A method and apparatus for planning a drone path

Abstract

Some embodiments of the present application provide a method and device for planning a path of a UAV, applied to the technical field of UAV path planning, wherein the method for planning the path of the UAV comprises: acquiring map information of a path planning area of the UAV, a starting point and an end point of a planned path; acquiring a plurality of density gradient rings between the starting point and the end point and a landing point probability of each density gradient ring in the plurality of density gradient rings; determining an intermediate node of the planned path according to the plurality of density gradient rings and the landing point probability, with the map information, the starting point and the end point as initial conditions; and generating a target path according to the starting point, the intermediate node and the end point. Some embodiments of the present application can realize efficient and accurate planning of the path of the UAV, and have good versatility.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned aerial vehicles, in particular to a method and device for planning a path of an unmanned aerial vehicle. BACKGROUND

[0002] In recent years, unmanned aerial vehicles have received more and more attention because they can efficiently and conveniently complete a variety of tasks in some application fields. As the application fields of unmanned aerial vehicles gradually expand, the research on path planning of unmanned aerial vehicles is becoming more and more important.

[0003] At present, in order to quickly plan a path of an unmanned aerial vehicle, a fast search random sampling is usually performed between a starting point and an end point of the unmanned aerial vehicle until the sampling point is the end point. However, it is found that the fast search random sampling method of the prior art cannot take into account both time and path cost. When the time cost is guaranteed, a good quality path of the unmanned aerial vehicle cannot be obtained, or when the path quality is guaranteed, a high time cost is required.

[0004] Therefore, how to provide a technical solution of an efficient method for planning a path of an unmanned aerial vehicle becomes a technical problem to be solved urgently. SUMMARY

[0005] Some embodiments of the present application aim to provide a method and device for planning a path of an unmanned aerial vehicle. The technical solution of the embodiments of the present application can realize accurate planning of a path of an unmanned aerial vehicle, obtain an optimal target path with good quality, and reduce time cost, and has good versatility.

[0006] In a first aspect, some embodiments of the present application provide a method for planning a path of an unmanned aerial vehicle, comprising: obtaining map information of a path planning area of the unmanned aerial vehicle, a starting point and an end point of a planned path; obtaining a plurality of density gradient rings between the starting point and the end point and a landing point probability of each density gradient ring in the plurality of density gradient rings, wherein the plurality of density gradient rings represent a direction of the planned path, and the landing point probability of each density gradient ring represents a probability of a node in the planned path of the unmanned aerial vehicle landing in a position of each density gradient ring; determining an intermediate node of the planned path according to the plurality of density gradient rings and the landing point probability with the map information, the starting point and the end point as initial conditions, and generating a target path according to the starting point, the intermediate node and the end point.

[0007] Some embodiments of the present application determine an intermediate node of a planned path of an unmanned aerial vehicle by obtaining a plurality of density gradient rings and corresponding landing point probabilities between a starting point and an end point of the unmanned aerial vehicle, and then obtain a target path. The method realizes accurate planning of a path of an unmanned aerial vehicle, obtains an optimal target path with good quality, reduces time cost, and has good versatility.

[0008] In some embodiments, the obtaining the plurality of density gradient rings between the start point and the end point comprises: equally dividing a straight line segment between the start point and the end point into a preset number of sub-line segments; constructing a plurality of gradient rings with the same center, the center being the end point, and the distance from the center to the proximal end point of the sub-line segment being the gradient ring radius, a circular ring between each two adjacent gradient rings in the plurality of gradient rings constituting one of the plurality of density gradient rings.

[0009] Some embodiments of the present application can obtain a plurality of density gradient rings by taking the end point as the center and equally dividing the line segment between the start point and the end point to obtain the gradient ring radius, thereby obtaining a higher-accuracy unmanned aerial vehicle planning path direction and a region in which intermediate nodes are scattered, providing exact direction support for subsequent unmanned aerial vehicle path planning.

[0010] In some embodiments, the intermediate nodes include at least one node, and the determining the intermediate nodes of the planning path according to the plurality of density gradient rings and the landing point probability comprises: obtaining a density gradient ring number in the plurality of gradient rings; generating a target density gradient ring in which any node of the at least one node is located according to the density gradient ring number and the corresponding landing point probability; generating an azimuth angle of the any node; obtaining an initial node according to the target density gradient ring and the azimuth angle; and obtaining the any node according to the initial node and a set step length.

[0011] Some embodiments of the present application can obtain the position of any node by the density gradient ring number and the landing point probability, thereby achieving accurate and rapid determination of the intermediate nodes.

[0012] In some embodiments, the obtaining the initial node according to the target density gradient ring and the azimuth angle comprises: taking the direction of the gradient ring radius corresponding to the target density gradient ring as a reference line; obtaining a target line segment between the end point and a target position point on the target density gradient ring, and the included angle between the target line segment and the reference line being the azimuth angle; and taking the target position point as the initial node.

[0013] Some embodiments of the present application can quickly obtain the initial node by the target density gradient ring and the azimuth angle.

[0014] In some embodiments, the obtaining the any node according to the initial node and the set step length comprises: obtaining an extended node closest to the initial node; taking the extended node as a start point, the direction of the line segment between the initial node and the extended node being a planning direction, and a position point satisfying the set step length being a to-be-detected node; and detecting the to-be-detected node to obtain the any node.

[0015] Some embodiments of the present application can quickly obtain a qualified to-be-detected node by an initial node and a set step length, and then obtain any node, thereby reducing node search time and improving efficiency.

[0016] In some embodiments, the detecting the to-be-detected node and obtaining the any node includes: when an intersection between a connection region between the to-be-detected node and the extended node and an obstacle region in the map information is an empty set, confirming that the to-be-detected node is a to-be-verified node; and when a distance between the to-be-verified node and the end point is less than a distance between the start point and the end point, confirming that the to-be-verified node is the any node.

[0017] Some embodiments of the present application can obtain an effective any node by detecting a to-be-detected node based on map information, and ensure that an obtained target path can not be affected by an obstacle.

[0018] In some embodiments, the method further includes: when the intersection between the connection region and the obstacle region is not an empty set, confirming that the to-be-detected node is any invalid node; and under uniform sampling conditions, re-collecting the any invalid node to obtain any valid node, wherein the any valid node is taken as one of the intermediate nodes.

[0019] Some embodiments of the present application can obtain a more accurate intermediate node by re-collecting any invalid node by uniform sampling.

[0020] In some embodiments, the generating a target path according to the start point, the intermediate nodes, and the end point includes: connecting the start point, the intermediate nodes, and the end point to obtain an initial planning path of the unmanned aerial vehicle; and optimizing the initial planning path to obtain the target path.

[0021] Some embodiments of the present application can ensure the quality of the target path by optimizing the initial planning path to obtain the target path.

[0022] In some embodiments, the optimizing the initial planning path to obtain the target path includes: in a sequence from the end point to the start point, cyclically performing the following operations until an a+1+bth node is the start point and the end point is connected to all updated nodes to obtain the target path: obtaining an ath node in the initial planning path; obtaining the a+1+bth node and detecting whether the ath node and the a+1+bth node are connected to obtain a detection result; confirming that the detection result is that the ath node and the a+1+bth node are not connected, and updating the ath+bth node to an a+1th node.

[0023] Some embodiments of the present application can achieve verification and optimization of the initial planning path by iterative traversal from the end point to the start point, so that the target path is the optimal path between the start point and the end point, and the quality of the target path is optimal.

[0024] In a second aspect, some embodiments of the present application provide a device for planning a path of a UAV, comprising: an information acquisition module configured to acquire map information of a path planning area of a UAV, a start point and an end point of a planning path; a gradient ring acquisition module configured to acquire a plurality of density gradient rings between the start point and the end point and a landing point probability of each density gradient ring in the plurality of density gradient rings, wherein the plurality of density gradient rings represent a direction of the planning path, and the landing point probability of each density gradient ring represents a probability of a node in the planning path of the UAV landing in a position in each density gradient ring; and a path generation module configured to determine an intermediate node of the planning path according to the plurality of density gradient rings and the landing point probability, and generate a target path according to the start point, the intermediate node and the end point, with the map information, the start point and the end point as initial conditions.

[0025] In a third aspect, some embodiments of the present application provide a computer readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, can implement the method of any one of the embodiments of the first aspect.

[0026] In a fourth aspect, some embodiments of the present application provide an electronic device, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, can implement the method of any one of the embodiments of the first aspect.

[0027] In a fifth aspect, some embodiments of the present application provide a computer program product, comprising a computer program, wherein the computer program, when executed by a processor, can implement the method of any one of the embodiments of the first aspect. BRIEF DESCRIPTION OF DRAWINGS

[0028] In order to more clearly illustrate the technical solutions of some embodiments of the present application, the following will briefly introduce the drawings needed to be used in some embodiments of the present application. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor.

[0029] Figure 1 A system diagram for planning a path of a UAV is provided for some embodiments of the present application;

[0030] Figure 2 One of method flow charts for planning a UAV path provided for some embodiments of the present application;

[0031] Figure 3 Obstacle region diagram provided for some embodiments of the present application;

[0032] Figure 4 One of density gradient ring diagram provided for some embodiments of the present application;

[0033] Figure 5 Second density gradient ring diagram provided for some embodiments of the present application;

[0034] Figure 6 Second method flow chart for planning a UAV path provided for some embodiments of the present application;

[0035] Figure 7 Target path diagram provided for some embodiments of the present application;

[0036] Figure 8 Device composition block diagram for planning a UAV path provided for some embodiments of the present application;

[0037] Figure 9 Electronic device diagram provided for some embodiments of the present application. DETAILED DESCRIPTION

[0038] The technical solutions in the embodiments of the present application will be described below with reference to the drawings.

[0039] It should be noted that similar reference numerals and letters refer to similar items throughout the accompanying drawings, and thus once an item is defined in one drawing, it is not necessary to further define and explain it in subsequent drawings. Meanwhile, in the description of the present application, the terms "first", "second", etc. are only used to distinguish description, and cannot be understood as indicating or implying relative importance.

[0040] In the related art, a Rapidly-exploring Random Tree (RRT) algorithm is used to solve a robot path planning problem. By iteratively expanding through the application of control inputs, the system is gradually biased towards a selected random point, such that the path spreads out like a tree from the root to the branches, and an initial feasible path can be obtained when the branches search to the target point (i.e., the end point). In other words, the path solving idea of the RRT algorithm is to quickly expand like a tree, and the path spreads out from the starting point to fill the search space step by step until it expands to the target point. As a common sampling-based path planning algorithm, the RRT algorithm has the greatest advantage of being able to quickly search for a feasible path. From the above related art, it can be seen that the RRT algorithm can only guarantee to search for an initial feasible path as quickly as possible, but cannot guarantee the quality of the initial feasible path (for example, the length of the initial feasible path), and in a complex map, the RRT algorithm is easy to fall into a local error zone, causing time redundancy.

[0041] In view of this, some embodiments of the present application provide a method and device for planning a UAV path. The method can determine intermediate nodes in the planned path by using multiple density gradient rings and landing point probabilities between the starting point and the end point of the UAV path, and then generate a target path for the UAV. Some embodiments of the present application can accurately plan the UAV path, obtain a good quality and optimal target path between the starting point and the end point, and reduce the time cost, which has good versatility.

[0042] As shown in Figure 1 some embodiments of the present application provide a system for planning a UAV path. The system includes a terminal 200. A user 100 can input map information, a starting point and an end point of a UAV in the terminal 200. The terminal 200 can construct multiple density gradient rings according to the starting point and the end point of the UAV, and then determine intermediate nodes of the UAV path according to the multiple density gradient rings and the landing point probabilities under the condition of the map information, the starting point and the end point of the UAV, and then obtain an initial planned path. Finally, the initial planned path is optimized to obtain a target path.

[0043] In some embodiments of the present application, the terminal 200 can be a mobile terminal or a non-portable computer terminal, which is not specifically limited herein.

[0044] The implementation process of the terminal 200 for planning a UAV path is described below in conjunction with the accompanying Figure 2 some embodiments of the present application provide an implementation process of the terminal 200 for planning a UAV path.

[0045] Please refer to the accompanying Figure 2 , Figure 2 flowchart of the method for planning a UAV path provided by some embodiments of the present application. The method for planning a UAV path includes:

[0046] S210, acquire map information of a path planning area of the UAV, a start point and an end point of a planned path.

[0047] In some embodiments of the present application, the map of the path planning of the UAV can be input into the terminal 200, that is, the information of the obstacle area, the start point and the target point (as an example of the end point) in the map information can be obtained. As shown in the figure, Figure 3 the gray square is the obstacle, Figure 3 the triangle in the lower half of the figure is the start point, Figure 3 and the triangle in the upper half of the figure is the target point.

[0048] S220, acquire a plurality of density gradient rings between the start point and the end point and a landing point probability of each density gradient ring in the plurality of density gradient rings, wherein the plurality of density gradient rings represent the direction of the planned path, and the landing point probability of each density gradient ring represents the probability of the node in the planned path of the UAV landing in the position of each density gradient ring.

[0049] For example, in some embodiments of the present application, the idea of sampling gradient ascent is adopted, that is, the rising density gradient rings between the start point and the target point are constructed, and the landing point probability of the sampling point (that is, the node in the planned path) is set, and the global sampling of the guide point is guided on the premise of the landing point probability. The guide point can be used as a candidate node for obtaining the final intermediate node.

[0050] In some embodiments of the present application, S220 can include: dividing a straight line segment between the start point and the end point into a preset number of sub-line segments; constructing a plurality of gradient rings with the same center, the distance from the center to the proximal end point of the sub-line segment as the radius of the gradient ring, the annulus between each two adjacent gradient rings in the plurality of gradient rings constitutes one of the plurality of density gradient rings.

[0051] For example, in some embodiments of the present application, the path expansion direction (that is, the direction from the start point to the target point) is the density concentration direction of the traction point, and the landing point probability is also a dynamic change process. The distance l between the start point x init and the target point x goal is calculated and divided into m equal sub-line segments, and x goal is used as the center to construct m concentric circles (one concentric circle corresponds to one gradient ring), as shown in Figure 4 The radius of each circle from inside to outside is:

[0052]

[0053] wherein r iis the radius of the ith concentric circle (also referred to as the ith density gradient ring), i is an integer, and i ∈ [1, m]. Wherein, the radius of each concentric circle is the distance between the nearest end point of the two end points of each sub-segment and the center of the circle.

[0054] In addition, Figure 4 x in new is an example of an intermediate node after the starting point, and Outside is an external area outside the mth circle.

[0055] In some embodiments of the present application, the probability of incrementally increasing the sampling points from the outside to the inside is set when the gradient ring is constructed, that is, the drop point probability. As shown in Figure 5 The annular area (one annular area is one density gradient ring) is shown in goal The first circle where the x i point is located is the internal area, and the annular area from the internal area to the outermost circle is Ω new . As the random tree of the planned path continues to explore, the x goal gradually advances to x new , the density gradient ring gradually shrinks, and the Outside continuously expands. With the x new position as a reference, when the x i+1 advances from Ω i , the drop point probability of each density gradient ring and the corresponding guide point also changes:

[0056]

[0057] Wherein, P i is the drop point probability of the ith density gradient ring, is the initial drop point probability of the ith density gradient ring, is the initial drop point probability of the (i+1)th density gradient ring, is the initial drop point probability of the jth density gradient ring, j ∈ [1, i], and P Outside is the drop point probability of the external area.

[0058] S230, taking the map information, the starting point and the end point as initial conditions, determining the intermediate nodes of the planned path according to a plurality of density gradient rings and the drop point probability, and generating a target path according to the starting point, the intermediate nodes and the end point.

[0059] For example, in some embodiments of the present application, the direction of the intermediate node planning can be obtained through the rising gradient direction of the density gradient ring, the position of the intermediate node can be determined through the drop point probability, and then the final target path is generated based on the starting point, the intermediate node and the target point.

[0060] In some embodiments of the present application, the intermediate nodes include at least one node, and S230 can include: obtaining a number of density gradient rings in the plurality of gradient rings; generating a target density gradient ring in which any node of the at least one node is located according to the number of density gradient rings and the corresponding landing probability; generating an azimuth angle of the any node; obtaining an initial node according to the target density gradient ring and the azimuth angle; and obtaining the any node according to the initial node and a set step length.

[0061] For example, in some embodiments of the present application, the density gradient ring is gradually reduced in the process of generating any intermediate node, as shown in Figure 5 After obtaining two nodes x1 and x2 (that is, two nodes in the intermediate nodes) in the mth density gradient ring, x new enters the (m-1)th density gradient ring, and the mth density gradient ring disappears at this time. Therefore, when obtaining any node, the number of the current density gradient ring needs to be known first, and then the target density gradient ring in which any node is located is determined based on the number of density gradient rings and the corresponding landing probability obtained by the above formula. Then, the position of any node is obtained in combination with the determined azimuth angle of any node and the set step length, where the position can be represented in the form of coordinates.

[0062] In some embodiments of the present application, S230 can include: taking the direction of the gradient ring radius corresponding to the target density gradient ring as a reference line; obtaining a target line segment of the terminal point and a target position point on the target density gradient ring, and the included angle between the target line segment and the reference line is the azimuth angle; and taking the target position point as the initial node.

[0063] For example, in some embodiments of the present application, the direction of l in Figure 4 is taken as the reference line, and then the intersection of the center of the circle as the target line segment and the reference line is taken, and the target position point with the target line segment having an azimuth angle size with the reference line and the terminal point of the target line segment falling on the arc boundary of the target density gradient ring is taken as the initial node x rand .

[0064] In some embodiments of the present application, S230 can include: obtaining an extended node closest to the initial node; taking the extended node as a starting point, the direction of the line connecting the initial node and the extended node as a planning direction, and a position point satisfying the set step length as a to-be-detected node; and detecting the to-be-detected node to obtain the any node.

[0065] For example, in some embodiments of the present application, the given map information also contains a plurality of traction points obtained by preliminary search. Then, the point closest to the initial node in the plurality of traction points is taken as the extended node xnearest Finally, the direction of the line connecting x rand and x nearest is taken as the planning direction of the node, and the point at which the step length is set in the planning direction is taken as the to-be-detected node x new , and then any node is obtained.

[0066] In some embodiments of the present application, the detecting the to-be-detected node and obtaining the any node include: when the intersection between the line region between the to-be-detected node and the extended node and the obstacle region in the map information is an empty set, confirming that the to-be-detected node is a to-be-verified node; and when the distance between the to-be-verified node and the terminal point is less than the distance between the starting point and the terminal point, confirming that the to-be-verified node is the any node.

[0067] For example, in some embodiments of the present application, if the line between the to-be-detected node and x nearest does not touch any obstacle in x Figure 3 , it is considered that the intersection between the line region between the to-be-detected node and the extended node and the obstacle region is an empty set, that is, it is confirmed that the to-be-detected node can be connected, and the to-be-verified node is obtained at this time. If the distance between the to-be-verified node and the terminal point is less than the distance between the starting point and the terminal point, it is indicated that the to-be-verified node is still any of the effective intermediate nodes and has not exceeded the planning path region.

[0068] In some embodiments of the present application, the method for planning the path of the unmanned aerial vehicle further includes: when the intersection between the line region and the obstacle region is not an empty set, confirming that the to-be-detected node is any invalid node; and under uniform sampling conditions, re-collecting the any invalid node to obtain any valid node, wherein the any valid node is taken as one of the intermediate nodes.

[0069] For example, in some embodiments of the present application, in the process of obtaining the intermediate node, an invalid node, that is, a node that fails to pass the detection, can be obtained. The above-mentioned method for collecting any node uses a dynamic sampling method, and the invalid node can be re-collected in a uniform sampling manner until the detection and verification pass, and then any valid node is obtained.

[0070] In order to obtain the optimal path, in some embodiments of the present application, S230 can include: connecting the starting point, the intermediate node, and the terminal point to obtain an initial planning path of the unmanned aerial vehicle; and optimizing the initial planning path to obtain the target path.

[0071] For example, in some embodiments of the present application, after the initial planning path is obtained by connecting the starting point, intermediate nodes and the end point, root node iteration is performed (as a specific example of optimization) to obtain the optimal target path.

[0072] In some embodiments of the present application, S230 can include: performing the following operations in a loop in the order of the end point to the starting point until the a+1+bth node is the starting point, and connecting the end point to all updated nodes to obtain the target path: obtaining the ath node in the initial planning path; obtaining the a+1+bth node and detecting whether the ath node and the a+1+bth node are connected, obtaining a detection result; confirming that the detection result is that the ath node and the a+1+bth node are not connected, and updating the a+bth node to the a+1th node.

[0073] For example, in some embodiments of the present application, root node iteration is performed in the direction from the end point to the starting point. For example, there are 10 nodes in the initial planning path, which are end point 1, node 2, node 3, …, node 9 and starting point 10. In the process of the first loop, end point 1 and node 2 must be connected, so it is necessary to determine whether end point 1 and node 3 are connected, and if the connection line between end point 1 and node 3 is an empty set, it is confirmed that the detection result is passed, that is, end point 1 and node 3 can be connected. Then, node 1 and node 4 are detected, and if they are not connected, node 3 is updated to a new node 2 in the target path, and node 2 is removed from the initial planning path. Next, the above process is performed on other nodes with the new node 2 as the starting point until the next node is traversed under the unconnected condition, and so on until the starting point 1 is traversed to end, and finally the optimal target path can be obtained.

[0074] The specific process of the method for planning a path provided by some embodiments of the present application will be described below with reference to the accompanying drawings. Figure 6 The specific process of the method for planning a path provided by some embodiments of the present application will be described below with reference to the accompanying drawings.

[0075] The specific process of the method for planning a path provided by some embodiments of the present application will be described below with reference to the accompanying drawings. Figure 6 , Figure 6 The specific process of the method for planning a path provided by some embodiments of the present application will be described below with reference to the accompanying drawings.

[0076] S610, obtaining the obstacle region in the map information, the distance L between the starting point and the end point, the step length len, the iteration number n=50, and the effective sampling number h=0.

[0077] S620, obtaining a plurality of density gradient circles from the starting point to the end point and corresponding landing point probabilities.

[0078] S630, determining whether n is greater than 0, if yes, performing S640, otherwise performing S690.

[0079] S640, acquiring x new and an extended node x nearest .

[0080] S650, determining whether the intersection of the distance k between x new and x nearest and the obstacle region is an empty set, if yes, executing S660, otherwise n = n-1, and then executing S630.

[0081] S660, determining whether k < L, if yes, executing S670, otherwise executing S680.

[0082] S670, adding x new to the intermediate node set, and setting h = h+1, and then executing S630.

[0083] S680, adding x new to the invalid point set, and then executing S630.

[0084] S690, determining whether H = 50-h is greater than 0, if yes, executing S691, otherwise executing S693.

[0085] S691, acquiring x new , an extended node x nearest and the distance Y between them by using uniform sampling corresponding to H.

[0086] S692, if the intersection of Y and the obstacle region is an empty set and less than L, executing S670, otherwise setting h = h+2a, and returning to S690.

[0087] Wherein, a is a decay coefficient.

[0088] In some embodiments of the present application, in order to adapt to complex map environment, after the dynamic density gradient sampling strategy is constructed, the map complexity coefficient Q and the decay coefficient a (a = 1-Q) are calculated in the sampling stage to obtain the iteration number H of uniform sampling after the dynamic density gradient sampling is completed. In this way, the use frequency of dynamic density gradient sampling and uniform sampling in different environments is dynamically changed to prevent the random tree of path planning from falling into a local error area in the process of search expansion, wherein:

[0089]

[0090] Wherein, n all is the total number of random guide points generated in the iteration process, h is the current number of valid guide points (that is, the number of nodes in the intermediate set), M is the number of stage random guide point dynamic gradient sampling inspection, and m effis the number of effective sampling points in M times, and H is the number of random and uniform sampling in the whole area. In the above execution process, H determines the number of iterations of uniform sampling in the sampling algorithm, and the size of H changes with the change of the map complexity coefficient Q. The larger the calculated Q is, the higher the complexity is, the smaller the decay coefficient a is, and the larger the finally calculated H is, which means that the frequency of using uniform sampling increases, and the sampling points will be freed from the local error area by mixing more times of global uniform sampling. On the contrary, in the simpler map, the calculated Q and H are smaller, and the sampling selection in the above algorithm execution will be more inclined to dynamic gradient sampling.

[0091] S693, obtaining an initial planning path according to the starting point, the intermediate node set and the ending point.

[0092] S694, optimizing the initial planning path to obtain a target path.

[0093] In some embodiments of the present application, the root iteration method provided in the embodiments of the present application can be used to finally obtain the target path shown in FIG. 8: x Figure 2 Figure 7 In some embodiments of the present application, the root iteration method provided in the embodiments of the present application can be used to finally obtain the target path shown in FIG. 8: x inti , x1, x2,..., x n , x new , x goal . To avoid repetition, the present application will not be described here.

[0094] In some embodiments of the present application, the root iteration method provided in the embodiments of the present application can be used to finally obtain the target path shown in FIG. 8: x

[0095] For example, as a specific example of the present application, before obtaining the target path, it is necessary to go through two stages of touch detection (that is, to judge whether the connection between two nodes is empty set with the obstacle area) and node update and constantly loop until the stop condition is met, that is, the updated current root node is the starting point or the current reference starting point is the starting point. In the state of the initial planning path, the target point x goal is the reference starting point x star iteratively reads its root node, and returns the parameter Free after being verified by the touch detection function:

[0096]

[0097] Wherein, x star is the reference starting point, is the root node of x star , and β obstacle represents the obstacle area. After touch detection, the return value of Free is iteratively updated, the root node of the current reference starting point is selected after updating, and the touch detection is continued, or the reference starting point is updated and the touch detection is continued:

[0098] ​

[0099] When returning Free = 1, x parent After storing the current root node information, the next root node is updated as its current root node. When returning Free = 0, the new reference starting point is updated as the stored x parent The above touch detection and root node update are repeated until the latest point iteration is the starting point x init And iteration is started from the target point x goal The final target path is obtained by reading the root node from the starting point.

[0100] Please refer to Figure 8 , Figure 8 The composition block diagram of the device for planning the path of the unmanned aerial vehicle is shown. It should be understood that the device for planning the path of the unmanned aerial vehicle corresponds to the above-mentioned method embodiments, and can perform various steps involved in the above-mentioned method embodiments. The specific functions of the device for planning the path of the unmanned aerial vehicle can be referred to the description in the above, and the detailed description is appropriately omitted here to avoid repetition.

[0101] Figure 8 The device for planning the path of the unmanned aerial vehicle includes at least one software function module stored in the form of software or firmware in the memory or solidified in the device for planning the path of the unmanned aerial vehicle, and the device for planning the path of the unmanned aerial vehicle includes: an information acquisition module 810 configured to acquire map information of a path planning area of an unmanned aerial vehicle, a starting point and an end point of a planned path; a gradient ring acquisition module 820 configured to acquire a plurality of density gradient rings between the starting point and the end point and a landing point probability of each density gradient ring in the plurality of density gradient rings, wherein the plurality of density gradient rings represent the direction of the planned path, and the landing point probability of each density gradient ring represents the probability of a node in the planned path of the unmanned aerial vehicle falling in the position of each density gradient ring; a path generation module 830 configured to determine an intermediate node of the planned path according to the plurality of density gradient rings and the landing point probability with the map information, the starting point and the end point as initial conditions, and generate a target path according to the starting point, the intermediate node and the end point.

[0102] In some embodiments of the present application, the gradient ring acquisition module 820 is configured to equally divide a straight line segment between the starting point and the end point into a preset number of sub-line segments; a plurality of gradient rings with the same center are constructed with the end point as the center, and the distance from the center to the proximal end point of the sub-line segment as the gradient ring radius, and the annulus between every two adjacent gradient rings in the plurality of gradient rings constitutes one of the plurality of density gradient rings.

[0103] In some embodiments of the present application, the intermediate nodes include at least one node, wherein the path generation module 830 is configured to obtain a number of density gradient rings in the plurality of gradient rings; generate a target density gradient ring in which any node of the at least one node is located according to the number of density gradient rings and the corresponding landing probability; generate an azimuth angle of the any node; obtain an initial node according to the target density gradient ring and the azimuth angle; and obtain the any node according to the initial node and a set step length.

[0104] In some embodiments of the present application, the path generation module 830 is configured to take a direction of the gradient ring radius corresponding to the target density gradient ring as a reference line; obtain a target line segment between the end point and a target position point on the target density gradient ring, and an included angle between the target line segment and the reference line is the azimuth angle; and take the target position point as the initial node.

[0105] In some embodiments of the present application, the path generation module 830 is configured to obtain an extended node closest to the initial node; take the extended node as a starting point, a connecting line direction between the initial node and the extended node as a planning direction, and a position point satisfying the set step length as a to-be-detected node; and detect the to-be-detected node to obtain the any node.

[0106] In some embodiments of the present application, the path generation module 830 is configured to confirm that the to-be-detected node is a to-be-verified node when an intersection between a connecting line region between the to-be-detected node and the extended node and an obstacle region in the map information is an empty set; and confirm that the to-be-verified node is the any node when a distance between the to-be-verified node and the end point is less than a distance between the starting point and the end point.

[0107] In some embodiments of the present application, the path generation module 830 is configured to confirm that the to-be-detected node is any invalid node when the intersection between the connecting line region and the obstacle region is not an empty set; and re-collect any valid node under uniform sampling conditions, wherein the any valid node is taken as one of the intermediate nodes.

[0108] In some embodiments of the present application, the path generation module 830 is configured to connect the starting point, the intermediate nodes and the end point to obtain an initial planning path of the unmanned aerial vehicle; and optimize the initial planning path to obtain the target path.

[0109] In some embodiments of the present application, the path generation module 830 is configured to perform the following operations in a loop in the order from the end point to the start point until the a+1+bth node is the start point and the end point is connected to all the updated nodes: obtaining the ath node in the initial planning path; obtaining the a+1+bth node and detecting whether the ath node is connected to the a+1+bth node to obtain a detection result; and confirming that the detection result is that the ath node is not connected to the a+1+bth node, and updating the a+bth node to the a+1th node.

[0110] Some embodiments of the present application also provide a computer readable storage medium having stored thereon a computer program, which, when executed by a processor, can implement the operations of the method corresponding to any of the above embodiments.

[0111] Some embodiments of the present application also provide a computer program product, which includes a computer program, wherein the computer program, when executed by a processor, can implement the operations of the method corresponding to any of the above embodiments.

[0112] As shown in Figure 9 Some embodiments of the present application provide an electronic device 900, which includes a memory 910, a processor 920, and a computer program stored in the memory 910 and executable on the processor 920, wherein the processor 920 reads the program from the memory 910 through a bus 930 and executes the program to implement the method of any of the above embodiments.

[0113] The processor 920 can process digital signals and can include various computing structures, such as a complex instruction set computer structure, a reduced instruction set computer structure, or a structure implementing a combination of multiple instruction sets. In some examples, the processor 920 can be a microprocessor.

[0114] The memory 910 can be used to store instructions executed by the processor 920 or data related to the execution of the instructions. These instructions and / or data can include code for implementing some or all of the functions of one or more modules described in the embodiments of the present application. The processor 920 of the embodiments of the present disclosure can be used to execute the instructions in the memory 910 to implement the above-described method. The memory 910 includes a dynamic random access memory, a static random access memory, a flash memory, an optical memory, or other memories well known to those skilled in the art.

[0115] The above merely provides an example of the present application and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application. It should be noted that similar reference numerals and letters represent similar items in the following drawings, and thus, once an item is defined in one drawing, it need not be further defined and explained in subsequent drawings.

[0116] The above merely provides an example of the present application and is not intended to limit the protection scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application. It should be noted that similar reference numerals and letters represent similar items in the following drawings, and thus, once an item is defined in one drawing, it need not be further defined and explained in subsequent drawings.

[0117] It should be noted that the relational terms herein such as first and second and the like are used solely to distinguish one entity or action from another, without necessarily requiring or implying any actual relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. An element proceeded by "comprises... a" does not, without more constraints, exclude the existence of additional identical elements in the process, method, article, or apparatus that comprises the element.

Claims

1. A method of planning a path for a drone, the method comprising: The method comprises: obtaining map information of a path planning area of a UAV, a starting point and an ending point of a planned path; obtaining a plurality of density gradient rings between the starting point and the ending point and a landing probability of each density gradient ring in the plurality of density gradient rings, wherein the plurality of density gradient rings represent a direction of the planned path, and the landing probability of each density gradient ring represents a probability of a node in the planned path of the UAV landing at a position in the each density gradient ring; determining an intermediate node of the planned path according to the plurality of density gradient rings and the landing probability based on the map information, the starting point and the ending point as initial conditions, and generating a target path according to the starting point, the intermediate node and the ending point; the obtaining of the plurality of density gradient rings between the starting point and the ending point comprises: equally dividing a straight line segment between the starting point and the ending point into a preset number of sub-line segments; constructing a plurality of gradient rings with the same center, taking the ending point as the center, and taking a distance from the center to a proximal end point of the sub-line segment as a gradient ring radius, wherein a circular ring between each two adjacent gradient rings in the plurality of gradient rings constitutes one of the plurality of density gradient rings; Density gradient rings from target point The first circle where the target point is located is the inner region, and the annular regions from the inner region to the outermost circle are the intermediate regions ; as the random tree of the planned path continues to explore, the intermediate nodes gradually advance to , the density gradient rings gradually shrink, Outside and the density gradient rings continue to expand; with the position as the reference, when advancing from to , the landing probability of each density gradient ring and the corresponding guide point also changes: wherein, is the landing probability of the i-th density gradient ring, is the initial landing probability of the i-th density gradient ring, is the initial landing probability of the i+1-th density gradient ring, is the initial landing probability of the j-th density gradient ring, j ∈ [1, i ], is the landing probability of the outer region; the intermediate node comprises at least one node, wherein the determining of the intermediate node of the planned path according to the plurality of density gradient rings and the landing probability comprises: obtaining a number of density gradient rings in the plurality of gradient rings; generating a target density gradient ring in which any node of the at least one node is located according to the number of density gradient rings and the corresponding landing probability; generating an azimuth angle of the any node; obtaining an initial node according to the target density gradient ring and the azimuth angle; and obtaining the any node according to the initial node and a set step length; the generating of the target path according to the starting point, the intermediate node and the ending point comprises: connecting the starting point, the intermediate node and the ending point to obtain an initial planned path of the UAV; and performing the following operations in a loop in a sequence from the ending point to the starting point until an a+1+bth node is the starting point, and connecting the ending point to all updated nodes to obtain the target path: obtaining an ath node in the initial planned path; obtaining the a+1+bth node and detecting whether the ath node and the a+1+bth node are connected to obtain a detection result; and confirming that the detection result is that the ath node and the a+1+bth node are not connected, and updating the a+bth node to an a+1th node.

2. The method of claim 1, wherein, the obtaining of the initial node according to the target density gradient ring and the azimuth angle comprises: taking a direction of the gradient ring radius corresponding to the target density gradient ring as a reference line; obtaining a target line segment between the ending point and a target position point on the target density gradient ring, and taking an included angle between the target line segment and the reference line as the azimuth angle; taking the target position point as the initial node.

3. The method of claim 1 or 2, wherein, the obtaining of the any node according to the initial node and the set step length comprises: acquire an extended node closest to the initial node; take the extended node as a starting point, and a direction from the initial node to the extended node as a planning direction, and take a position point satisfying the set step length as a to-be-detected node; detect the to-be-detected node, and acquire any node.

4. The method of claim 3, wherein, The detection of the to-be-detected node and the acquisition of the any node include: when an intersection between a connection region between the to-be-detected node and the extended node and an obstacle region in the map information is an empty set, confirm that the to-be-detected node is a to-be-verified node; when a distance between the to-be-verified node and the terminal point is less than a distance between the starting point and the terminal point, confirm that the to-be-verified node is the any node.

5. The method of claim 4, wherein, The method further includes: when the intersection between the connection region and the obstacle region is not an empty set, confirm that the to-be-detected node is any invalid node; under uniform sampling conditions, reacquire the any invalid node, and acquire any valid node, wherein the any valid node is taken as one of the intermediate nodes.

6. An apparatus for planning a path of a drone, the apparatus comprising: The device is configured to execute the method of claim 1, and includes: an information acquisition module configured to acquire map information of a path planning region of a UAV, a starting point and a terminal point of a planned path; a gradient ring acquisition module configured to acquire a plurality of density gradient rings between the starting point and the terminal point and a landing point probability of each density gradient ring in the plurality of density gradient rings, wherein the plurality of density gradient rings represent directions of the planned path, and the landing point probability of each density gradient ring represents a probability that a node in the planned path of the UAV lands at a position in the each density gradient ring; a path generation module configured to take the map information, the starting point and the terminal point as initial conditions, determine intermediate nodes of the planned path according to the plurality of density gradient rings and the landing point probability, and generate a target path according to the starting point, the intermediate nodes and the terminal point.

Images