A target search route planning method based on cumulative probability optimization

Abstract

The present invention discloses a target search route planning method based on cumulative probability optimization. The method comprises the following steps: 1) selecting multiple search formations suitable for the search task from a formation route type library, then delineating a grid to generate an effective solution space based on the total number of search platforms in the multiple search formations, the target search range required to complete the given task, the turning radius of the search platform, and the detection range; 2) obtaining the detection probability density distribution of the search platform; 3) obtaining the probability density distribution of the target; 4) calculating the time-varying detection probability of each waypoint in the i-th solution space; and 5) obtaining a target search route planning scheme based on a specified time and an integrated cumulative probability weight according to a priority index F. The present invention utilizes an effective planning strategy solution space and an integrated probability model to rapidly obtain a strategy suitable for an unmanned surface search platform to perform the task, thereby ensuring a high cumulative detection probability and simplifying the search procedure of the unmanned surface search platform.

Description

Technical Field

[0001] The present invention relates to intelligent route planning technology, and in particular to a target search route planning method based on cumulative probability optimization. Background Art

[0002] Traditional target search route planning methods are aimed at single-task situations. Commonly used target search route planning methods include optimization algorithms such as A* heuristic algorithm, genetic algorithm, and particle swarm algorithm. These optimization algorithms seek local convergence optimal solutions through long-term loop iterations. The routes contained in the solutions are complex and difficult to execute, and increase the difficulty of platform control. However, it is difficult for an unmanned surface search platform to navigate completely along the trajectory of the optimal solution. Moreover, since unmanned surface search platforms have a certain search width, traditional route planning methods are difficult to adapt to search tasks with high timeliness requirements and simple collaboration. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a target search route planning method based on cumulative probability optimization in response to the defects in the existing technology.

[0004] The technical solution adopted by the present invention to solve the technical problem is: a target search route planning method based on cumulative probability optimization, comprising the following steps:

[0005] 1) Based on the search mission, multiple search formations suitable for the mission are selected from the formation route type library. Then, based on the total number of search platforms in the multiple search formations, the target search range required to complete the mission, the turning radius of the search platforms, and the detection range, a grid is delineated to generate an effective solution space (including parameters such as transverse and longitudinal ship spacing, turning transition length, number of search platforms, and time-varying waypoints). The selection of transverse and longitudinal ship spacing should consider detection capability and interference between platform noise; the turning transition length refers to the distance between the platform's route after turning and the original route, focusing on the platform's search capability and mission completion; the number of search platforms should consider the total number of search platforms and the mission completion; and time-varying waypoints are preset waypoints based on different solution spaces, with the waypoint spacing generally being the minimum turning radius.

[0006] 2) Obtain the detection probability density distribution of the search platform, calculated as follows:

[0007] Define the instantaneous detection probability density pk(x,y,R) when the jth surface search platform detects a target at any point (x,y) at the kth time-varying waypoint. x ,R y )for

[0008]

[0009] Where, σ=RT / 3, RT is the effective working distance of the detection equipment in the search platform; san is the search sector angle of the search platform; (R x ,R y ) is the kth time-varying waypoint coordinate;

[0010] When multiple search platforms act simultaneously, their total probability density distribution is the sum of the detection probability densities of all search platforms;

[0011] 3) Obtain the probability density distribution of the target, calculated as follows:

[0012] Taking the origin of the effective solution space as the origin, the target probability density of any point (x, y) in the integral region G when the uniform distribution is defined is

[0013]

[0014] Where G is the integral region used to calculate the target appearance probability based on the target probability density distribution during the search process. It is a dynamic region that changes with time and has the time-varying waypoint of the first search platform as its origin. The side length of the integral region G is XL = XH = (n + 1) * max(RC), where max(RC) is the maximum value in the set of minimum turning radii of each surface search platform, and n is the number of platforms participating in the search.

[0015] 4) Grid the integration area G, which is generally divided into equal intervals. The grid spacing can be adjusted at any time to obtain the best calculation time efficiency. Then, the integration area is delineated according to different solution spaces, and the time-varying detection probability of each waypoint in the i-th solution space is cyclically calculated. and the time distribution t(i,k);

[0016] 5) Target search route planning scheme generation

[0017] According to the priority index F, the time and integral cumulative probability weight w(i) are specified; the corresponding parameters of the i-th candidate solution corresponding to the maximum value of w(i) are used as the target search route planning scheme.

[0018] According to the above scheme, the search formations in step 1) include: single horizontal team, flank front single horizontal team, reverse V-shaped team, and V-shaped team.

[0019] According to the above scheme, the cumulative probability weight w(i) in step 5) is:

[0020]

[0021] The beneficial effects produced by the present invention are:

[0022] 1. Comprehensively consider the constraints such as mission area, search platform performance, and hydrological conditions to achieve the optimal configuration of the search platform's target search route;

[0023] 2. Utilize the effective planning strategy solution space and integral probability model to quickly obtain a planning strategy suitable for the surface unmanned search platform to carry out the mission, which can not only ensure a high cumulative discovery probability, but also reduce the number of turns of the surface unmanned search platform and simplify the search procedure. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0025] Figure 1 is a flow chart of a method according to an embodiment of the present invention;

[0026] Figure 2 This is a schematic diagram of a typical formation route type library according to an embodiment of the present invention;

[0027] Figure 3 2. Schematic diagram of a single-row search for an effective solution space according to an embodiment of the present invention;

[0028] Figure 4 2. Schematic diagram of an effective solution space for reverse human-shaped search according to an embodiment of the present invention;

[0029] Figure 5 2 is a schematic diagram of the division of the time integration region G according to an embodiment of the present invention;

[0030] Figure 6 2 is a schematic diagram of time-varying detection probability according to an embodiment of the present invention. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0032] like Figure 1 As shown, a target search route planning method based on cumulative probability optimization includes the following steps:

[0033] First, a grid is drawn based on the number of search platforms, the given area or target location, the turning radius, and the detection distance to generate the corresponding effective solution space (such as time-varying waypoints) for the selected formation route. Then, the probability density distribution of the target and detection equipment is generated according to the target situation and the detection equipment indicators (detection distance and search sector). Finally, the target search path of the surface unmanned search platform is planned and generated by cyclically calculating the time-varying detection probability and the cumulative detection probability, and optimizing the time and probability weights.

[0034] The specific implementation steps are as follows:

[0035] (1) Generation of effective solution space.

[0036] like Figure 2 As shown in the figure, multiple search formations suitable for the task are selected from the formation route type library according to the search task, such as Figure 3 and Figure 4 As shown, the effective solution space is generated by delineating a grid based on the number of search platforms, the sea area or suspicious target range required to complete the given mission, the turning radius, and the detection distance. The effective solution space includes parameters such as the transverse and longitudinal spacing of ships, the turning transition length, the number of search platforms, and time-varying waypoints.

[0037] Among them, the selection of the lateral and longitudinal spacing of ships should take into account the detection capability and the mutual interference between platform noises; the turning transition length refers to the distance between the route after the platform turns and the original route, focusing on the platform search capability and mission completion; the number of search platforms should consider the total number of search platforms and the completion of the mission; time-varying waypoints are waypoints preset according to different solution spaces, and the waypoint interval is generally the minimum turning radius.

[0038] The following uses a single horizontal formation as an example to illustrate how to generate an effective solution space.

[0039] When using unmanned surface search platforms for target search, the selection of the horizontal spacing L (j = 1, 2, ..., n-1) and the vertical spacing a of each search platform should take into account the mutual interference between the detection equipment and the self-noise. The value range of L (j) can be agreed based on experience and the performance of the detection equipment; a (j = 1, 2, ..., n-1) for a single horizontal team is fixed to 0. This generates an effective solution space for single horizontal team search as follows: Figure 3 As shown, the formation parameters include the ship lateral spacing L (meters), the ship longitudinal spacing a (meters), the turning transition length T (meters), the number of search platforms num, and the waypoint interval jg (meters).

[0040] (2) Calculation of the probability density distribution of the search platform.

[0041] The effective range of the search platform's detector equipment refers to the distance within a given sea area at which the probability of detecting any target within a designated search sector is 50%. This range is variable and depends on factors such as the performance of the search platform's detector equipment, hydrological conditions, the ship's speed, and the target's heading, speed, and reflectivity. Therefore, the effective operating range of the detector equipment should be measured before entering the search area.

[0042] An important objective criterion for evaluating sensor performance is the detection probability, that is, the probability of a sensor detecting a target at a certain distance in a given environment. This probability is called the instantaneous detection probability. Assuming that each completed detection is a random variable, independent of each other, and satisfies the normal distribution, it can be assumed that the instantaneous detection probability at the maximum detection distance of the detection equipment under given current hydrological conditions reaches 50%. Define the instantaneous detection probability density of the jth surface unmanned search platform detecting a target at any point (x, y) at the kth time-varying waypoint as

[0043]

[0044] Where, σ=RT / 3, RT is the effective working distance of the measuring equipment; san is the search sector angle; (R x ,R y ) is the kth time-varying waypoint coordinate.

[0045] It can be seen from the above formula that the detection probability density of a detection device is a two-dimensional distribution. When multiple detection devices act simultaneously, the total probability density distribution is the sum of the detection probability densities of all detection devices.

[0046] (3) Calculation of the probability density distribution of the target.

[0047] Taking the origin of the effective solution space as the origin, the target probability density of any point (x, y) in the integral region G when the uniform distribution is defined is

[0048]

[0049] Among them, G is the integral area for calculating the target occurrence probability based on the target probability density distribution during the search process, and is the dynamic area that changes with time (such as Figure 5 As shown in Figure 2, the time-varying waypoint of the first search platform is taken as the origin; the side length of the integration region G is XL = XH = (n + 1) * max(RC), where max(RC) is the maximum value in the set of minimum turning radii of each surface search platform, and n is the number of platforms participating in the search;

[0050] (4) Calculation of integrated cumulative detection probability.

[0051] First, the integration region G is divided into equal intervals, and then the time-varying detection probability of each waypoint in the i-th solution space is calculated cyclically. And the time distribution t(i,k) (unit, hour), the time-varying detection probability results are as follows Figure 6 shown.

[0052] (5) Generation of target search route planning scheme

[0053] The time and integral cumulative probability weight w(i) are specified according to the priority index F, as shown in the following formula.

[0054]

[0055] The parameter of the i-th solution to be selected corresponding to the maximum value of w(i) is the target search route planning scheme.

[0056] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention.

Claims

1. A target search route planning method based on cumulative probability optimization, characterized by: The following steps are involved: 1) Based on the search mission, multiple search formations suitable for the mission are selected from the formation route type library. Then, based on the total number of search platforms in the multiple search formations, the target search range required to complete the given mission, the turning radius and detection range of the search platforms, a grid is delineated to generate an effective solution space. The effective solution space includes parameters such as the transverse and longitudinal spacing of the ships, the turning transition length, the number of search platforms, and time-varying waypoints. The turning transition length refers to the distance between the platform's route after turning and the original route; the time-varying waypoint is a waypoint preset according to the search mission; 2) Obtain the detection probability density distribution of the search platform, calculated as follows: Define the instantaneous detection probability density pk(x,y,R) when the jth surface search platform detects a target at any point (x,y) at the kth time-varying waypoint. x ,R y )for Where, σ=RT / 3, RT is the effective working distance of the detection equipment in the search platform; san is the search sector angle of the search platform; (R x ,R y ) is the kth time-varying waypoint coordinate; When multiple search platforms act simultaneously, their total probability density distribution is the sum of the detection probability densities of all search platforms; 3) Obtain the probability density distribution of the target, calculated as follows: Taking the origin of the effective solution space as the origin, the target probability density of any point (x, y) in the integral region G when the uniform distribution is defined is Where G is the integral region used to calculate the target appearance probability based on the target probability density distribution during the search process. It is a dynamic region that changes with time and has the time-varying waypoint of the first search platform as its origin. The side length of the integral region G is XL = XH = (n + 1) * max(RC), where max(RC) is the maximum value in the set of minimum turning radii of each surface search platform, and n is the number of platforms participating in the search. 4) Grid-divide the integration area G, then delineate the integration area according to different solution spaces, and cyclically calculate the time-varying detection probability of each waypoint in the i-th solution space and the time distribution t(i,k); 5) Target search route planning scheme generation According to the priority index F, the time and integral cumulative probability weight w(i) are specified; the corresponding parameters of the candidate solution corresponding to the maximum value of w(i) are used as the target search route planning scheme.

2. The target search route planning method based on cumulative probability optimization according to claim 1 is characterized in that: The searched formations in step 1) include: single horizontal formation, flank front single horizontal formation, reverse V formation, and V formation.

3. The target search route planning method based on cumulative probability optimization according to claim 1 is characterized in that: The cumulative probability weight w(i) in step 5) is:

Images