A method for deducing a target sea voyage path of a UAV based on a Venn diagram

Abstract

The application discloses a kind of based on venne diagram UAV target sea navigation path deduction method, for the island and reconnaissance station and each space constraint region involved in UAV target sea navigation construction space constraint region set, on this basis, construct venne diagram model, sea space region is expressed with venne diagram unit and its adjacency relationship, so as to divide UAV evasion and normal navigation space;Then based on distance constraint condition screening UAV target possible unit, and then the recursion search of unit path that UAV target possible unit passes;Finally output the unit path set that satisfies space constraint condition.Compared with traditional path planning method, the application can deduce multiple navigation paths that satisfy time constraint condition, more comprehensive, better meet the actual demand of task;It can be based on venne diagram on demand division according to sea space, effectively improve the exploration efficiency of navigation path, greatly save the cost of calculation, flexible adaptation multi-granularity space constraint.

Description

Technical Field

[0001] This invention relates to the field of UAV target path deduction technology, and in particular to a method for deducing the maritime navigation path of UAV targets based on the Vinyk diagram. Background Technology

[0002] With the development of artificial intelligence technology, unmanned aerial vehicles (UAVs) have been applied in many military and civilian fields, such as transporting emergency supplies like medical rescue packages, maritime patrol and surveillance, and maritime search and rescue. UAV path planning has become a hot research topic. Researchers have conducted extensive studies in the field of UAV path planning, proposing methods such as the artificial potential field method (see CHEN Yongbo et al., UAV path planning using artificial potential field method updated by optimal control theory), the A* algorithm (He Yan. UAV path planning based on dynamic weighted A* algorithm, Journal of Hebei University of Science and Technology, 2018), the particle swarm algorithm (Xiong Huajie et al., UAV path planning method based on improved particle swarm algorithm, Computer Measurement & Control, 2020), the fast expanding random tree algorithm (Wu Xiaojing et al., UAV path planning algorithm based on dynamic step size BI-RRT, Journal of Hebei University of Science and Technology, 2019), the D* algorithm (Zhao Juan. UAV path planning strategy based on heuristic point-guided D* algorithm expansion, Mechanical Design & Manufacturing, 2020), the artificial fish swarm algorithm (Xu Jiangbo and Liu Linlan. UAV 3D path planning based on improved artificial fish swarm algorithm, Computer Engineering & Design, 2019), and the locust algorithm (Cheng Zexin et al., UAV 3D path planning based on locust algorithm, Flight Mechanics, 2019). The aforementioned studies aimed to explore the shortest path that satisfies obstacle constraints, focusing on improving path planning efficiency. However, these studies treated the shortest path as the optimal solution, neglecting the different requirements of UAV paths under various mission scenarios. For example, in maritime patrol, aerial reconnaissance, and maritime search and rescue missions, the coverage area of ​​the UAV path is more meaningful than its length, provided that timeliness requirements are met. Therefore, the shortest path is not necessarily the optimal solution in different mission scenarios. This invention focuses on the study of UAV path derivation under distance constraints. Summary of the Invention

[0003] Purpose of the Invention: The technical problem this invention aims to solve is that existing path search technologies only focus on finding the shortest path while ignoring the possibility of multiple path choices. This invention proposes a method for extrapolating the maritime navigation path of UAV targets based on Venn diagrams, capable of searching for all possible paths that satisfy distance constraints, including:

[0004] To ensure the safety and stealth of UAVs, UAV target routes are typically constrained by both temporal and spatial factors: firstly, time constraints, such as the maximum travel time; and secondly, spatial constraints. Specifically, UAV target routes need to avoid obstacles on the sea surface, such as islands, as well as maritime reconnaissance stations, such as microwave radars and infrared sensors deployed on the sea surface.

[0005] The present invention specifically includes the following steps:

[0006] Step 1: Construct a set of spatially constrained regions;

[0007] Step 2: Initialize the Vinyasa diagram seed set;

[0008] Step 3: Construct the seed set of the Venn diagram;

[0009] Step 4: Construct the Venn diagram;

[0010] Step 5: Initialize the adjacency matrix of the Vinograph;

[0011] Step 6: Calculate the adjacency distance in the Venn diagram;

[0012] Step 7: Update the adjacency matrix of the Venn diagram;

[0013] Step 8: Filter units based on distance constraints;

[0014] Step 9: Initialize the path set: Initialize the path set pSet to an empty set.

[0015] Step 10: Determine if the cell set vSet is empty. If it is, skip to step 13; otherwise, continue to step 11.

[0016] Step 11: Initialize the current path;

[0017] Step 12: Construct the path recursive search function;

[0018] Step 13: Run the recursive path search function: using the current path sequence curPath = [v start The path length at the previous time step, prePathLen = 0, is used as input to call the recursive function PathSearch(curPath, prePathLen);

[0019] Step 14: Output the path set pSet.

[0020] Step 1 includes: setting the starting point P of the known UAV target's sea route. start Latitude and longitude (X) start ,Y start Destination point P end Latitude and longitude (X) end,Y end Given the UAV target's fastest flight speed maxV and the longest flight time Δt required to complete its navigation path, and assuming that the distances between the islands, maritime reconnaissance stations, and other spatially constrained areas that the UAV target needs to avoid are all relatively large, it is necessary to deduce the set of maritime routes pSet for the UAV target that satisfies the spatiotemporal constraints. Then, it can be deduced that the longest maritime distance for the UAV target is maxd = maxV × Δt.

[0021] Obtain the barycenter coordinates o of the various spatially constrained areas i, including islands and reconnaissance stations, involved in the UAV target's maritime navigation. i Calculate the minimum radius r of the circle that can cover the spatially constrained region, using the centroid coordinates as the center. i The number of spatially constrained regions is denoted as n. area Constructing a region consisting of all spatial constraints (o i ,r i The set of spatially constrained regions formed by ) |ResSet|=n area ;

[0022] Step 2 includes: initializing the Vinyson map seed set SeedSet to empty: SeedSet = {}, and setting the number of edges k of the Vinyson map cells corresponding to the constraint regions.

[0023] Step 3 includes: for each spatial constraint region i in the set of spatial constraint regions ResSet, setting its centroid o i Add it as a seed to the Vinyson diagram seed set SeedSet: With the centroid o of spatially constrained region i i Let r be the center of the circle, and its minimum radius be r. i Twice the radius R = 2r i The outer perimeter of the constrained region i is formed by the circle C′. i Take outsourced garden C′ i k equal parts P k(i-1)+1 P k(i-1)+2 ,…,P ik Add the k equally divided points as seeds to the Vinonic diagram seed set. Assume there are no other seeds within the outer circle, and the number of seeds in the seed set SeedSet is N = (k+1)n. area If there are no other seeds in the outlying garden, then the center of gravity is o. i Center and minimum radius r i A circle with radius ω must belong to the seed o in the Venn diagram. i The unit it belongs to is because of the outsourced garden C′ i The k-partition points are the distances from the seed o. i The most recent seed, k-partition seed and seed o iThe perpendicular bisector must be about the centroid o i The minimum circle tangent centered at the center of the circle, the constrained region corresponding to the Vinograph element must contain the centroid o. i Center and radius r i The smallest circle, i.e. containing the spatially constrained region i (Voronoi graph construction reference Jan Oliver Vallgrün, Voronoi Graph Matching for RobotLocalization and Mapping, Transactions on Computational Science IX, 2010).

[0024] Step 4 includes: constructing a Venn map Vor(SeedSet) based on N seeds in the Venn map seed set SeedSet, and dividing the plane into N Venn map units v. j 1≤j≤N, each cell contains one and only one seed from the Vinyau diagram seed set SeedSet. j 1≤j≤N, satisfying that the distance between a point in a cell and its corresponding seed is shorter than the distance between points in the cell and other seeds. The edges of the Vor(SeedSet) are the perpendicular bisectors or rays of a pair of seeds. The vertices in the Vor(SeedSet) correspond to the centers of the circumcircles of the three seeds, and the circumcircles do not contain any seeds from the SeedSet (VorOi graph construction reference Jan Oliver Vallgrün, Voronoi Graph Matching for Robot Localization and Mapping, Transactions on Computational Science IX, 2010).

[0025] Step 5 includes: for each cell v in the Venn diagram Vor(SeedSet) j Calculate the number of other adjacent units m. j , building unit v j m j ×m j ∑-dimensional adjacency matrix j initialize its elements to an empty set: 1≤f,g≤m j Where f and g represent unit v j The adjacency unit index, v f and v g Representation unit v j Two adjacent units;

[0026] In the initialization of the adjacency relation matrix of the Vinograph, each element v jThe number of adjacent units m j Equal to the unit v j The number of sides.

[0027] Step 6 includes: based on unit v j m j ×m j ∑-dimensional adjacency matrix j Calculate the pairwise adjacent units of v f and v g Minimum diameter between With the maximum diameter The smallest diameter This refers to unit v j The inner can connect its adjacent unit v f With v g Shortest straight-line distance, maximum diameter This refers to unit v j The inner can connect its adjacent unit v f With v g The longest straight-line distance between two adjacent units v j With v f The edge between them is denoted as l. j,f Two adjacent units v j With v g Let the edge between them be denoted as l j,g The calculation process is as follows:

[0028]

[0029] Where min represents taking the minimum value, l j,f and l j,g "are connected" indicates that edge l j,f With edge l j,g They are connected;

[0030]

[0031] Where dist(x,y) represents the straight-line distance between points x and y, when edge l j,f With edge l j,g Minimum diameter at intersection With the maximum diameter All are 0, l j,f .X1 and l j,f .X2 represent edges l respectively j,f The two endpoints, l j,g .X1 and l j,g .X2 represent edges l respectively j,g The two endpoints, MinDist(l j,f .X1,l j,g) represents edge l j,f endpoint X1 to edge l j,g The shortest distance, MaxDist(l j,f .X1,l j,g ) represents edge l j,f endpoint X1 to edge l j,g The longest distance, will unit v j Pairs of adjacent units v f -v g Minimum diameter between With the maximum diameter The average value is used as the unit v j For the intermediate time unit v f With v g Distance between j (v f ,v g ):

[0032]

[0033] In the calculation of adjacency distance in a Vinograph, in cell v j The inner can connect its adjacent unit v f With v g The adjacency distance is equal to the length of the line connecting the midpoints of its two sides, because the shortest straight distance and the longest straight distance correspond to the lengths of the lines connecting the endpoints of its two sides, respectively.

[0034] Step 7 includes: updating unit v j m j ×m j ∑-dimensional adjacency matrix j , the adjacency relation matrix ∑ j Each off-diagonal element in the array is updated to be the corresponding minimum diameter. (Formula (1)) and maximum diameter The set formed by (formula (2)): ∑ j [v f ,v g ] = dist j (v f ,v g )(Formula (3)), f≠g, 1≤f, g≤m j , where v f With v g Represents unit v j Two adjacent adjacent cells, dist j (v f ,v g ) indicates that the unit v j For the intermediate time unit v f With v gThe distance between them (as defined in formula (3)).

[0035] Step 8 includes: for each unit v j Calculate its relationship with the starting point P. start With destination P end The minimum distance and MinSumDist(v j ,P start ,P end ):

[0036]

[0037] Among them, X K ∈v j Representation unit v j The point in the middle;

[0038] Calculate the maximum flight distance maxd of the UAV target:

[0039] maxd=maxV×Δt (5)

[0040] Select the minimum distance and MinSumDist(v) in the non-spatial constrained region j ,P start ,P end For cells whose maximum flight distance is less than or equal to maxd, construct an optional cell set vSet, as shown in equation (6):

[0041] vSet = {v j |MinSumDist(v j ,P start ,P end )≤maxd} (6)

[0042] In step 8, the cell selection based on distance constraints, for each cell v j Calculate its relationship with the starting point P. start With destination P end The minimum distance and MinSumDist(v j ,P start ,P end This method filters out cells that do not meet the distance constraint maxd, which can significantly reduce the computational cost of subsequent recursive path search.

[0043] Step 11 includes: initializing the current path sequence curPath to start from point P. start Unit v start The constructed sequence, i.e., curPath = [v start ], where P start ∈v start v start∈vSet, the path length prePathLen of the previous time step is initialized to zero: prePathLen = 0;

[0044] Step 12 includes: constructing a recursive path search function PathSearch(curPath, prePathLen), whose recursive parameters include: a path from the start time t0 to the current time t. c The sequence of Vinograph units traversed (assuming it consists of two or more Vinograph units) is represented as curPath = [v t0 ,v t1 ,...,v t(c-1) ,v tc ], from the initial time t0 to the previous time point t c-1 The path length traversed, i.e., the path length at the previous time step, is denoted as prePathLen, where v tc This indicates the unit reached by the current path sequence, i.e., the current unit, and then the following steps are performed:

[0045] Step 12-1: Calculate the shortest distance from the current path sequence to the destination, curPathLen, which is equivalent to calculating the path length prePathLen from the previous time step and the distance between edge l. t(c-1),tc Sum of minimum distances to the destination:

[0046] curPathLen=prePathLen+min(dist(l t(c-1),tc .midpt,P end ))(7)

[0047] Where dist(l t(c-1),tc .midpt,P end ) indicates from edge l t(c-1),tc From the midpoint to the destination P end The distance;

[0048] Step 12-2: If the shortest distance from the current path sequence to the destination, curPathLen, exceeds the longest flight distance, maxd (i.e., curPathLen > maxd), it means that the current path does not meet the longest flight distance requirement. Skip to step 12-8; otherwise, continue to step 12-3.

[0049] Step 12-3: If the current path has reached the destination P end The unit in question, namely P end ∈v cur Add the current path to the path set pSet = pSet∪{curPath}, then jump to step 12-8; otherwise, continue to step 12-4.

[0050] Step 12-4: Based on the current cell v in the adjacency matrix of the Vinograph... tc The adjacency matrix ∑ tc Find unit v tc m tc Candidate adjacent cells that belong to the cell set vSet but not to the current path curPath Indicates the m-th tc If there are m candidate adjacent units, and the number of units is greater than zero, then... tc If the value is greater than 0, initialize h = 1 and continue to step 12-5; otherwise, skip to step 12-8.

[0051] Step 12-5: Determine if the adjacent cell index satisfies h≤m tc If yes, continue to step 12-6; otherwise, skip to step 12-8.

[0052] Step 12-6: For the unit v at the current time point tc Candidate adjacent units v l h performs a recursive search and updates the current path sequence to curPath = [v t0 ,v t1 ,...,v t(c-1) ,v tc ,v lh ], where c represents the index of the current time point, and the length of the path before the update is:

[0053]

[0054] The path search function PathSearch(curPath, prePathLen) is called recursively.

[0055] Step 12-7: Update the candidate adjacent cell index h = h + 1, and return to step 12-5;

[0056] Step 12-8: End the current path search function;

[0057] In step 12, the path recursive search function is constructed by recursively searching for cell paths that satisfy the distance constraint maxd.

[0058] Beneficial effects: Compared with traditional path planning methods, the present invention has the following advantages: (1) It can deduce multiple navigation paths that meet time constraints, with a more comprehensive coverage and better meet the actual needs of the mission; (2) It can divide the sea surface space as needed based on the Venn diagram, effectively improve the exploration efficiency of navigation paths, save a lot of computational costs, and flexibly adapt to multi-granularity spatiotemporal constraints. Attached Figure Description

[0059] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, and the advantages of the present invention as described above or otherwise will become clearer.

[0060] Figure 1 This is a flowchart of the method of the present invention.

[0061] Figure 2 This is the Venn diagram model provided in Embodiment 2 of the present invention.

[0062] Figure 3 This is the Venn diagram model provided in Embodiment 3 of the present invention. Detailed Implementation

[0063] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0064] Given a known UAV target's sea route, start point P. start Latitude and longitude (X) start ,Y start Destination P end Latitude and longitude (X) end ,Y end The maximum flight speed (maxV) and the longest flight time (Δt) required for the UAV target to complete this flight path are given. The distances between the islands and maritime reconnaissance stations to be avoided and other spatially constrained areas are relatively large. Therefore, it is necessary to deduce the set of maritime routes (pSet) for the UAV target that satisfies the spatiotemporal constraints. Figure 1 As shown, the deduction steps are as follows:

[0065] Step 1: Construct a set of spatially constrained regions: Obtain the centroid coordinates o of each spatially constrained region i, including islands and reconnaissance stations involved in the UAV target's maritime navigation. i Calculate the minimum radius r of the circle that can cover the spatially constrained region, using the centroid coordinates as the center. i The number of spatially constrained regions is denoted as n. area Construct a region consisting of all spatial constraints (o i ,r i The set of spatially constrained regions formed by ) |ResSet|=n area ;

[0066] Step 2: Initialize the Veno map seed set: Initialize the Veno map seed set to empty, SeedSet = {}, and set the number of edges k of the region cells;

[0067] Step 3: Construct the seed set of the Venn diagram: For each spatial constraint region i in the ResSet of spatial constraint regions, set its centroid o i As a seed, it is added to the SeedSet of the Venn diagram, that is... With the centroid o of spatially constrained region ii Let r be the center of the circle, and its minimum radius be r. i The radius is twice the radius, i.e., R = 2r. i The outer perimeter of the constrained region i is formed by the area C'. i ,take outsourcing garden C' i k equal parts P k(i-1)+1 P k(i-1)+2 ,…,P ik Add it as a seed to the Vinyson map seed set SeedSet, i.e. Assuming there are no other seeds inside the outer circle, the number of seeds in the seed set SeedSet is N = (k+1)n. area ;

[0068] Step 4: Construct the Venn map: Based on the N seeds in the Venn map seed set SeedSet, construct the Venn map Vor(SeedSet), dividing the plane into N units v. j 1≤j≤N, each cell contains one and only one seed from the Vinyau diagram seed set SeedSet. j 1≤j≤N, satisfying that the distance between a point in a cell and its corresponding seed is shorter than the distance between points in the cell and other seeds, the edges of the Vor(SeedSet) are the perpendicular bisectors or rays of a pair of seeds, and the points of the Vor(SeedSet) correspond to the centers of the circumcircles of the three seeds, and the circumcircles do not contain any seeds from the seed set SeedSet (VorOi graph construction reference Jan Oliver Vallgrün, Voronoi Graph Matching for Robot Localization and Mapping, Transactions on Computational Science IX, 2010);

[0069] Step 5: Initialize the adjacency matrix of the Venn diagram: For each cell v in the Venn diagram Vor(SeedSet) j Calculate the number of other adjacent units m. j , building unit v j m j ×m j ∑-dimensional adjacency matrix j initialize its elements to an empty set, that is 1≤f,g≤m j ;

[0070] Step 6: Calculate the adjacency distance in the Venn diagram: based on cell v j m j ×m j ∑-dimensional adjacency matrix j Calculate the values ​​of its two adjacent units vf -v g Minimum diameter between With the maximum diameter The smallest diameter This refers to unit v j The inner can connect its adjacent units v f With v g Shortest straight-line distance, maximum diameter This refers to unit v j The inner can connect its adjacent unit v f With v g The longest straight-line distance, which will be the unit v j With v f Let the edge between them be denoted as l j,f Unit v j With v g Let the edge between them be denoted as l j,g The calculation process is as follows:

[0071]

[0072] Where, min represents taking the minimum value, "l j,f and l j,g "are connected" means that the edges are connected. j,f With edge l j,g They are connected;

[0073]

[0074] l j,f .X1 and l j,f .X2 represent edges l respectively j,f The two endpoints, l j,g .X1 and l j,g .X2 represent edges l respectively j,g The two endpoints, MinDist(l j,f .X1,l j,g ) represents edge l j,f endpoint X1 to edge l j,g The shortest distance, MaxDist(l j,f .X1,l j,g ) represents edge l j,f endpoint X1 to edge l j,g The longest distance, and so on, for unit v j Two adjacent units v f -v g Minimum diameter between With the maximum diameter The average value is used as its adjacent distance:

[0075]

[0076] Step 7: Update the adjacency matrix of the Venn diagram: Update cell v j m j ×m j Σ adjacency matrix j Update its off-diagonal elements to be composed of the corresponding minimum diameter. With the maximum diameter The set that is formed, namely Σ j [v f ,v g ]=dist j (v f ,v g f≠g, 1≤f,g≤m j ;

[0077] Step 8: Filter cells based on distance constraints: For each cell v j Calculate its relationship with the starting point P. start With destination P end The minimum distance and MinSumDist(v j ,P start ,P end As shown in equation (4):

[0078]

[0079] The maximum flight distance maxd of the UAV target is calculated as shown in equation (5):

[0080] maxd=maxV×Δt (5)

[0081] Select the minimum distance and MinSumDist(v) in the non-spatial constrained region j ,P start ,P end For cells whose maximum flight distance is less than or equal to the maximum flight distance maxd = maxV × Δt, construct an optional cell set vSet, as shown in equation (6):

[0082] vSet = {v j |MinSumDist(v j ,P start ,P end )≤maxd} (6)

[0083] Step 9: Initialize the path set: Initialize the path set to an empty set, i.e.

[0084] Step 10: Determine if the cell set vSet is empty. If it is, skip to step 13; otherwise, continue to step 11.

[0085] Step 11: Initialize the current path: Initialize the current path sequence from the starting point P. start Unit v start The constructed sequence, i.e., curPath = [v start ], where P start ∩v start v start ∩vSet initializes the path length of the previous time step to zero, i.e., prePathLen = 0;

[0086] Step 12: Construct a recursive path search function: Construct a recursive path search function PathSearch(curPath, prePathLen), whose recursive parameters include: the current path sequence consisting of several units, curPath = [v t0 ,v t1 ,...,v t(c-1) ,v tc ], and from the initial time t0 to the previous time point t c-1 The path length traversed, i.e., the path length prePathLen at the previous time step, and the unit reached by the current path sequence, i.e., the current unit, denoted as v. tc :

[0087] Step 12-1: Calculate the shortest distance from the current path sequence to the destination, curPathLen, which is equivalent to calculating the path length prePathLen from the previous time step and the distance between edge l. t(c-1),tc The sum of the minimum distances to the destination.

[0088] curPathLen=prePathLen+min(dist(l t(c-1),tc .midpt,P end (7)

[0089] Step 12-2: If the shortest distance from the current path sequence to the destination, curPathLen, exceeds the longest flight distance, maxd (i.e., curPathLen > maxd), it means that the current path does not meet the longest flight distance requirement. Skip to step 12-8. Otherwise, continue to step 12-3.

[0090] Step 12-3: If the current path has reached the destination P end The unit in question, namely P end ∈v cur Add the current path to the path set pSet = pSet∪{curPath}, jump to step 12-8, otherwise continue to step 12-4;

[0091] Step 12-4: Based on the current cell v in the adjacency matrix of the Vinograph... tc The adjacency matrix ∑ tc Find unit v tc m tc An adjacent cell that belongs to the cell set vSet but is not part of the current path curPath. If the number of units is greater than zero, i.e., m tc If the value is greater than 0, initialize h = 1 and continue to step 12-5; otherwise, skip to step 12-8.

[0092] Step 12-5: Determine if the adjacent cell index satisfies h≤m tc If yes, continue to step 12-6; otherwise, skip to step 12-8.

[0093] Step 12-6, for adjacent unit v lh Perform a recursive search and update the current path sequence to curPath = [v t0 ,v t1 ,...,v t(c-1) ,v tc ,v lh The path length before the update was:

[0094]

[0095] The path search function PathSearch(curPath, prePathLen) is called recursively.

[0096] Step 12-7: Update the adjacent cell index, h = h + 1, and return to step 12-5;

[0097] Step 12-8: End the current path search function;

[0098] Step 13: Run the recursive path search function: using the current path sequence curPath = [v start The path length at the previous time step, prePathLen = 0, is used as input to call the recursive function PathSearch(curPath, prePathLen);

[0099] Step 14: Output the path set pSet.

[0100] This invention uses the flight time of the UAV as a time constraint, and divides the sea surface avoidance and normal navigation space based on the Venn diagram model. On this basis, it realizes the deduction of the sea navigation path that meets the time constraint.

[0101] Example 1:

[0102] Constructing a Veno diagram:

[0103] If a UAV target's sea route needs to avoid an island, then the Veno map constructed for that island is as follows:

[0104] Step 1: Construct a set of spatially constrained regions: Obtain the centroid coordinates O of the spatially constrained regions involved in the island, calculate the minimum circle radius R that can cover the spatially constrained region with the centroid coordinates O as the center, and construct a set of spatially constrained regions ResSet = {(O,R)} consisting of all spatially constrained regions (O,R), where |ResSet| = 1;

[0105] Step 2: Initialize the Veno map seed set: Initialize the Veno map seed set to empty, SeedSet = {}, and set the number of edges of the region cells k = 6;

[0106] Step 3: Construct the Venn diagram seed set: Add the centroid O as a seed to the Venn diagram seed set SeedSet, i.e., SeedSet={O}. With the centroid O as the center and 2R as the radius, form the outer circle C′ of the constraint region. Take the 6 equal parts P1, P2, ..., P6 of the outer circle C′ and add them as seeds to the Venn diagram seed set SeedSet, i.e., SeedSet={O,P1,P2,...,P6}. The number of seeds in the seed set SeedSet is N=7.

[0107] Step 4: Construct the Veno graph: Based on the 7 seeds in the Veno graph seed set SeedSet, construct the Veno graph Vor(SeedSet). Divide the plane into 7 units, denoted as v1 to v7. Each unit contains one and only one seed from the Veno graph seed set SeedSet. j .

[0108] Example 2:

[0109] Construct the adjacency matrix of the Vinograph:

[0110] The Vinograph is set to consist of 7 units. Figure 2 As shown, unit v1 is adjacent to two other units v2 and v6. The minimum diameter between two adjacent units v2-v6 within unit v1 is... Maximum diameter is The adjacency matrix of the Vinograph unit v1 is constructed as follows:

[0111] Step 6: Calculate the adjacency distance in the Venn diagram: The distance between two adjacent cells v2-v6 within cell v1 is calculated as follows:

[0112]

[0113] Step 7: Update the adjacency matrix of the Venn diagram: Update the adjacency matrix ∑1 of cell v1, ∑1[v2,v6]=dist1(v2,v6)=4.5.

[0114] Example 3: Path recursive search.

[0115] The adjacency matrix of the Venn diagram for unit v2 is shown in Table 1:

[0116] Table 1

[0117] <![CDATA[v1]]> <![CDATA[v3]]> <![CDATA[v4]]> <![CDATA[v1]]> - 3.464 3 <![CDATA[v3]]> 3.464 - 1.732 <![CDATA[v4]]> 3 1.732 -

[0118] Let the current path sequence be curPath=[v1,v2], the cell set be vSet={v3,v4}, the set of adjacent cells belonging to cell set vSet that can be visited further be {v3,v4}, and the number of adjacent cell sets that can be visited further be m. tc =2, the adjacency matrix of the Vinograph for unit v2 is shown in xx, the path length at the previous time step prePathLen = 1.732, the longest flight distance maxd = 9, as shown in xx. Figure 3 As shown, the recursive path search process is as follows:

[0119] Step 12-1: Calculate the shortest distance from the current path sequence to the destination, curPathLen, which is equivalent to calculating the path length prePathLen from the previous time step and the distance between edge l. t(c-1),tc The sum of the minimum distances to the destination, curPathLen = 1.732 + 5.196 = 6.928;

[0120] Step 12-2: The shortest distance from the current path sequence to the destination, curPathLen, does not exceed the longest flight distance, maxd=9. Continue to step 12-3.

[0121] Step 12-3: The current path has not reached the destination P. end In unit v3, continue with steps 12-4;

[0122] Step 12-4: Based on the adjacency relation matrix ∑2 of the current cell v2 in the Venn diagram, find the two adjacent cells v3 and v4 belonging to the cell set vSet of cell v2. tc =2, initialize h=1, continue with steps 12-5;

[0123] Step 12-5: Determine if the adjacent cell index satisfies h≤2, then continue to step 12-6;

[0124] Step 12-6: Perform a recursive search on the adjacent unit υ3, update the current path sequence to curPath=[v1,v2,v3], update the path length of the previous time step to: prePathLen=1.732+3.464=5.196, and recursively call the path search function PathSearch(curPath,prePathLen);

[0125] Step 12-a1: Calculate the shortest distance from the current path sequence to the destination: curPathLen = 5.196 + 1.732 = 6.928;

[0126] Step 12-a2: The shortest distance from the current path sequence to the destination, curPathLen, does not exceed the longest flight distance, maxd. Continue to step 12-a3.

[0127] Step 12-a3: The current path has reached the destination P. end The unit in question, namely P end ∈v cur Add the current path to the path set pSet = pSet∪{curPath} = {[v1,v2,v3]}, and jump to step 12-a8;

[0128] Step 12-a8: End the current path search function and continue the previous level recursive search 12-7;

[0129] Step 12-7: Update the adjacent cell index, h=2, and return to step 12-5;

[0130] Step 12-5: Determine if the adjacent cell index satisfies h≤m tc Continue with steps 12-6;

[0131] Step 12-6: Perform a recursive search on adjacent cell v4, update the current path sequence to curPath = [v1, v2, v4], update the path length of the previous time step to prePathLen = 1.732 + 3 = 4.732, and recursively call the path search function PathSearch(curPath, prePathLen);

[0132] Step 12-b1: Calculate the shortest distance from the current path sequence to the destination, curPathLen, which is calculated by multiplying the path length prePathLen from the previous time step by the edge l. t(c-1),tc The sum of the minimum distances to the destination, curPathLen = 4.732 + 3 = 7.732;

[0133] Step 12-b2: Determine if the shortest distance curPathLen from the current path sequence to the destination does not exceed the longest flight distance maxd, then continue to step 12-b3;

[0134] Step 12-b3: Determine if the current path has not reached the destination P. end Continue to step 12-b4 in the current unit;

[0135] Step 12-b4: Based on the adjacency relation matrix ∑4 of the current cell v4 in the Vinograph adjacency relation matrix, find the adjacent cell v3 of cell v4 that belongs to the cell set vSet but does not belong to the current path curPath. tc =1, initialize h=1, continue with step 12-b5;

[0136] Step 12-b5: Determine if the adjacent cell index satisfies h≤m tc Continue with step 12-b6;

[0137] Step 12-b6, for adjacent unit v lh Perform a recursive search, update the current path sequence to curPath = [v1, v2, v4, v3], update the path length of the previous time step to prePathLen = 4.732 + 1.732 = 6.464, and recursively call the path search function PathSearch(curPath, prePathLen);

[0138] Step 12-c1: Calculate the shortest distance from the current path sequence to the destination, curPathLen, which is equivalent to calculating the path length prePathLen from the previous time step and the edge length l. c-1,c The sum of the minimum distances to the destination, curPathLen = 6.464 + 1.732 = 8.196;

[0139] Step 12-c2: Determine if the shortest distance curPathLen from the current path sequence to the destination does not exceed the longest flight distance maxd, then continue to step 12-c3;

[0140] Step 12-c3: Determine if the current path has reached the destination P. end The unit in question, namely P end ∈v cur Add the current path to the path set pSet = pSet∪{[v1,v2,v4,v3]} = {[v1,v2,v3],[v1,v2,v4,v3]}, then jump to step 12-c8.

[0141] Step 12-c8: End the current path search function and continue the previous level recursive search in step 12-b7;

[0142] Step 12b7: Update the adjacent cell index, h = h + 1 = 3, return to step 12-b5;

[0143] Step 12-b5: Determine if the adjacent cell index does not satisfy h≤m tc Skip to step 12-b8;

[0144] Step 12-b8: End the current path search function and continue to the previous level recursive search function, step 12-a7;

[0145] Step 12-a7: Update the adjacent cell index, h = h + 1 = 2, return to step 12-a5;

[0146] Step 12-a5: Determine if the adjacent cell index does not satisfy h≤m tc Skip to step 12-a8;

[0147] Step 12-a8: End the current path search function and continue to the previous level recursive search function, step 12-7;

[0148] Step 12-7: Update the adjacent cell index, h = h + 1 = 3, and return to step 12-5;

[0149] Step 12-5: Determine if the adjacent cell index does not satisfy h≤m tc Skip to steps 12-8;

[0150] Steps 12-8: End the current path search function and continue to step 14;

[0151] Step 14: Output path set pSet = {[v1,v2,v3],[v1,v2,v4,v3]}.

[0152] When extrapolating maritime navigation paths for UAV targets, the method of this invention can be used to construct a set of spatially constrained regions, including islands and reconnaissance stations. Based on this, a Venn diagram model and its adjacency matrix are constructed to delineate the UAV avoidance and normal navigation spaces. The maritime navigation path of the UAV target is then recursively extrapolated using units within the Venn diagram. This method overcomes the limitation of traditional path extrapolation methods that prioritize the shortest path, considering multiple possibilities for the UAV target's maritime navigation path under distance constraints, thus providing more comprehensive coverage. Furthermore, the recursive search method based on Venn diagram units effectively improves the efficiency of navigation path exploration, significantly reduces computational costs, and flexibly adapts to multi-granularity spatial constraints.

[0153] The research work of this invention was supported by the Collaborative Innovation Center for New Software Technologies and Industrialization.

[0154] In its specific implementation, this application provides a computer storage medium and a corresponding data processing unit. The computer storage medium is capable of storing a computer program, which, when executed by the data processing unit, can run the invention's content regarding a method for calculating the maritime navigation path of a UAV target based on a Venn diagram, as well as some or all of the steps in various embodiments. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0155] Those skilled in the art will clearly understand that the technical solutions in the embodiments of the present invention can be implemented using computer programs and their corresponding general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of computer programs, i.e., software products. These computer program software products can be stored in a storage medium and include several instructions to cause a device containing a data processing unit (which may be a personal computer, server, microcontroller, MUU, or network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present invention.

[0156] This invention provides a method for predicting the maritime navigation path of a UAV target based on a Vinio chart. Many methods and approaches exist for implementing this technical solution; the above description is merely a preferred embodiment of the invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this invention, and these improvements and modifications should also be considered within the scope of protection of this invention. All components not explicitly stated in this embodiment can be implemented using existing technologies.

Claims

1. A method for deriving the maritime navigation path of a UAV target based on a Vinio chart, characterized in that, Includes the following steps: Step 1: Construct a set of spatially constrained regions: Obtain the various spatially constrained regions involved in the maritime navigation of the UAV target. barycentric coordinates Calculate the minimum radius of the circle that can cover the spatially constrained region, using the centroid coordinates as the center. The number of spatially constrained regions is denoted as Constructing regions with all spatial constraints The set of spatially constrained regions , ; Step 2: Initialize the Vinyasa diagram seed set: Initialize the Vinyasa diagram seed set... Initialize to empty: ; Step 3: Construct the seed set of the Venn diagram: for the set of spatially constrained regions Each spatial constraint region to shift its center of gravity Add as a seed to the Vinotu seed set : To constrain the region by space center of gravity The minimum radius of the circle centered at a point is... twice the radius Forming a constrained region Outsourcing Park Outsourcing Park of Equal parts , ,…, ,Will Equal division points are added as seeds to the Vinonic diagram seed set. ; Step 4: Construct the Venn diagram: based on the seed set of the Venn diagram In Construct a Vinograph from seeds ; Step 5: Initialize the adjacency matrix of the Venn diagram: For the Venn diagram Each unit in Calculate the number of other adjacent units. Building unit of 3D adjacency matrix initialize its elements to an empty set: , ,in and Representation unit Adjacency unit index, and Representation unit Two adjacent units; Step 6: Calculate the adjacency distance in the Venn diagram: based on cells of 3D adjacency matrix Calculate the pairs of adjacent cells. Minimum diameter between With the maximum diameter The smallest diameter This refers to the unit It can connect its adjacent units. and Shortest straight-line distance, maximum diameter This refers to the unit It can connect its adjacent units. and The longest straight-line distance; Step 7: Update the adjacency matrix of the Venn diagram: Update the unit of 3D adjacency matrix Adjacency matrix Each off-diagonal element in the array is updated to be the corresponding minimum diameter. With the maximum diameter The set that constitutes; Step 8: Filter cells based on distance constraints: For each cell Calculate its relationship with the starting point With destination minimum distance and ; Step 9: Initialize the path set: Set the path set... Initialize to an empty set: ; Step 10: Determine the set of units Is it empty? If yes, skip to step 13; otherwise, continue to step 11. Step 11: Initialize the current path: Initialize the current path sequence. Starting from Unit The sequence formed; Step 12: Constructing the path recursive search function: Constructing the path recursive search function. ; Step 13: Run the recursive path search function: using the current path sequence Path length compared to the previous time For input, call the recursive function ; Step 14: Output path set .

2. The method according to claim 1, characterized in that, Step 1 includes: setting the starting point of the known UAV target's sea route. Destination Maximum flight speed of UAV targets And the longest flight time required to complete the flight path. .

3. The method according to claim 2, characterized in that, Step 2 includes: setting the number of edges of the corresponding Venn diagram element for the constrained region. .

4. The method according to claim 3, characterized in that, Step 3 includes: setting the outer circle to contain no other seeds, and setting the seed set. The number of seeds in .

5. The method according to claim 4, characterized in that, Step 4 includes: dividing the plane into Each Vinaus diagram unit , Each unit contains one and only one set of Vinaut diagram seeds. One of the seeds , A Venn diagram satisfies the condition that the distance between a point within a cell and its corresponding seed is shorter than its distance to any other seed. The sides are the perpendicular bisectors or rays of a pair of seeds, as shown in the Vinno diagram. The vertices in the graph correspond to the centers of the circumcircles of the three seeds, and the interiors of the circumcircles do not contain the set of seeds in the Venn diagram. The seeds in the middle.

6. The method according to claim 5, characterized in that, Step 5 includes: initializing the adjacency relation matrix of the Vinograph, for each cell... Number of adjacent units equal to this unit The number of sides.

7. The method according to claim 6, characterized in that, Step 6 includes: connecting two adjacent units and The edge between them is denoted as Two adjacent units and The edge between them is denoted as The calculation process is as follows: (1)， Where min represents taking the minimum value. and "are connected" indicates that the edges are connected. With edge They are connected; (2)， in Represents the straight-line distance between points x and y, when the side With edge Minimum diameter at intersection With the maximum diameter All are 0. and Representing edges respectively The two endpoints, and Representing edges respectively The two endpoints, Representing an edge endpoints to the edge The shortest distance, Representing an edge endpoints to the edge The longest distance, will be the unit Pairs of adjacent units Minimum diameter between With the maximum diameter The average value is used as the unit as a mediating unit and Distance between : (3)。 8. The method according to claim 7, characterized in that, Step 7 includes: the set is: , , ,in and Representation and Unit Two adjacent adjacent cells, Indicated by unit as a mediating unit and The distance between them.

9. The method according to claim 8, characterized in that, Step 8 includes: The calculation formula is: (4)， in, Representation unit The point in the middle; Calculate the longest flight distance of a UAV target : (5)， Select the minimum distance in the non-spatial constrained region. Less than or equal to the longest flight distance The units, constructing an optional unit set As shown in equation (6): (6) 。 10. The method according to claim 9, characterized in that, Step 11 includes: ,in , The path length of the previous time period The value is zero before initialization. ; Step 12 includes: constructing a path recursive search function Its recursive parameters include: from the start time From start to current time The sequence of Vinograph units traversed is represented as follows: From the initial time up to the previous time point The path length traversed, i.e. the path length at the previous time step, is denoted as ,in This indicates the unit reached by the current path sequence, i.e., the current unit, and then the following steps are performed: Step 12-1: Calculate the shortest distance from the current path sequence to the destination. : (7)， in Indicates from edge Midpoint to destination The distance; Step 12-2: If the shortest distance from the current path sequence to the destination... Exceeding the longest flight distance ,Right now If the current path does not meet the maximum flight distance requirement, proceed to step 12-8; otherwise, continue to step 12-3. Step 12-3: If the current path has reached the destination The unit in which it is located, i.e. Add the current path to the path set. If yes, proceed to step 12-8; otherwise, continue to step 12-4. Step 12-4: Based on the current cell in the adjacency matrix of the Vinograph Adjacency matrix Find the unit of Each belongs to the unit set Not belonging to the current path Candidate Adjacency Units , Indicates the first If the number of candidate adjacent units is greater than zero, that is... ,initialization =1, continue to step 12-5; otherwise, skip to step 12-8. Step 12-5: Determine if the adjacent cell index satisfies the condition. If yes, continue to step 12-6; otherwise, skip to step 12-8. Step 12-6: For the unit at the current time point Candidate Adjacency Units Perform a recursive search and update the current path sequence to... Where c represents the index of the current time point, and the path length of the previous time point is: (8)， Recursive call pathfinding function ; Step 12-7: Update the candidate adjacent cell index Return to steps 12-5; Step 12-8: End the current path search function; In step 12, during the construction of the path recursive search function, a recursive search is performed to find the path that satisfies the distance constraint. The unit path.

Images