Unmanned aerial vehicle path planning method for military simulation and application thereof

Abstract

The invention discloses an unmanned aerial vehicle path planning method for military simulation and application thereof. The method comprises the following steps: acquiring static obstacle information and dynamic obstacle information of a simulation scene; initializing starting point and terminal point information of navigation, and constructing a priority queue; sequentially traversing each navigation point by taking the terminal point as a starting point, and screening neighbor navigation points based on the static obstacle information to obtain an initial path; and re-planning an initial path based on the dynamic obstacle information. According to the method, an area which is not influenced by a dynamic obstacle is not calculated during path re-planning, a linear combination of a Manhattan distance and a Chebyshev distance approximates an Euclidean distance to serve as a heuristic value, the calculation time is remarkably shortened, a basic cost value is introduced, the calculation rate of an actual cost value and a current minimum cost value is remarkably improved, and compared with a conventional algorithm, the method has the advantage that the calculation efficiency is improved. The path re-planning speed is obviously improved, and the requirement of a complex environment in military simulation is met.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of path planning, and particularly relates to a UAV path planning method for military simulation and application thereof. BACKGROUND

[0002] In military simulation, how to quickly and accurately plan an optimal path for a combat task for a UAV and how to quickly respond to unexpected situations for path re-planning in a complex combat environment are a big challenge for current technology. Conventional path planning algorithms are more suitable for static path planning in a completely static and known environment, and the time complexity of re-planning and first-time planning is the same when facing environmental changes, which cannot respond more quickly and is difficult to quickly and accurately re-plan a path when facing various unexpected situations. SUMMARY

[0003] The application aims to provide a UAV path planning method for military simulation and application thereof to solve the technical problem that conventional path planning algorithms are difficult to quickly and accurately re-plan a path when facing various unexpected situations.

[0004] To achieve the above-mentioned purpose, the first aspect of the application provides a UAV path planning method for military simulation, comprising:

[0005] obtaining static obstacle information and dynamic obstacle information of a simulation scene;

[0006] initializing start point and end point information of navigation and constructing a priority queue based on key value size arrangement;

[0007] taking the end point as the start point, sequentially traversing each navigation point, when traversing to a target navigation point, screening neighbor navigation points of the target navigation point based on the static obstacle information, adding the screened neighbor navigation points into the priority queue according to key value sorting, taking the neighbor navigation point with the smallest key value, updating information, and taking it as the next traversed navigation point, until the traversal is completed, to obtain an initial path;

[0008] judging whether there is a navigation point affected by a dynamic obstacle in the initial path based on the dynamic obstacle information and the initial path;

[0009] if yes, selecting the first affected navigation point as a new obstacle, re-planning a path between the new obstacle and the start point to obtain an updated path;

[0010] judging whether there is a navigation point affected by a dynamic obstacle in the updated path based on the dynamic obstacle information and the updated path;

[0011] if no, taking the updated path as a target path and outputting.

[0012] In one or more embodiments, the static obstacle information includes the outline position of the static obstacle.

[0013] In one or more embodiments, the dynamic obstacle information includes the outline position of the dynamic obstacle at any given time.

[0014] In one or more embodiments, the key value of the navigation point s is calculated using the following formula:

[0015]

[0016] In the formula, Let be the historical minimum cost from navigation point s to the destination, with an initial value of infinity. Let be the minimum cost from navigation point s to the destination. The heuristic value for the navigation point s to the starting point. For key value correction, The coefficient of thermal expansion is , and Greater than or equal to 1 Starting point;

[0017] in, The calculation formula is as follows:

[0018]

[0019] In the formula, Let be the set of neighboring navigation points of navigation point s. Representing navigation points To the navigation point The actual value of the transaction. The endpoint;

[0020] The calculation formula is as follows:

[0021]

[0022]

[0023] The calculation formula is as follows:

[0024] In the formula, Let s be the Manhattan distance from the navigation point to the starting point. Let be the Chebyshev distance from navigation point s to the starting point, and a and b be linear coefficients. The cost correction factor from navigation point s to the starting point. From navigation point s to navigation point Cost correction factor To be with navigation points The basic value of the type association.

[0025] In one or more embodiments, a Cartesian coordinate system is constructed centered on navigation point s. Neighboring navigation points are classified based on the number of coordinate axes in which their positions relative to navigation point s change within this Cartesian coordinate system; the greater the change in coordinate axis position, the higher the classification. The larger the preset value, the better.

[0026] In one or more embodiments, the step of updating information by selecting the neighbor navigation point with the smallest key value specifically includes:

[0027] Compare the historical minimum cost value to the destination from the neighboring navigation point s with the smallest key value with the current minimum cost value.

[0028] When the historical minimum value is greater than the current minimum value, the current minimum value is used as the historical minimum value of the neighbor navigation point s.

[0029] When the historical minimum value is less than the current minimum value, the historical minimum value of the neighboring navigation point s is set to infinity.

[0030] In one or more embodiments, the step of obtaining the initial path until the traversal is complete specifically includes:

[0031] When the key value of the target navigation point is less than or equal to the key values ​​of all neighboring navigation points in the priority queue, and the historical minimum cost value of the target navigation point is equal to the current minimum cost value, the traversal is complete and the initial path is obtained.

[0032] In one or more embodiments, it further includes:

[0033] If the initial path construction fails, the inflation coefficient is reduced or the maximum traversal count is increased, and then replanning is performed until the initial path is obtained; and / or,

[0034] When replanning the path between the newly added obstacle and the starting point based on the added obstacle, the expansion coefficient is set to 1.

[0035] In one or more embodiments, the step of filtering neighboring navigation points of the target navigation point based on the static obstacle information includes:

[0036] A Cartesian coordinate system is constructed with the target navigation point as the center. Based on the size of the UAV, the minimum bounding box of the UAV is constructed at the center of the Cartesian coordinate system.

[0037] Based on the preset minimum safety distance and the minimum bounding box, a passable bounding box is constructed;

[0038] Obtain the target coordinate axis where the position of the neighbor navigation point changes relative to the target navigation point;

[0039] Based on the passable bounding box, the target coordinate axis, and the static obstacle information, determine whether the neighbor navigation point is passable;

[0040] If not, delete the neighbor navigation point.

[0041] In one or more embodiments, the step of determining whether there are navigation points affected by dynamic obstacles in the initial path based on the dynamic obstacle information and the initial path includes:

[0042] Based on the dynamic obstacle information and the initial path, the relative velocity vector and relative coordinate vector between the dynamic obstacle and the UAV are obtained;

[0043] Based on the relative velocity vector and the relative coordinate vector, the minimum distance time between the dynamic obstacle and the drone is obtained, and it is determined whether the minimum distance time is less than 0.

[0044] If not, calculate the minimum distance between the dynamic obstacle and the drone;

[0045] Based on the radius of the minimum enclosing sphere of the dynamic obstacle, the radius of the minimum enclosing sphere of the drone, and the preset minimum safety distance, the minimum non-collision distance is obtained.

[0046] Determine whether the minimum distance is greater than the minimum non-collision distance;

[0047] If not, calculate the navigation points in the initial path that collide with the dynamic obstacles, mark the candidate new obstacles, and collect them into the queue;

[0048] The step of selecting the navigation point that is first affected and marking it as a new obstacle is as follows:

[0049] According to the minimum distance-time, the candidate new obstacles in the queue are sorted, and the candidate new obstacle with the smallest minimum distance-time is selected as the new obstacle.

[0050] To achieve the above objectives, a second aspect of this application provides a UAV path planning device for military simulation, comprising:

[0051] The information acquisition module is used to acquire static and dynamic obstacle information of the simulation scene;

[0052] The initialization module is used to initialize the start and end information of the navigation and build a priority queue based on the key value size.

[0053] The initial path planning module is used to traverse each navigation point sequentially, starting from the destination. When a target navigation point is reached, the neighbor navigation points of the target navigation point are filtered based on the static obstacle information. The filtered neighbor navigation points are added to the priority queue and sorted by key value. The neighbor navigation point with the smallest key value is selected, the information is updated, and it is used as the navigation point for the next traversal until the traversal is completed, and the initial path is obtained.

[0054] The first dynamic obstacle determination module is used to determine, based on the dynamic obstacle information and the initial path, whether there are navigation points in the initial path that are affected by dynamic obstacles.

[0055] The path replanning module is used to, when there are navigation points affected by dynamic obstacles in the initial path, select the navigation point that is affected first and mark it as a new obstacle, then replan the path between the new obstacle and the starting point to obtain an updated path.

[0056] The second dynamic obstacle judgment module is used to determine whether there are navigation points affected by dynamic obstacles in the updated path based on the dynamic obstacle information and the updated path.

[0057] The output module is used to output the updated path as the target path when there are no navigation points affected by dynamic obstacles in the updated path.

[0058] The advantages of this application, which differ from existing technologies, are:

[0059] In the method of this application, areas unaffected by dynamic obstacles are not calculated during path replanning, and the average replanning calculation speed is more than two orders of magnitude faster than the conventional A* algorithm.

[0060] The method in this application uses a linear combination of Manhattan distance and Chebyshev distance to approximate Euclidean distance as a heuristic, which can significantly reduce the computation time with small errors; and introduces the fundamental cost of integer quantization. It can significantly improve the calculation rate of actual cost value and current minimum cost value, and thus significantly improve the path replanning speed compared with conventional algorithms, meeting the needs of complex environments in military simulation. Attached Figure Description

[0061] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0062] Figure 1 This is a flowchart illustrating one implementation of the UAV path planning method for military simulation used in this application;

[0063] Figure 2 This is a schematic diagram of one implementation method of the neighbor navigation point in this application;

[0064] Figure 3 yes Figure 1 A flowchart illustrating one embodiment corresponding to S300;

[0065] Figure 4 This is a schematic diagram of one embodiment of the minimum bounding box and the passable bounding box of this application;

[0066] Figure 5 This is a schematic diagram of the obstacle analysis surface for Class B candidate navigation points;

[0067] Figure 6 yes Figure 1 A flowchart illustrating one embodiment corresponding to S400;

[0068] Figure 7 This is a schematic diagram of one embodiment of the UAV path planning device for military simulation used in this application;

[0069] Figure 8 This is a schematic diagram of one embodiment of the electronic device of this application. Detailed Implementation

[0070] To enable those skilled in the art to better understand the technical solutions in this disclosure, the technical solutions in the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this disclosure.

[0071] With the maturation of drone technology, the innovation of operational concepts, and changes in the strategic environment, the form of modern warfare has undergone significant changes. The economic efficiency and low risk of drone warfare have made it an important trend in military operations. Military simulation is used to address the extreme complexity of modern warfare, reduce the cost and risk of real military activities, and leverage the achievements of the information technology revolution. It provides a safe, economical, controllable, repeatable, and measurable virtual battlefield laboratory that runs through the entire process of war preparation (research, planning, training, equipment development) and war execution (scheme evaluation, decision support). It is an indispensable strategic support technology for modern armies to enhance combat effectiveness, optimize resource allocation, reduce casualty risks, and seize the commanding heights of future warfare. However, in military simulation, how drones can quickly and accurately plan the optimal path for combat missions and how to rapidly respond to emergencies and replan paths in complex combat environments are major challenges currently facing the technology.

[0072] There is limited domestic research on this topic, and existing technologies mainly utilize and improve upon the A* algorithm to achieve optimal path planning. However, the A* algorithm is better suited for static path planning in completely static and known environments. When faced with environmental changes, the time complexity of replanning is the same as the initial planning, making it unable to provide a faster response and failing to comprehensively consider the speed and accuracy of path planning, as well as how to quickly and accurately replan in the face of various unforeseen circumstances.

[0073] To address the aforementioned issues, the applicant has developed a novel UAV path planning method for military simulation. This method significantly improves the replanning calculation speed of UAVs in military simulations when facing various unforeseen circumstances, while also increasing the average speed during the initial planning phase.

[0074] Specifically, please refer to Figure 1 , Figure 1 This is a flowchart illustrating one implementation method of the UAV path planning method for military simulation used in this application.

[0075] like Figure 1 As shown, the method includes:

[0076] S100: Obtain static and dynamic obstacle information of the simulation scene.

[0077] In one implementation, static obstacle information may include the outline position of the static obstacle.

[0078] Specifically, static obstacles can be fortifications, bunkers, buildings, trees, etc. that are designated as indestructible in the battlefield environment. Their outlines are calculated and the outline information is voxelized and saved as static impassable obstacle points in the database for subsequent path planning. This can reduce the obstacle analysis pressure during system operation and improve the path planning speed.

[0079] Since static obstacle information only includes outline information and not internal information, the amount of data that needs to be stored can be significantly reduced, improving runtime analysis speed. For example, for a static building of 20m*20m*100m with navigation point voxels of 1m*1m*1m, the complete information is... One navigation point, but only need to record the outline With each navigation point, the amount of information recorded is reduced by 78%, and the recording efficiency is 4.54 times that of complete information, while achieving the same effect.

[0080] In one implementation, dynamic obstacle information may include the outline position of the dynamic obstacle at any given time.

[0081] Specifically, similar to static obstacle information, dynamic obstacle information can also include only the outline position. Dynamic obstacles can be various moving entities in the battlefield environment, as well as fortifications, bunkers, buildings, trees, etc., that are set to be destructible during the simulation. Their size, position at each moment, velocity detection, etc. are calculated and obtained, which can be used for dynamic obstacle avoidance during path planning. This can trigger path replanning to avoid obstacles when a collision is possible, or trigger path replanning to update the optimal route when obstacles are removed (such as when a building collapses).

[0082] S200: Initialize the start and end information of the navigation and build a priority queue based on the key value.

[0083] Specifically, the algorithm of this application uses a key value to represent the priority of each navigation point, and the smaller the key value, the higher the priority of the navigation point.

[0084] Taking navigation point s as an example, the formula for calculating its key value is as follows:

[0085]

[0086] In the formula, Let be the historical minimum cost from navigation point s to the destination, with an initial value of infinity. Let be the minimum cost from navigation point s to the destination. The heuristic value for the navigation point s to the starting point. For key value correction, The coefficient of thermal expansion is , and Greater than or equal to 1 Starting point;

[0087] in, The calculation formula is as follows:

[0088]

[0089] In the formula, Let be the set of neighboring navigation points of navigation point s. Representing navigation points To the navigation point The actual value of the transaction. The endpoint;

[0090] The calculation formula is as follows:

[0091]

[0092]

[0093] The calculation formula is as follows:

[0094] In the formula, Let s be the Manhattan distance from the navigation point to the starting point. Let be the Chebyshev distance from navigation point s to the starting point, and a and b be linear coefficients.

[0095] The cost correction factor from navigation point s to the starting point. From navigation point s to navigation point The cost correction factor can be determined comprehensively based on battlefield factors such as enemy and friendly situation, weather factors, and degree of danger, and can be quantified by integer model.

[0096] To be with navigation points The basic value of the type association.

[0097] The following details how, during key-value calculation, an expansion coefficient is constructed. It can improve planning speed, and its value can be set based on obstacle density, for example, when the obstacle density is around 30%. Try to quickly plan a path, and the cost of the planned path should not be higher than the cost of the actual optimal path. times.

[0098] Heuristic value in this embodiment In the calculation process, the linear combination of Manhattan distance and Chebyshev distance is used to approximate Euclidean distance as a heuristic value, and linear parameters a and b are introduced, which can significantly reduce the calculation time with small error.

[0099] For example, in one implementation, , The heuristic values ​​obtained based on Manhattan distance and Chebyshev distance have an average error of 6% compared to the true Euclidean values, and the average computation time is reduced to 39.1%.

[0100] In addition, the actual cost in this embodiment The basic cost was introduced during the calculation. Its neighboring navigation points with navigation point s It is related to the type.

[0101] For example, since drones have volume, neighbor navigation points can be defined in consideration of the drone's volume, and a minimum bounding box surrounding the drone can be constructed based on the drone's size. , The length, width and height are respectively , , This allows the system to perfectly surround the drone without touching it. After constructing a Cartesian coordinate system centered on this minimum bounding box, the drone's selectable neighbor navigation point for each step can be an adjacent cube. (See [link to relevant documentation]). Figure 2 , Figure 2 This is a schematic diagram of one implementation method of the neighbor navigation point in this application.

[0102] like Figure 2 As shown, the drone's minimum bounding box is located at the center, with 26 selectable neighbor navigation points on the periphery. Among these 26 neighbor navigation points, there are three different types: some neighbor navigation points change only the value of one coordinate axis relative to the drone's position, such as... Figure 2 The C-class candidate navigation points; some neighboring navigation points have changes in the values ​​of two coordinate axes, such as... Figure 2 Category B candidate navigation points; some neighboring navigation points have changes in the values ​​of three coordinate axes, such as... Figure 2 Category A candidate navigation points.

[0103] Understandably, different types of neighboring navigation points result in varying distances from the drone. Therefore, we can define: as the number of coordinate axes with changing values ​​increases, the base cost... Increase.

[0104] In one implementation, through integer quantization, the basic cost of three categories of candidate navigation points, A, B, and C, can be defined. as follows:

[0105]

[0106] Based on this fundamental value, navigation points can be significantly improved. To the navigation point The actual cost value and the current minimum cost value calculation rate.

[0107] S300. Starting from the destination, traverse each navigation point in turn. When a target navigation point is reached, filter the neighbor navigation points of the target navigation point based on the static obstacle information, add the filtered neighbor navigation points to the priority queue, sort them according to the key value, take the neighbor navigation point with the smallest key value, update the information, and use it as the navigation point for the next traversal, until the traversal is completed and the initial path is obtained.

[0108] After initialization, path planning can begin from the destination. First, the accessible neighbor navigation points of the destination can be added to a priority queue. The key values ​​are calculated using the formula shown in S200 and then arranged by key value size.

[0109] You can select the neighbor navigation point with the smallest key value. As the next waypoint, i.e., the navigation point for the next traversal; it can also be used for neighboring navigation points. The information will be updated.

[0110] The following details the method for filtering neighbor navigation points. First, based on static obstacle information, the neighbor navigation points of the target navigation point s are filtered to identify neighbor navigation points with static obstacles.

[0111] For specific screening methods, please refer to [link / reference]. Figure 3 , Figure 3 yes Figure 1 A flowchart of one embodiment corresponding to S300.

[0112] like Figure 3 As shown, methods for filtering neighbor navigation points may include:

[0113] S301. Construct a Cartesian coordinate system centered on the target navigation point. Based on the size of the UAV, construct the minimum bounding box of the UAV at the center of the Cartesian coordinate system.

[0114] Similar to the relevant description in S200, the minimum bounding box of the UAV can be constructed. , The length, width and height are respectively , , It can just surround the drone without touching it.

[0115] S302. Construct a passable bounding box based on the preset safety distance and minimum bounding box.

[0116] For example, please refer to Figure 4 , Figure 4 This is a schematic diagram of one embodiment of the minimum bounding box and the passable bounding box of this application.

[0117] like Figure 4 As shown, the definition Represents a rectangle Circumcircle radius, defined , Let represent the minimum safe distances from obstacles in the horizontal and vertical directions, respectively. In this case, a sufficient condition for navigation point s to be a passable navigation point for the UAV is that the base of navigation point s has a length of [missing information]. Gao Wei All navigation points within the rectangular bounding box are passable navigation points, meaning all navigation points within the bounding box are passable navigation points.

[0118] S303. Obtain the target coordinate axes where the positions of neighboring navigation points relative to the target navigation point have changed.

[0119] S304. Based on the passable bounding box, target coordinate axis, and static obstacle information, determine whether the neighboring navigation point is passable.

[0120] In this embodiment, a dimensionality reduction analysis method is used to analyze only the coordinate axis directions in which the position of the neighboring navigation points changes relative to the target navigation point, and whether there are obstacle points.

[0121] Specifically, firstly, it is assumed that the initial position of the UAV, i.e. the accessible bounding box centered on the target navigation point, does not overlap with other obstacles. Then, for neighboring navigation points, it is determined whether there are static obstacles on the plane (analysis plane) perpendicular to the target coordinate axis. If not, it means that the neighboring navigation point is accessible.

[0122] For example, for Class A candidate navigation points, all three coordinate axes have changed, so it's necessary to analyze whether the navigation points corresponding to the three affected faces are obstacles. For Class B candidate navigation points, it's necessary to analyze the navigation points corresponding to two faces, and for Class C candidate navigation points, it's necessary to analyze the navigation points corresponding to one face. A total of 6 faces need to be detected. , , , , , The total number of navigation points to be analyzed is ,remember , The above formula simplifies to .

[0123] However, to fully analyze all waypoints, the number of points to be analyzed would be... Clearly, the dimensionality reduction analysis method in this embodiment significantly reduces the amount of analysis.

[0124] For example, please refer to Figure 5 , Figure 5 This is a schematic diagram of the obstacle analysis surface for Class B candidate navigation points. For example... Figure 5 As shown, the candidate navigation points for category B involve and Analysis of the two faces: when there are no static obstacles on the two faces, the candidate navigation point of class B can be considered passable and added to the priority queue.

[0125] The following is a detailed explanation of Neighbor Navigation Points The information update method initially assumes that the historical minimum cost of all navigation points is infinite, and then updates the information for neighboring navigation points. When its historical minimum value > hour, The state is locally over-consistent, indicating that the navigation point An improved environment can make = ;

[0126] When its historical minimum value Less than hour, The state of partial inconsistency indicates that the navigation point A deteriorating environment, or even impassable conditions, can make .

[0127] Based on the above updates, the next traversal can be performed, i.e., traversal... After filtering the neighbor navigation points, add them to the priority queue and sort them according to the key value. Repeat the above steps until the traversal is complete.

[0128] Specifically, in one implementation, traversal completion can mean that when the key value of the target navigation point is less than or equal to the key values ​​of all neighboring navigation points in the priority queue, and the historical minimum cost value of the target navigation point is equal to the current minimum cost value, then the planning is complete and the initial path is obtained.

[0129] That is, the target navigation point currently being traversed. key value ,and = This indicates that the optimal path has been found and the initial planning has been completed.

[0130] During the traversal, if an untravelable neighbor navigation point is encountered, or if the initial path construction is not completed even after reaching the maximum number of traversals, the initial path construction fails. In this case, the inflation coefficient can be reduced. Alternatively, increase the maximum number of traversals, then replan until the initial path is obtained.

[0131] S400: Based on dynamic obstacle information and the initial path, determine whether there are navigation points in the initial path that are affected by dynamic obstacles.

[0132] The initial path planning process achieves avoidance of static obstacles. Furthermore, dynamic obstacle avoidance can be performed based on dynamic obstacle information.

[0133] Specifically, please refer to Figure 6 , Figure 6 yes Figure 1 A flowchart of one embodiment corresponding to S400.

[0134] like Figure 6 As shown, methods for determining whether there are navigation points affected by dynamic obstacles in the initial path may include:

[0135] S401. Based on the dynamic obstacle information and the initial path, obtain the relative velocity vector and relative coordinate vector between the dynamic obstacle and the UAV.

[0136] With time Taking dynamic obstacle obscuration (OBS) as an example, assuming the coordinates of the UAV (User Av) are... The velocity vector is The coordinates of obs are The velocity vector is Then the relative coordinate vectors of uav and obs The relative velocity vectors of UAV and OBS .

[0137] S402. Based on the relative velocity vector and relative coordinate vector, obtain the minimum distance time between the dynamic obstacle and the UAV, and determine whether the minimum distance time is less than 0.

[0138] Furthermore, the time required for UAV and OBS to reach the minimum distance - minimum distance time .

[0139] If the minimum distance time is less than 0, it means that the minimum distance has already occurred in the past. The two sides will not collide while maintaining their current speed, and it can be determined that the dynamic obstacle has no effect on the initial path.

[0140] If the minimum distance time is greater than 0, it indicates that both sides may be expanding, then:

[0141] S403, Calculate the minimum distance between dynamic obstacles and drones.

[0142] Based on the relative velocity vector and the relative coordinate vector, the minimum distance can be calculated as follows: .

[0143] S404. Based on the radius of the minimum enclosing sphere of the dynamic obstacle, the radius of the minimum enclosing sphere of the drone, and the preset minimum safe distance, the minimum non-collision distance is obtained.

[0144] Based on the dimensions of the dynamic obstacle, the circumscribed sphere of the minimum bounding box of the dynamic obstacle can be constructed. The radius of this circumscribed sphere is defined as... .

[0145] Similar to S301 and S302, based on the size of the drone, the circumscribed sphere of the drone's minimum bounding box can be constructed. The sum of the circumscribed sphere's radius and the minimum safe distance is defined as... .

[0146] Therefore, the minimum non-collision distance between the two parties can be obtained. .

[0147] S405. Determine whether the minimum distance is greater than the minimum non-collision distance.

[0148] like This indicates that the two sides are within a safe range when they are closest to each other and will not collide.

[0149] like If , it means that a collision will occur between the two parties, then:

[0150] S406. Calculate the navigation points that collide with dynamic obstacles in the initial path, mark the candidate new obstacles, and collect them into the queue.

[0151] Specifically, with As the origin of time, time The distance between the two sides is Therefore, when a collision occurs, the following conditions are met: The quadratic equation of ,obvious, , , Solve the equation to obtain the solution. and The time of the collision for and The smaller of the negative ones in the middle and lower classes.

[0152] Get collision time Subsequently, based on the initial path, navigation points that collide with dynamic obstacles in the initial path can be obtained, marked as candidate new obstacles, and collected into the queue.

[0153] S500: Select the navigation point that is first affected and mark it as a new obstacle. Replan the path between the new obstacle and the starting point to obtain the updated path.

[0154] After analyzing all dynamic obstacles within the range, you can select... Prioritize obstacle avoidance for the smallest dynamic obstacle, select the navigation point where the dynamic obstacle collides with the initial path, and mark it as a new obstacle.

[0155] After obtaining the coordinates of the newly added obstacle, the S200 method can be used for replanning. In the replanning, it is only necessary to replan the path between the newly added obstacle and the starting point, that is, to traverse the path starting from the navigation point of the successor neighbor of the newly added obstacle, which effectively improves the efficiency of replanning and obtains the updated path.

[0156] Additionally, it should be noted that during the replanning process, the inflation coefficient can be... Change it to 1 to ensure that when the drone encounters an emergency during its journey, it can replan a new path if there is an avoidance solution, thus ensuring the algorithm's adoptability and consistency and avoiding replanning failure.

[0157] S600: Based on dynamic obstacle information and the updated path, determine whether there are navigation points affected by dynamic obstacles in the updated path.

[0158] Similar to S400, it is possible to re-evaluate whether the updated path contains navigation points affected by dynamic obstacles.

[0159] Understandably, if so, the update can continue until there are no navigation points affected by dynamic obstacles in the update path.

[0160] If there are no navigation points affected by dynamic obstacles in the updated path, then:

[0161] S700: Use the updated path as the target path and output it.

[0162] The UAV path planning method based on the above embodiments has a replanning time of [time missing]. Where K is the number of navigation points affected by obstacles, and N is the size of the priority queue. By reusing historical information, unaffected areas are not calculated at all, and its average replanning calculation speed is more than two orders of magnitude faster than the existing technology A*.

[0163] In addition, by constructing the expansion coefficient It can significantly improve the planning speed of the initial path; by using a linear combination of Manhattan distance and Chebyshev distance to approximate Euclidean distance as a heuristic, it can significantly reduce the computation time with a small error; and by introducing the fundamental cost of integer quantization... It can significantly improve the calculation rate of actual cost value and current minimum cost value; thus, compared with conventional algorithms, it significantly improves the speed of initial path planning and replanning, meeting the needs of complex environments in military simulation.

[0164] This application also provides a UAV path planning device for military simulation; please refer to [link / reference]. Figure 7 , Figure 7This is a schematic diagram of one embodiment of the UAV path planning device for military simulation used in this application.

[0165] like Figure 7 As shown, the device includes an information acquisition module 21, an initialization module 22, an initial path planning module 23, a first dynamic obstacle judgment module 24, a path replanning module 25, a second dynamic obstacle judgment module 26, and an output module 27.

[0166] Among them, the information acquisition module 21 is used to acquire static obstacle information and dynamic obstacle information of the simulation scene;

[0167] Initialization module 22 is used to initialize the start and end information of navigation and to build a priority queue based on the key value size;

[0168] The initial path planning module 23 is used to traverse each navigation point sequentially, starting from the destination. When a target navigation point is reached, the neighbor navigation points of the target navigation point are filtered based on static obstacle information. The filtered neighbor navigation points are added to a priority queue and sorted by key value. The neighbor navigation point with the smallest key value is selected, the information is updated, and it is used as the navigation point for the next traversal until the traversal is completed, and the initial path is obtained.

[0169] The first dynamic obstacle judgment module 24 determines whether there are navigation points affected by dynamic obstacles in the initial path based on dynamic obstacle information and the initial path;

[0170] The path replanning module 25 is used to select the navigation point that is first affected by dynamic obstacles in the initial path, mark it as a new obstacle, and replan the path between the new obstacle and the starting point to obtain an updated path.

[0171] The second dynamic obstacle judgment module 26 is used to determine whether there are navigation points affected by dynamic obstacles in the updated path based on dynamic obstacle information and updated path;

[0172] The output module 27 is used to output the updated path as the target path when there are no navigation points affected by dynamic obstacles in the updated path.

[0173] As mentioned above Figures 1 to 6 This specification describes a method for unmanned aerial vehicle (UAV) path planning for military simulation based on embodiments thereof. The details mentioned in the above description of the method embodiments also apply to the UAV path planning device for military simulation based on embodiments thereof. The above-described UAV path planning device for military simulation can be implemented in hardware, software, or a combination of hardware and software.

[0174] This application also provides an electronic device, please refer to...Figure 8 , Figure 8 This is a schematic diagram of one embodiment of the electronic device of this application. For example... Figure 8 As shown, the electronic device 30 may include at least one processor 31, a memory 32 (e.g., non-volatile memory), a RAM 33, and a communication interface 34, and the at least one processor 31, memory 32, RAM 33, and communication interface 34 are connected together via a bus 35. The at least one processor 31 executes at least one computer-readable instruction stored or encoded in the memory 32.

[0175] It should be understood that the computer-executable instructions stored in memory 32, when executed, cause at least one processor 31 to perform the above-described combinations in the various embodiments of this specification. Figures 1-6 The description includes various operations and functions.

[0176] In the embodiments of this specification, electronic device 30 may include, but is not limited to: personal computer, server computer, workstation, desktop computer, laptop computer, notebook computer, mobile electronic device, smartphone, tablet computer, cellular phone, personal digital assistant (PDA), handheld device, messaging device, wearable electronic device, consumer electronic device, etc.

[0177] According to one embodiment, a program product, such as a machine-readable medium, is provided. The machine-readable medium may have instructions (i.e., the elements implemented in software as described above), which, when executed by a machine, cause the machine to perform the above-described combinations of the various embodiments of this specification. Figures 1-5 The various operations and functions described. Specifically, a system or apparatus equipped with a readable storage medium storing software program code that implements the functions of any of the embodiments described above, and enabling the computer or processor of the system or apparatus to read and execute the instructions stored in the readable storage medium.

[0178] In this case, the program code read from the readable medium itself can perform the functions of any of the above embodiments, and therefore the machine-readable code and the readable storage medium storing the machine-readable code constitute a part of this specification.

[0179] Examples of readable storage media include floppy disks, hard disks, magneto-optical disks, optical disks (such as CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD-RW), magnetic tapes, non-volatile memory cards, and ROMs. Alternatively, program code can be downloaded from a server computer or the cloud via a communication network.

[0180] Those skilled in the art will understand that the various embodiments disclosed above can be modified and varied without departing from the spirit of the invention. Therefore, the scope of protection of this specification should be defined by the appended claims.

[0181] It should be noted that not all steps and units in the above process and system structure diagrams are mandatory; some steps or units can be omitted according to actual needs. The execution order of each step is not fixed and can be determined as needed. The device structure described in the above embodiments can be a physical structure or a logical structure. That is, some units may be implemented by the same physical client, or some units may be implemented by multiple physical clients, or they may be jointly implemented by certain components in multiple independent devices.

[0182] In the above embodiments, the hardware units or modules can be implemented mechanically or electrically. For example, a hardware unit, module, or processor may include permanent dedicated circuitry or logic (such as a dedicated processor, FPGA, or ASIC) to perform the corresponding operation. The hardware unit or processor may also include programmable logic or circuitry (such as a general-purpose processor or other programmable processor), which can be temporarily configured by software to perform the corresponding operation. The specific implementation method (mechanical, dedicated permanent circuitry, or temporarily configured circuitry) can be determined based on cost and time considerations.

[0183] The specific embodiments described above with reference to the accompanying drawings are exemplary embodiments, but do not represent all embodiments that can be implemented or fall within the scope of the claims. The term "exemplary" as used throughout this specification means "serving as an example, instance, or illustration" and does not imply that it is "preferred" or "advantageous" compared to other embodiments. Specific details are included to provide an understanding of the described techniques. However, these techniques can be practiced without these specific details. In some instances, well-known structures and apparatuses are shown in block diagram form to avoid obscuring the concepts of the described embodiments.

[0184] The foregoing description of this disclosure is provided to enable any person skilled in the art to implement or use this disclosure. Various modifications to this disclosure will be apparent to those skilled in the art, and the general principles applicable herein can be applied to other variations without departing from the scope of this disclosure. Therefore, this disclosure is not limited to the examples and designs described herein, but is consistent with the widest scope of the principles and novel features disclosed herein.

Claims

1. A method for unmanned aerial vehicle path planning for military simulation, characterized in that, The method comprises the following steps: obtaining static obstacle information and dynamic obstacle information of a simulation scene; initializing start point and end point information of navigation, and constructing a priority queue based on key value size arrangement; taking the end point as the starting point, sequentially traversing each navigation point, when a target navigation point is traversed, screening neighbor navigation points of the target navigation point based on the static obstacle information, and adding the screened neighbor navigation points into the priority queue according to key value sorting, taking the neighbor navigation point with the minimum key value, updating information, and taking the neighbor navigation point as the next traversed navigation point, until the traversal is completed, to obtain an initial path; judging whether there is a navigation point affected by a dynamic obstacle in the initial path based on the dynamic obstacle information and the initial path; if yes, selecting the first affected navigation point as a new obstacle, re-planning a path between the new obstacle and the start point, and obtaining an updated path; judging whether there is a navigation point affected by a dynamic obstacle in the updated path based on the dynamic obstacle information and the updated path; if no, taking the updated path as a target path and outputting. 2.The UAV path planning method of claim 1, wherein, The static obstacle information comprises a contour position of a static obstacle; and / or The dynamic obstacle information comprises a contour position of a dynamic obstacle at each time. 3.The method of claim 1, wherein, A calculation formula of the key value of a navigation point s is as follows: wherein is the historical minimum cost value from the navigation point s to the end point, and is initially set to infinity, is the current minimum cost value from the navigation point s to the end point, is the heuristic value from the navigation point s to the start point, is the key value correction value, is the inflation coefficient, and is greater than or equal to 1, is the start point; wherein The calculation formula is as follows: wherein is a set of neighbor navigation points of the navigation point s, represents a navigation point to the navigation point the actual cost value of, is an end point; The calculation formula is as follows: The calculation formula is as follows: wherein, is the Manhattan distance from the start point s, is the Chebychev distance from the start point s, a, b are linear coefficients, is the cost modification factor from the start point s to the navigation point s, is the cost modification factor from the navigation point s to the navigation point , is the base cost value associated with the type of the navigation point . 4.The method of claim 3, wherein, A rectangular coordinate system is constructed with the navigation point s as the center, and the neighbor navigation points are classified according to the number of coordinate axes that change in the position of the neighbor navigation points relative to the navigation point s in the rectangular coordinate system, and the greater the coordinate axis change amount, the greater the preset value. 5.The UAV path planning method of claim 3, wherein, The step of taking the neighbor navigation point with the minimum key value and updating information is specifically as follows: comparing a historical minimum cost value of the neighbor navigation point s to the end point and a current minimum cost value; when the historical minimum cost value is greater than the current minimum cost value, taking the current minimum cost value as the historical minimum cost value of the neighbor navigation point s; when the historical minimum cost value is less than the current minimum cost value, setting the historical minimum cost value of the neighbor navigation point s as infinite. 6.The method of claim 3, wherein, The step of until the traversal is completed, to obtain the initial path is specifically as follows: when the key value of the target navigation point is less than or equal to the key values of all neighbor navigation points in the priority queue, and the historical minimum cost value of the target navigation point is equal to the current minimum cost value, the traversal is completed, and the initial path is obtained. 7.The UAV path planning method of claim 3, wherein, The method further comprises the following steps: when the initial path construction fails, attenuating an inflation coefficient or increasing a maximum traversal number, and then re-planning until the initial path is obtained; and / or when the path between the new obstacle and the start point is re-planned based on the new obstacle, setting an inflation coefficient as 1. 8.The method of Claim 1, wherein, The step of screening neighbor navigation points of the target navigation point based on the static obstacle information comprises the following steps: constructing a rectangular coordinate system with the target navigation point as the center, constructing a minimum bounding box of the unmanned aerial vehicle at the center of the rectangular coordinate system based on the size of the unmanned aerial vehicle; constructing a passable bounding box based on a preset minimum safety distance and the minimum bounding box; obtaining a target coordinate axis in which the position of the neighbor navigation point relative to the target navigation point changes; judging whether the neighbor navigation point is passable based on the passable bounding box, the target coordinate axis, and the static obstacle information; if no, deleting the neighbor navigation point. 9.The UAV path planning method of claim 1, wherein, The step of judging whether there is a navigation point affected by the dynamic obstacle in the initial path based on the dynamic obstacle information and the initial path comprises: Based on the dynamic obstacle information and the initial path, the relative speed vector and the relative coordinate vector of the dynamic obstacle and the UAV are obtained; Based on the relative speed vector and the relative coordinate vector, the minimum distance time of the dynamic obstacle and the UAV is obtained, and it is judged whether the minimum distance time is less than 0; If not, the minimum distance between the dynamic obstacle and the UAV is calculated; Based on the radius of the minimum enclosing sphere of the dynamic obstacle, the radius of the minimum enclosing sphere of the UAV, and the preset minimum safety distance, the minimum non-collision distance is obtained; It is judged whether the minimum distance is greater than the minimum non-collision distance; If not, the navigation point in the initial path that collides with the dynamic obstacle is calculated, and the to-be-selected new obstacle is marked and collected into the queue; The step of selecting the first affected navigation point as a new obstacle is specifically: According to the size of the minimum distance time, the to-be-selected new obstacles in the queue are sorted, and the to-be-selected new obstacle with the minimum minimum distance time is selected as a new obstacle.

10. An unmanned aerial vehicle path planning apparatus for military simulation, characterized by, Comprise: An information acquisition module is configured to acquire static obstacle information and dynamic obstacle information of a simulation scene; An initialization module is configured to initialize start point and end point information of navigation, and construct a priority queue based on key value size arrangement; An initial path planning module is configured to take the end point as the start point, and sequentially traverse each navigation point. When a target navigation point is traversed, neighbor navigation points of the target navigation point are filtered based on the static obstacle information, and the filtered neighbor navigation points are added to the priority queue according to key value sorting. The neighbor navigation point with the minimum key value is taken, information is updated, and the navigation point is taken as the next traversal navigation point, until the traversal is completed, and an initial path is obtained; A first dynamic obstacle judgment module is configured to judge whether there is a navigation point affected by a dynamic obstacle in the initial path based on the dynamic obstacle information and the initial path; A path re-planning module is configured to select the first affected navigation point as a new obstacle when there is a navigation point affected by a dynamic obstacle in the initial path, re-plan a path between the new obstacle and the start point, and obtain an updated path. A second dynamic obstacle judgment module is configured to judge whether there is a navigation point affected by a dynamic obstacle in the updated path based on the dynamic obstacle information and the updated path; An output module is configured to take the updated path as a target path and output when there is no navigation point affected by a dynamic obstacle in the updated path.

Images