitch angle values ​​for the next frame of the current point are obtained.

[0030] The trajectory data of the target point is determined based on the distance prediction, azimuth prediction, and pitch prediction values ​​of the next frame for the current point.

[0031] In some possible implementations, the step of optimizing the trajectory data of the target point to obtain the optimal trajectory data of the target point includes:

[0032] Based on the trajectory motion characteristics weights and the error correlation rate weights, the trajectory accuracy weights are used to obtain trajectory performance evaluation indicators.

[0033] The flight path is divided into several levels according to the aforementioned flight path performance evaluation indicators;

[0034] The optimal trajectory data for the target point is obtained based on the aforementioned levels.

[0035] In some possible implementations, the weight of the trajectory motion characteristics is equal to the target velocity change value multiplied by the target azimuth change value multiplied by the target altitude change value; the error correlation rate weight is obtained based on the matching of radar trajectory data and secondary radar trajectory data; and the trajectory accuracy weight is equal to the system range accuracy index multiplied by the system azimuth accuracy index multiplied by the system pitch accuracy index.

[0036] In some possible implementations, obtaining the optimal trajectory data of the target point based on the aforementioned levels includes:

[0037] When the trajectory performance evaluation index is at the first level, the first weighted convergence coefficient and the first point-of-flight correlation gate are obtained. Based on the first weighted convergence coefficient and the first point-of-flight correlation gate, the optimal trajectory data of the target point is obtained. The first weighted convergence coefficient is obtained by increasing the weighted convergence coefficient of the point-of-flight correlation gate, and the first point-of-flight correlation gate is obtained by decreasing the value of the point-of-flight correlation gate.

[0038] When the trajectory performance evaluation index is at level two, the second point-linked wave gate, the first filter parameter, and the second filter parameter are obtained. Based on the second point-linked wave gate, the first filter parameter, and the second filter parameter, the optimal trajectory data of the target point is obtained. The second point-linked wave gate is obtained by increasing the value of the point-linked wave gate. The first filter parameter is obtained by increasing the value of the α parameter in the α-β filtering algorithm. The second filter parameter is obtained by increasing the value of the β parameter in the α-β filtering algorithm.

[0039] When the trajectory performance evaluation index is at level three, the third point-related wave gate, the third filtering parameter, and the fourth filtering parameter are obtained. Based on the third point-related wave gate, the third filtering parameter, and the fourth filtering parameter, the optimal trajectory data of the target point is obtained. The third point-related wave gate is obtained by adjusting the value of the point-related wave gate. The third filtering parameter is obtained by adjusting the value of the α parameter in the α-β filtering algorithm. The fourth filtering parameter is obtained by adjusting the value of the -β parameter in the α-β filtering algorithm.

[0040] On the other hand, the present invention also provides a trajectory optimization device, comprising:

[0041] The data acquisition unit is used to acquire the initial point data for each scanning cycle of the radar;

[0042] The first dot data acquisition unit is used to preprocess the initial dot data to obtain the first dot data;

[0043] The point-to-point associative initial gate setting unit is used to obtain the point-to-point associative initial gate.

[0044] The dot-matrix association determination unit is used to calculate the spatial relative displacement between the first dot-matrix data in adjacent scanning cycles in each scanning cycle of the radar, and determine whether the dot-matrix in adjacent scanning cycles is successfully associated based on the spatial relative displacement and the dot-matrix association initial gate.

[0045] The point navigation association gate setting unit is used to obtain the initial prediction value of point-to-point association based on the spatial relative displacement when it is determined that the point trace association of the adjacent scanning cycle is successful, obtain the point navigation association gate, set the weight of the point navigation association gate based on the r-order convergence idea to obtain the optimized point navigation association gate, and set the optimized point navigation association gate based on the initial prediction value of point-to-point association to obtain the first point navigation association gate.

[0046] The track data acquisition unit is used to filter the target point track based on the α-β filtering algorithm to obtain the track data of the target point track when it is determined that the target point track falls into the first point track association gate;

[0047] The optimal trajectory data acquisition unit is used to optimize the trajectory data of the target point to obtain the optimal trajectory data of the target point.

[0048] The beneficial effects of the above embodiments are as follows: The trajectory optimization method provided by the present invention acquires the initial point track data of each radar scanning cycle, preprocesses the initial point track data to obtain the first point track data, acquires the point-to-point association initial gate, determines whether the point tracks of adjacent scanning cycles are successfully associated based on the spatial relative displacement between the first point track data of adjacent cycles and the point-to-point association initial gate, acquires the point-to-point association gate when the point tracks are successfully associated, obtains the point-to-point association gate based on the spatial relative displacement, sets the weight of the point-to-point association gate based on the r-order convergence idea to obtain the optimized point-to-point association gate, sets the optimized point-to-point association gate based on the initial prediction value of point-to-point association to obtain the first point-to-point association gate, when it is determined that the target point track falls into the first point-to-point association gate, filters the target point track based on the α-β filtering algorithm to obtain the trajectory data of the target point track, and finally optimizes the trajectory data of the target point track to obtain the optimal trajectory data of the target point track. This invention sets the point-to-point correlation gate based on the initial preset value of point-to-point correlation and performs track processing based on the α-β filtering algorithm, which reduces the system resource consumption of the target track and optimizes the track offline, thereby improving the quality of the target track and the system performance. Attached Figure Description

[0049] Figure 1 A flowchart illustrating an embodiment of a trajectory optimization method provided by the present invention;

[0050] Figure 2 This is a schematic diagram of an embodiment of a trajectory optimization device provided by the present invention. Detailed Implementation

[0051] Preferred embodiments of the present invention will now be described in detail with reference to the accompanying drawings, which form part of this application and are used together with the embodiments of the present invention to illustrate the principles of the present invention, but are not intended to limit the scope of the present invention.

[0052] Figure 1 This is a schematic flowchart of an embodiment of the trajectory optimization method provided by the present invention, as shown below. Figure 1 As shown, a trajectory optimization method includes:

[0053] S101. Acquire the initial spot data for each radar scan cycle;

[0054] S102. Preprocess the initial dot data to obtain the first dot data;

[0055] S103, Obtain the initial gate associated with each point;

[0056] S104. Calculate the spatial relative displacement between the first point data in adjacent scanning cycles in each scanning cycle of the radar, and determine whether the point data in adjacent scanning cycles are successfully associated based on the spatial relative displacement and the point association initial gate.

[0057] S105. If the point association between adjacent scanning cycles is successfully determined, the initial prediction value of point association is obtained according to the spatial relative displacement, the point navigation association gate is obtained, the weight of the point navigation association gate is set according to the r-order convergence idea to obtain the optimized point navigation association gate, and the optimized point navigation association gate is set according to the initial prediction value of point association to obtain the first point navigation association gate.

[0058] S106. If it is determined that the target point track falls into the first point flight path gate, the target point track is filtered based on the α-β filtering algorithm to obtain the target point track data.

[0059] S107. Optimize the trajectory data of the target point to obtain the optimal trajectory data of the target point.

[0060] Compared with existing technologies, this embodiment provides a trajectory optimization method that acquires initial point data for each radar scan cycle, preprocesses the initial point data to obtain first point data, acquires an initial gate for point-to-point association, determines whether the point data in adjacent scan cycles are successfully associated based on the spatial relative displacement between the first point data in adjacent cycles and the initial gate for point-to-point association, acquires a point-to-navigation association gate, obtains the initial prediction value of point-to-point association based on the spatial relative displacement, sets the weight of the point-to-navigation association gate based on the r-order convergence idea to obtain an optimized point-to-navigation association gate, sets the optimized point-to-navigation association gate based on the initial prediction value of point-to-navigation association to obtain a first point-to-navigation association gate, and, when it is determined that the target point falls into the first point-to-navigation association gate, filters the target point based on the α-β filtering algorithm to obtain the trajectory data of the target point, and finally optimizes the trajectory data of the target point to obtain the optimal trajectory data of the target point. This invention sets the point-to-point correlation gate based on the initial preset value of point-to-point correlation and performs track processing based on the α-β filtering algorithm, which reduces the system resource consumption of the target track and optimizes the track offline, thereby improving the quality of the target track and the system performance.

[0061] In a specific embodiment of the present invention, in step S101, point data for each radar scanning cycle is acquired. The radar point data expression is point_data(N,para), where N represents the number of points and para represents the point characteristics. para includes time, range (R), azimuth (Az), and elevation (El) information.

[0062] It should be noted that, due to differences in the characteristics of the points within the same detection airspace, there may be a lag in the formation time of the points. The concept of differentiation is used to rearrange the temporal sequence of the points. In a specific embodiment of the present invention, in step S102, the initial point data is preprocessed to obtain the first point data. Firstly, the initial point data is rearranged in the temporal domain, specifically including the following steps:

[0063] Step 1: Obtain the time difference Time_gap between adjacent points by calculating the difference through decreasing time between adjacent points. Time_gap(N-1) = point_data(p+1,time) - point_data(p,time); where N represents the total number of points.

[0064] Step 2: When Time_gap is less than zero, it is determined that there is a time lag between adjacent points;

[0065] Step 3: Swap the positions of adjacent dots to complete the dot rearrangement;

[0066] Step 4: Repeat steps 1 to 3 until there is no lag.

[0067] It should be noted that in the azimuth scanning logic of modern phased array radar, each frame of azimuth scanning starts from the initial azimuth, and this characteristic is used to extract the scan points in each frame. In a specific embodiment of the present invention, in step S102, based on the temporal rearrangement of the initial point data, each frame of the initial point data is extracted to obtain the first point data, specifically including the following steps:

[0068] Step 1: Calculate the difference between the azimuth angles of adjacent points in the initial trace data, decreasing sequentially, to obtain the azimuth difference Az_gap between adjacent points.

[0069] Az_gap(N-1) = point_data(p+1,Az) - point_data(p,Az); where p∈(1,N-1), and N represents the total number of points;

[0070] Step 2: When Az_gap > Frame_Az_thr, it is determined that the adjacent points belong to different frame ranges. Frame_Az_thr is the inter-frame azimuth threshold. In this embodiment, the azimuth threshold is set to 80.

[0071] Step 3: Extract and save the point data of each frame to obtain the first point data.

[0072] It should be noted that in step S103, the initial gate for point-to-point association is obtained based on the radar system's operating mode and characteristics. According to the system's operating mode and characteristics, and combined with three-dimensional (range point_R, azimuth point_Az, elevation point_E1) spatial information, the initial gates for point-to-point association, R_thr0, Az_thr0, and E1_thr0, are set.

[0073] In some embodiments of the present invention, in step S103, the point-to-point association initial gate includes a range initial gate, an azimuth initial gate, and a pitch initial gate, and the expression for the range initial gate is:

[0074] R_thr0=W0*tgtV_max*Frame_updateTime

[0075] The initial gate expression for the azimuth angle is:

[0076] Az_thr0=W1*(3*Az_index+Az_max(i))

[0077] The initial gate expression for the pitch angle is:

[0078]

[0079] In the formula, R_thr0 represents the initial range gate; tgtV_max represents the maximum target detection speed; Frame_updateTime represents the inter-frame update time; Az_th0 represents the initial azimuth gate; Az_max(i) represents the azimuth change in each range segment, and the azimuth change is the largest when the target is flying tangentially. Using geometric relationships, the azimuth change Az_max(i) for each distance segment after the target flies tangentially at a distance R_thr0 is calculated. This invention sets 6 distance intervals, spaced 60km apart. When the target distance exceeds 360km, the Az_max(i) of the 6th distance interval is used. Az_index represents the system azimuth measurement accuracy; El_thr0 represents the initial gate of the pitch angle; W0, W1, and W2 represent weighting values; El_BeamWidth represents the pitch beamwidth; K is a constant. In this embodiment, an excessively large initial gate design will increase the false correlation rate, while an excessively small one will reduce the target detection probability. Therefore, it is necessary to optimize the design based on system performance and actual air situation. W0, W1, and W2 are set to 1, 0.5, and 0.5, respectively.

[0080] In some embodiments of the present invention, in step S105, the spatial relative displacement between the initial point data of adjacent scanning cycles in each scanning cycle of the radar includes: the relative displacement of the distance between the initial point data of adjacent scanning cycles, the relative displacement of the azimuth angle between the initial point data of adjacent scanning cycles, and the relative displacement of the elevation angle between the initial point data of adjacent scanning cycles.

[0081] In some embodiments of the present invention, step S105, which involves determining whether the traces of adjacent scanning cycles are successfully associated based on the spatial relative displacement and the point-to-point associated initial gate, includes:

[0082] When the spatial relative displacement is less than the initial gate of the point-to-point association, the point-to-point association of the adjacent scanning cycles is determined to be successful.

[0083] In a specific embodiment of the present invention, the current frame's trace is extracted, and its spatial relative displacement with all traces in the next frame is calculated:

[0084] R_dif0=point_data(p+1,R)-point_data(p,R);

[0085] Az_dif0=point_data(p+1,Az)-point_data(p,Az);

[0086] El_dif0=point_data(p+1,El)-point_data(p,El).

[0087] When the relative spatial displacement falls within the initial gate, the point-to-point association is considered successful, meaning that the following conditions are met simultaneously:

[0088]

[0089] Calculate the initial predicted values ​​of distance, azimuth, and elevation: R_Pre, Az_Pre, and El_Pre.

[0090] R_pre=point_data(p+1,R)+vR*Frame_updateTime,

[0091] Az_pre=point_data(p+1,Az)+vAz*Frame_updateTime,

[0092] El_pre=point_data(p+1,El)+vEl*Frame_updateTime.

[0093] Where vR, vAz, and vEl are respectively:

[0094] vR = R_dif0 / Frame_updateTime

[0095] vAz=Az_dif0 / Frame_updateTime,

[0096] vEl=El_dif0 / Frame_updateTime.

[0097] Track Initiation: Once the point-to-point association is completed, the new track batch is considered to have started successfully, the track batch number is increased, and the process of linking points to the track begins.

[0098] In some embodiments of the present invention, when the length of the target point trace is greater than a preset value, the expression for the weight of the point navigation associated gate is:

[0099]

[0100] In the formula, W(p0) represents the weighting value currently used for the target point trace; W(p0+1) represents the weighting value of the next point navigation gate; r represents the convergence coefficient; and L represents the length of the target point trace.

[0101] In a specific embodiment of the invention, the point navigation association refers to the initial gate design scheme of point-to-point association, and the weighting value of the point navigation association gate is designed in two stages:

[0102] When the track length L ≤ L_thr0, W0, W1 and W2 are set to 0.5, 0.25 and 0.25 respectively. L_thr0 reflects the track tracking stability, and is set to 10 in this invention.

[0103] When the track length L > L_thr0, i.e., after stable track tracking, the gate size of the convergent track association is gradually reduced to decrease system resource consumption and improve system detection performance. The weighting value is calculated based on the r-order convergence concept:

[0104]

[0105] In the formula, W(p0) represents the weighting value currently used for the track, and W(p0+1) represents the weighting value of the next point-related gate. The convergence coefficient r of this invention is set to 0.3. When the convergence reaches 1 / X of the initial weighting value, the values ​​of W0, W1 and W2 are no longer updated. The value of X of this invention is set to 5.

[0106] It should be noted that the point-to-point association is successful only if the target point track falls within the point-to-point navigation association gate. In some embodiments of the present invention, in step S106, the filtering process of the target point track based on the α-β filtering algorithm to obtain the target point track trajectory data includes:

[0107] Based on the α-β filtering algorithm, the filtered value of the trace data of the current point of the target trace is obtained according to the trace data of the current point;

[0108] According to the filtered value of the trace data of the current point, the distance prediction value, azimuth prediction value, and pitch angle prediction value of the next frame of the current point are obtained;

[0109] According to the distance prediction value, azimuth prediction value, and pitch angle prediction value of the next frame of the current point, the track data of the target trace is determined.

[0110] In some embodiments of the present invention, the expression of the a parameter in the α-β filtering algorithm is:

[0111]

[0112] The expression of the β parameter in the α-β filtering algorithm is:

[0113]

[0114] In the formula, L represents the target track length. When α < X1 or β < X2, the α-β value is no longer updated. In this embodiment, X1 and X2 are respectively set to 0.1 and 0.01.

[0115] In some embodiments of the present invention, after obtaining the parameters α and β of the α-β filtering algorithm, the filtered values of distance, azimuth, and pitch, R_flt, Az_flt, and El_flt, are calculated:

[0116] R_flt = R_pre + α * (point_data(p0, R) - R_pre),

[0117] Az_flt = Az_pre + α * (point_data(p0, Az) - Az_pre),

[0118] El_flt = El_pre + α * (point_data(p0, El) - El_pre).

[0119] Calculate the velocity filtered values of distance, azimuth, and pitch, vR_flt, vAz_flt, and vEl_flt:

[0120]

[0121]

[0122]

[0123] In the formula, point_data(p0,R), point_data(p0,Az), and point_data(p0,El) represent the distance, azimuth, and elevation information of the currently associated point, respectively.

[0124] The predicted values ​​of range, azimuth, and pitch are calculated using the α-β filter value. Point navigation association is then performed on the next frame of data. R_prr represents the predicted range value of the current point in the next frame, Az_pre represents the predicted azimuth value of the current point in the next frame, and El_pre represents the predicted pitch value of the current point in the next frame.

[0125] R_pre=R_flt+vR_flt*Frame_updateTime,

[0126] Az_pre=Az_flt+vAz_flt*Frame_updateTime,

[0127] El_pre=El_flt+vEl_flt*Frame_updateTime.

[0128] If the target point does not fall within the point-to-way association gate, it indicates that the point-to-way association has failed. In this case, blind inference is performed on the track, and the range, azimuth, and pitch of the blind-inferenced point are taken from the predicted values ​​of the previous frame. The range, azimuth, and pitch velocities of the blind-inferenced point are taken from the filtered values ​​of the previous frame.

[0129] Track termination. If a track is blindly advanced Y times consecutively, point navigation association will no longer be performed, indicating that the track has terminated. In this embodiment, Y is set to 5.

[0130] Short track and stationary feature optimization: After removing blind-throw points from the track, if the track length is less than L_thr1, it is judged as a short track and removed. In this invention, L_thr1 is set to 10. The standard deviations of track distance, azimuth, and pitch are calculated. If they are all less than the standard deviation thresholds for distance, azimuth, and pitch, they are judged as stationary targets and removed. The standard deviation thresholds for distance, azimuth, and pitch are set according to different system states.

[0131] It should be noted that the trajectory performance evaluation of this invention is performed in an offline state. In some embodiments of this invention, step S107, which optimizes the trajectory data of the target point to obtain the optimal trajectory data of the target point, includes:

[0132] Based on the trajectory motion characteristics weights and the error correlation rate weights, the trajectory accuracy weights are used to obtain trajectory performance evaluation indicators.

[0133] The flight path is divided into several levels according to the aforementioned flight path performance evaluation indicators;

[0134] The optimal trajectory data for the target point is obtained based on the aforementioned levels.

[0135] In some embodiments of the present invention, the weight of the trajectory motion characteristics is equal to the target velocity change value multiplied by the target azimuth change value multiplied by the target altitude change value; the error correlation rate weight is obtained based on the matching of radar trajectory data and secondary radar trajectory data; the trajectory accuracy weight is equal to the system range accuracy index multiplied by the system azimuth accuracy index multiplied by the system pitch accuracy index.

[0136] In a specific embodiment of the present invention, the track performance evaluation index U is obtained based on the track motion characteristic weight A, the error correlation rate weight B, and the track accuracy weight C. The larger the U value, the better the track performance, U = A*B*C*255.

[0137] The trajectory motion characteristic weight A is mainly composed of distance, bearing, and elevation angle characteristics;

[0138] A = F R *F Az *F El

[0139] 1) Based on the change in target velocity, F R The value can be:

[0140]

[0141] 2) Based on the change in azimuth angle of the track relative to the radar, F Az The value can be:

[0142]

[0143] 3) Based on the change in target altitude, F El The value can be:

[0144]

[0145] By matching with secondary radar track data, the false correlation rate μ is calculated, and the weight B is set based on the false correlation rate.

[0146]

[0147] The trajectory accuracy weight C is mainly composed of range, azimuth, and elevation angle, and is calculated by comparing it with secondary radar trajectory data to determine the range accuracy P. R Azimuth accuracy P Az and pitch accuracy P El ;

[0148] C = Q R *Q Az *Q El

[0149] 1) If the system distance accuracy index is set to R_index, then Q R for:

[0150]

[0151] 2) Set the system azimuth accuracy index to Az_index, then Q Az for:

[0152]

[0153] 3) Set the system pitch accuracy index to El_index, then Q El for:

[0154]

[0155] In some embodiments of the present invention, obtaining the optimal trajectory data of the target point based on the aforementioned levels includes:

[0156] When the trajectory performance evaluation index is at the first level, the first weighted convergence coefficient and the first point-of-flight correlation gate are obtained. Based on the first weighted convergence coefficient and the first point-of-flight correlation gate, the optimal trajectory data of the target point is obtained. The first weighted convergence coefficient is obtained by increasing the weighted convergence coefficient of the point-of-flight correlation gate, and the first point-of-flight correlation gate is obtained by decreasing the value of the point-of-flight correlation gate.

[0157] When the trajectory performance evaluation index is at level two, the second point-linked wave gate, the first filter parameter, and the second filter parameter are obtained. Based on the second point-linked wave gate, the first filter parameter, and the second filter parameter, the optimal trajectory data of the target point is obtained. The second point-linked wave gate is obtained by increasing the value of the point-linked wave gate. The first filter parameter is obtained by increasing the value of the α parameter in the α-β filtering algorithm. The second filter parameter is obtained by increasing the value of the β parameter in the α-β filtering algorithm.

[0158] When the trajectory performance evaluation index is at level three, the third point-related wave gate, the third filtering parameter, and the fourth filtering parameter are obtained. Based on the third point-related wave gate, the third filtering parameter, and the fourth filtering parameter, the optimal trajectory data of the target point is obtained. The third point-related wave gate is obtained by adjusting the value of the point-related wave gate. The third filtering parameter is obtained by adjusting the value of the α parameter in the α-β filtering algorithm. The fourth filtering parameter is obtained by adjusting the value of the -β parameter in the α-β filtering algorithm.

[0159] In a specific embodiment of the present invention, the track performance is divided into four levels: excellent, good, average, and poor, based on the target track performance evaluation results. Track optimization is performed according to different levels to ensure track quality while achieving the best system performance.

[0160] When the performance evaluation index of the track is at the first level, it means that the track performance is excellent. That is, when U≥220, the weight convergence coefficient and the size of the bandgap gate at the contraction point can be increased, the system resource consumption can be reduced, and the system detection performance can be improved.

[0161] When 170≤U<220, it indicates that the trajectory performance evaluation index is good, and the current point navigation process can be locked.

[0162] When the track performance evaluation index is at level two, it means that the track performance is average. That is, when 127≤U<170, the size of the point navigation correlation gate and the track filtering parameters can be increased to improve the track performance evaluation level.

[0163] When the track performance evaluation index is at level three, it indicates that the track performance is poor. That is, when U < 127, it is necessary to comprehensively adjust the point-to-point correlation gate and filtering parameters to ensure that the track accuracy meets the system requirements.

[0164] To better implement the trajectory optimization method in this embodiment of the invention, based on a trajectory optimization method, correspondingly, as follows: Figure 2 As shown, this embodiment of the invention also provides a trajectory optimization device, a trajectory optimization device 200 comprising:

[0165] The data acquisition unit 201 is used to acquire the initial spot data for each scanning cycle of the radar;

[0166] The first dot data acquisition unit 202 is used to preprocess the initial dot data to obtain the first dot data;

[0167] Point-to-point associative initial gate setting unit 203, obtains the point-to-point associative initial gate;

[0168] The dot-matter association determination unit 204 is used to calculate the spatial relative displacement between the first dot-matter data of adjacent scanning cycles in each scanning cycle of the radar, and determine whether the dot-matter association of adjacent scanning cycles is successful based on the spatial relative displacement and the dot-matter association initial gate.

[0169] The point navigation association gate setting unit 205 is used to obtain the initial prediction value of point-to-point association based on the spatial relative displacement when it is determined that the point-to-point association of adjacent scanning cycles is successful, obtain the point navigation association gate, set the weight of the point navigation association gate based on the r-order convergence idea to obtain the optimized point navigation association gate, and set the optimized point navigation association gate based on the initial prediction value of point-to-point association to obtain the first point navigation association gate.

[0170] The track data acquisition unit 206 is used to filter the target point track based on the α-β filtering algorithm to obtain the track data of the target point track when it is determined that the target point track falls into the first point track association gate;

[0171] The optimal trajectory data acquisition unit 207 is used to optimize the trajectory data of the target point to obtain the optimal trajectory data of the target point.

[0172] The apparatus 200 for forwarding a remote server interface provided in the above embodiments can implement the technical solutions described in the above embodiments of the method for forwarding a remote server interface. The specific implementation principles of each module or unit can be found in the corresponding content of the above embodiments of the method for forwarding a remote server interface, and will not be repeated here. Those skilled in the art will understand that all or part of the process of the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.

[0173] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

Claims

1. A method of track optimization, characterized by, The method comprises the following steps: acquiring initial track data of each scanning cycle of a radar; preprocessing the initial track data to obtain first track data; acquiring an initial point-point association gate; calculating a spatial relative displacement between the first track data of adjacent scanning cycles in each scanning cycle of the radar, and determining whether the track of the adjacent scanning cycle is successfully associated according to the spatial relative displacement and the initial point-point association gate; In a case where the track association of the adjacent scan cycle is determined to be successful, an initial prediction value of point-point association is obtained according to the amount of spatial relative displacement, a point-horizon association gate is obtained, a weight of the point-horizon association gate is set based on an r-order convergence thought to obtain an optimized point-horizon association gate, and a first point-horizon association gate is obtained based on the initial prediction value of the point-point association and the optimized point-horizon association gate, wherein the r-order convergence thought is calculated as follows: r r-order convergence thought calculation: wherein, W(p 0 ) represents the current weight value used for the track, W(p 0+1 ) represents the weight value of the next point track association gate, r represents the convergence coefficient; L represents the target track length; in a case where it is determined that the target track falls within the first point-track association gate, performing filtering processing on the target track based on an α-β filtering algorithm to obtain track data of the target track; performing track optimization on the track data of the target track to obtain optimal track data of the target track.

2. The track optimization method of claim 1, wherein, The initial point-point association gate comprises a distance initial gate, an azimuth angle initial gate and an elevation angle initial gate, and the distance initial gate is expressed as: The azimuth angle initial gate is expressed as: The elevation angle initial gate is expressed as: In the formula, R_thr0 represents the initial range gate; tgtV_max represents the maximum target detection speed; Frame_updateTime represents the inter-frame update time; 0 represents the initial azimuth gate; Az_max(i) represents the azimuth change of each range segment; Az_index represents the system azimuth measurement accuracy;​ El_thr0 denotes the initial elevation beam for the elevation angle; W0 , W1 and W2 denotes a weighting value; El_BeamWidth denotes the elevation beam width; K is a constant.

3. The track optimization method of claim 1, wherein, The spatial relative displacement between the initial track data of adjacent scanning cycles in each scanning cycle of the radar comprises: a relative displacement of distance between the initial track data of adjacent scanning cycles, a relative displacement of azimuth angle between the initial track data of adjacent scanning cycles and a relative displacement of elevation angle between the initial track data of adjacent scanning cycles.

4. The track optimization method of claim 1, wherein, The determination of whether the track of the adjacent scanning cycle is successfully associated according to the spatial relative displacement and the initial point-point association gate comprises: when the spatial relative displacement is smaller than the initial point-point association gate, it is determined that the track of the adjacent scanning cycle is successfully associated.

5. The track optimization method of claim 1, wherein, when the length of the target track is greater than a preset value, the expression of the weight of the point-track association gate is: In the formula, W(p0) represents the current weighting value used by the target track; and W(p0+1) represents the weighting value of the next point track associated with the gating window. r represents a convergence coefficient; L represents the length of the target track.

6. The track optimization method of claim 1, wherein, The filtering processing on the target track based on the α-β filtering algorithm to obtain the track data of the target track comprises: obtaining a filtering value of the track data of a current point of the target track based on the α-β filtering algorithm according to the track data of the current point of the target track; obtaining distance prediction values, azimuth angle prediction values and elevation angle prediction values of a next frame of the current point according to the filtering value of the track data of the current point; determining the track data of the target track according to the distance prediction values, the azimuth angle prediction values and the elevation angle prediction values of the next frame of the current point.

7. The track optimization method of claim 1, wherein, The track optimization on the track data of the target track to obtain the optimal track data of the target track comprises: obtaining a track performance evaluation index according to a track motion characteristic weight value, a false association rate weight value and a track precision weight value; dividing the track into several levels according to the track performance evaluation index; obtaining the optimal track data of the target track according to the several levels.

8. The track optimization method of claim 7, wherein, The track motion characteristic weight value is equal to a target speed change value multiplied by a target azimuth angle change value multiplied by a target height change value; the false association rate weight value is obtained based on matching of radar track data and secondary radar track data; and the track precision weight value is equal to a system distance precision index multiplied by a system azimuth precision index multiplied by a system elevation precision index.

9. The track optimization method of claim 7, wherein, The obtaining of the optimal track data of the target track according to the several levels comprises: When the track performance evaluation index is the first level, a first weight convergence coefficient and a first point-horizon association gate are obtained, and optimal track data of the target track are obtained based on the first weight convergence coefficient and the first point-horizon association gate, the first weight convergence coefficient being obtained by increasing a weight convergence coefficient of the point-horizon association gate, and the first point-horizon association gate being obtained by decreasing a value of the point-horizon association gate; When the track performance evaluation index is the second level, a second point-horizon association gate, a first filter parameter and a second filter parameter are obtained, and the optimal track data of the target track are obtained based on the second point-horizon association gate, the first filter parameter and the second filter parameter, the second point-horizon association gate being obtained by increasing a value of the point-horizon association gate, the first filter parameter being obtained by increasing a value of an α parameter in the α-β filter algorithm, and the second filter parameter being obtained by increasing a value of a β parameter in the α-β filter algorithm; When the track performance evaluation index is the third level, a third point-horizon association gate, a third filter parameter and a fourth filter parameter are obtained, and the optimal track data of the target track are obtained based on the third point-horizon association gate, the third filter parameter and the fourth filter parameter, the third point-horizon association gate being obtained by adjusting a value of the point-horizon association gate, the third filter parameter being obtained by adjusting a value of the α parameter in the α-β filter algorithm, and the fourth filter parameter being obtained by adjusting a value of the β parameter in the α-β filter algorithm.

10. A track optimization device, characterized by The method comprises the following steps: a data acquisition unit is configured to acquire initial track data of each scanning period of a radar; a first track data acquisition unit is configured to pre-process the initial track data to obtain first track data; a point-point association initial gate setting unit is configured to obtain a point-point association initial gate; a track association judgment unit is configured to calculate a spatial relative displacement between the first track data of adjacent scanning periods in each scanning period of the radar, and determine whether the track association of the adjacent scanning periods is successful according to the spatial relative displacement and the point-point association initial gate; The point-horizon association wave gate setting unit is configured to, in a case where the track association of the adjacent scan cycle is determined to be successful, obtain an initial prediction value of point-point association according to the spatial relative displacement amount, obtain a point-horizon association wave gate, set a weight of the point-horizon association wave gate based on an r-order convergence thought to obtain an optimized point-horizon association wave gate, and set the optimized point-horizon association wave gate based on the initial prediction value of the point-point association to obtain a first point-horizon association wave gate, wherein the r The r-order convergence thought is calculated as follows: wherein W(p 0 ) represents the current use weighting value of the track, W(p 0+1 ) represents the weighting value of the next point track association gate, r represents the convergence coefficient; L represents the target track length; a track data acquisition unit is configured to perform filter processing on the target track based on an α-β filter algorithm to obtain track data of the target track when it is determined that the target track falls within the first point-horizon association gate; an optimal track data acquisition unit is configured to perform track optimization on the track data of the target track to obtain optimal track data of the target track.

Images