An unmanned aerial vehicle path planning method based on a triangle partitioning technique and related devices
By combining triangulation technology and genetic algorithms, the path planning of unmanned aerial vehicles (UAVs) is optimized, solving the problems of large computational load and large error in existing technologies, and achieving efficient and reliable path planning.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GUANGDONG POWER GRID CO LTD
- Filing Date
- 2022-12-30
- Publication Date
- 2026-05-29
AI Technical Summary
Among existing UAV path planning methods, route planning based on genetic algorithms suffers from high computational cost and large error, resulting in poor efficiency in engineering applications.
A UAV path planning method based on triangulation technology is adopted. Multiple UAV inspection routes are obtained by pre-setting triangulation route method to form an inspection route set. This set is used as the initial particle swarm and combined with a pre-set genetic algorithm and a pre-set collision and crossover strategy to perform route optimization calculation, thereby optimizing route selection and crossover operations.
This reduces computational load, improves the reliability and efficiency of route planning, lowers computational pressure, and ensures the accuracy and efficiency of UAV path planning.
Smart Images
Figure CN116380063B_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of unmanned aerial vehicle (UAV) inspection technology, and in particular to a UAV path planning method and related apparatus based on triangulation technology. Background Technology
[0002] In recent years, due to their high mobility, ease of operation, and low cost, UAVs have been widely used in various monitoring scenarios. These advantages include the ability to traverse complex terrain using only cameras and various sensors, and the efficient, safe, and reliable data collection and transmission from distant targets. However, when UAVs are tasked with complex operating environments, planning shorter and safer routes to improve their efficiency has become a pressing issue.
[0003] In existing technologies, genetic algorithms are often used for route optimization. However, current genetic algorithms still have significant problems. For example, they require rasterization of the 2D map. When the 2D map has a high pixel count, the route dimension increases, thus increasing the computational demands. If the 2D map has a low pixel count, more raster cells need to be inserted, further increasing the algorithm's load. Furthermore, low pixel counts result in larger errors, hindering algorithm convergence. Based on these issues, using current genetic algorithms for route optimization in practical engineering is clearly limited and inefficient. Summary of the Invention
[0004] This application provides a UAV path planning method and related apparatus based on triangulation technology, which solves the technical problem that existing route planning methods based on genetic algorithms are limited by large computational load or large error, resulting in poor efficiency in engineering applications.
[0005] In view of this, the first aspect of this application provides a UAV path planning method based on triangulation technology, including:
[0006] Multiple drone inspection routes are obtained based on the preset triangulation method, forming a drone inspection route set.
[0007] Using the UAV inspection route set as the initial particle swarm, a pre-set genetic algorithm is used to perform route optimization calculations to obtain the optimal UAV route. The pre-set genetic algorithm optimizes route selection based on a preset triangular segmentation algorithm and performs crossover operations based on a preset collision crossover strategy.
[0008] Preferably, the step of obtaining multiple UAV inspection routes based on the preset triangulation method to form a UAV inspection route set includes:
[0009] Based on the preset triangulation method, various drone inspection routes are drawn according to the inspection start and end points;
[0010] A collision test is performed on the drawn UAV inspection route. If the test passes, the UAV inspection route is retained to form a UAV inspection route set.
[0011] Preferably, the step of initializing the UAV inspection route set as a particle swarm and using a pre-set genetic algorithm to perform route optimization calculations to obtain the optimal UAV route includes:
[0012] According to the pre-set genetic algorithm, the UAV inspection route set is used as the initial particle swarm, and the initial fitness value corresponding to each individual route is calculated at the same time.
[0013] Based on the initial fitness value and the preset triangular segmentation algorithm, an individual selection operation is performed on the initialized particle swarm to obtain a genetic particle swarm.
[0014] The genetic particle swarm is subjected to crossover operation based on a preset collision crossover strategy to obtain an updated particle swarm. The step of calculating the initial fitness value corresponding to each individual route is returned until the iteration condition is reached, at which point the optimal solution is obtained and the optimal solution is taken as the optimal UAV route.
[0015] Preferably, the step of performing crossover operations on the genetic particle swarm based on a preset collision crossover strategy to obtain an updated particle swarm includes:
[0016] Based on a preset collision crossover strategy, the genetic particle group that failed the collision test is crossovered with the genetic particle group that passed the collision test with a preset probability to obtain an updated particle group.
[0017] The second aspect of this application provides a UAV path planning device based on triangulation technology, comprising:
[0018] The route generation unit is used to obtain various UAV inspection routes based on the preset triangulation method, and form a UAV inspection route set.
[0019] The route selection unit is used to initialize the UAV inspection route set as a particle swarm, and use a preset genetic algorithm to perform route optimization calculation to obtain the optimal UAV route. The preset genetic algorithm optimizes route selection based on a preset triangular segmentation algorithm and performs crossover operation based on a preset collision crossover strategy.
[0020] Preferably, the route generation unit is specifically used for:
[0021] Based on the preset triangulation method, various drone inspection routes are drawn according to the inspection start and end points;
[0022] A collision test is performed on the drawn UAV inspection route. If the test passes, the UAV inspection route is retained to form a UAV inspection route set.
[0023] Preferably, the route selection unit includes:
[0024] The initial calculation subunit is used to initialize the particle swarm based on the UAV inspection route set according to the preset genetic algorithm, and simultaneously calculate the initial fitness value corresponding to each individual route.
[0025] The individual selection subunit is used to perform individual selection operations on the initialized particle swarm based on the initial fitness value and the preset triangular segmentation algorithm to obtain a genetic particle swarm.
[0026] The cross-iteration subunit is used to perform cross-operation on the genetic particle swarm based on a preset collision cross-operation strategy, obtain an updated particle swarm, and return to the initial calculation subunit until the iteration condition is met, then the optimal solution is obtained, and the optimal solution is used as the optimal UAV route.
[0027] Preferably, the cross-iteration subunit is specifically used for:
[0028] Based on a preset collision crossover strategy, the genetic particle group that failed the collision test is crossovered with the genetic particle group that passed the collision test with a preset probability to obtain an updated particle group.
[0029] A third aspect of this application provides a UAV path planning device based on triangulation technology, the device including a processor and a memory;
[0030] The memory is used to store program code and transmit the program code to the processor;
[0031] The processor is used to execute the UAV path planning method based on triangular segmentation technology as described in the first aspect, according to the instructions in the program code.
[0032] The fourth aspect of this application provides a computer-readable storage medium for storing program code for executing the UAV path planning method based on triangulation technology described in the first aspect.
[0033] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:
[0034] This application provides a UAV path planning method based on triangulation technology, including: obtaining multiple UAV inspection routes based on a preset triangulation method to form a UAV inspection route set; using the UAV inspection route set as an initial particle swarm, performing route optimization calculations using a preset genetic algorithm to obtain the optimal UAV route; optimizing route selection based on a preset triangulation algorithm using the preset genetic algorithm; and performing cross-operation based on a preset collision cross-operation strategy.
[0035] The UAV path planning method based on triangulation technology provided in this application does not obtain UAV inspection routes through a gridded model. Instead, it uses a pre-defined triangulation method to obtain multiple UAV inspection routes, forming a set of UAV inspection routes, thereby reducing computational load. In the process of route optimization using a genetic algorithm, specific algorithms are designed to optimize the selection and crossover operations of individual routes. This not only selects suitable genetic individuals but also reduces computational load, ensuring the reliability of route planning while minimizing computational pressure. Therefore, this application can solve the technical problem that existing route planning methods based on genetic algorithms are limited by large computational loads or large errors, resulting in poor efficiency in engineering applications. Attached Figure Description
[0036] Figure 1 A flowchart illustrating a UAV path planning method based on triangulation technology provided in this application embodiment;
[0037] Figure 2 A schematic diagram of the structure of a UAV path planning device based on triangulation technology provided in this application embodiment;
[0038] Figure 3 A schematic diagram illustrating the process of generating UAV inspection routes based on the preset triangulation method provided in this application embodiment;
[0039] Figure 4 This is a schematic diagram of the route optimization process based on a preset triangular segmentation algorithm provided in an embodiment of this application;
[0040] Figure 5 This is a schematic diagram illustrating the route optimization process based on a universally predefined triangular segmentation algorithm provided in an embodiment of this application.
[0041] Figure 6 A schematic diagram of the route optimization process based on a preset collision and intersection strategy provided in this application embodiment. Figure 1 ;
[0042] Figure 7 A schematic diagram of the route optimization process based on a preset collision and intersection strategy provided in this application embodiment. Figure 2 . Detailed Implementation
[0043] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present application.
[0044] For easier understanding, please refer to Figure 1 This application provides an embodiment of a UAV path planning method based on triangulation technology, comprising:
[0045] Step 101: Obtain multiple UAV inspection routes based on the preset triangulation method to form a UAV inspection route set.
[0046] Further, step 101 includes:
[0047] Based on the preset triangulation method, various drone inspection routes are drawn according to the inspection start and end points;
[0048] Collision tests are performed on the drawn drone inspection routes. If the test passes, the drone inspection routes are retained to form a set of drone inspection routes.
[0049] Please see Figure 3 The default triangulation method only requires initializing one vertex, for example... Figure 3 Point C in the diagram forms a complete triangular route L1(A, C, B) with the inspection start point A and end point B. This is the principle of the triangular route method. Based on this, various UAV inspection routes can be drawn. However, to improve the effectiveness of the routes, this embodiment includes a collision test, which checks whether the UAV will collide with obstacles while inspecting the currently drawn route.
[0050] For example, if a collision test is performed on route L1(A, C, B), and the test passes, it means that the drone will not collide with obstacles when flying on this route, and it can be included in the drone inspection route set; otherwise, this route is discarded and a new route is drawn.
[0051] Step 102: Using the UAV inspection route set as the initial particle swarm, a pre-set genetic algorithm is used to perform route optimization calculations to obtain the optimal UAV route. The pre-set genetic algorithm optimizes route selection based on a pre-set triangular segmentation algorithm and performs crossover operations based on a pre-set collision crossover strategy.
[0052] Further, step 102 includes:
[0053] According to the preset genetic algorithm, use the UAV inspection route set as the initial particle swarm, and calculate the initial fitness value corresponding to each individual route at the same time;
[0054] Based on the initial fitness value and the preset triangular segmentation algorithm, perform an individual selection operation on the initial particle swarm to obtain a genetic particle swarm;
[0055] Based on the preset collision crossover strategy, perform a crossover operation on the genetic particle swarm to obtain an updated particle swarm, and return to the step of calculating the initial fitness value corresponding to each individual route until the iteration condition is reached, then obtain the optimal solution, and use the optimal solution as the optimal UAV route.
[0056] Furthermore, based on the preset collision crossover strategy, perform a crossover operation on the genetic particle swarm to obtain an updated particle swarm, including:
[0057] Based on the preset collision crossover strategy, perform a crossover calculation on the genetic particle swarm that fails the collision test with the genetic particle swarm that passes the collision test with a preset probability to obtain an updated particle swarm.
[0058] It should be noted that in this embodiment, the UAV inspection route set directly generated by the preset triangular route method is used as the initial particle swarm, and there is no need to form a route based on the traditional grid processing. Compared with the traditional genetic algorithm, the triangular route method does not require grid processing of the map in the initialization stage, and naturally there is no operation of inserting grids. The entire initialization process is simple and fast, so it can reduce the amount of calculation to a certain extent.
[0059] In addition to selecting excellent individuals in the initial particle swarm through the evaluation method of the initial fitness value, this embodiment also sets a preset triangular segmentation algorithm to perform individual selection in the initial particle swarm, so as to obtain an optimized genetic particle swarm. The specific preset triangular segmentation algorithm utilizes the geometric theory that "the sum of the two sides of a triangle is greater than the third side". Please refer to Figure 4 , for the current route, some optimization operations can be performed. For example, randomly select a point D on side AC and a point E on side BC, and connect DE to obtain a sub-generation route L2(A, D, E, B). According to the route L1(A, C, B), its route length can be calculated as: DL1 = AD + CD + CE + EB; while the route length of the sub-generation route L2(A, D, E, B) obtained by the triangular segmentation algorithm is: DL2 = AD + DE + EB, DL1 - DL2 = (CD + CE) - DE; since CD + CE > DE, that is, DL1 - DL2 > 0, DL2 < DL1. Therefore, it can be seen that based on the preset triangular segmentation algorithm, the sub-generation route can be optimized, and a better route can be selected from the parent route, or a better sub-generation route can be generated; at this time, if the sub-generation route L2(A, D, E, B) passes the collision test, the parent route can be directly eliminated and the sub-generation route can be added to the genetic particle swarm.
[0060] To improve the universality of the preset triangular segmentation algorithm in this embodiment, the triangular segmentation is not limited to being used on two sides sharing a common vertex, and can also be performed on any two sides of the parent route. Please refer to Figure 5 , for example, the route L3(A, F, G, B) is one of the offspring routes obtained after triangular segmentation based on the route L2(A, D, E, B); it is not difficult to prove based on geometric properties that DL3 < DL2. Obviously, as long as the offspring route obtained after triangular segmentation passes the collision test, it will definitely eliminate the parent route, that is, the fitness of the qualified offspring route must be better than that of the parent route.
[0061] In this embodiment, the offspring routes that do not pass the collision test will not be directly eliminated. This embodiment configures a preset collision crossover strategy for crossover operations, and cross-calculates the genetic particle swarm that does not pass the collision test with the genetic particle swarm that passes the collision test with a preset probability to obtain an updated particle swarm.
[0062] Please refer to Figure 6 and Figure 7 , if S is an obstacle, L1 is the route that does not pass the collision test, and L2 is the route that passes the collision test. Assume that in a certain collision crossover operation, L1 and L2 are paired. After the collision crossover, new routes, that is, sub-routes L1' and L2', can be generated. Among them, L1' is the offspring route that passes the collision test, and L2' is the offspring route that does not pass the collision test. For the offspring routes obtained by performing collision crossover operations based on the preset collision crossover strategy, the following elimination rules are adopted: preferentially eliminate the offspring routes that do not pass the collision test; if both offspring do not pass the collision test, it means that this collision crossover operation is invalid, and the parent route is retained; if only one offspring route passes the collision test, it will replace the parent route that does not pass the collision test.
[0063] The traditional genetic algorithm pairs individuals pairwise without difference and then performs crossover operations; the purpose of the genetic algorithm using crossover operations is to generate more excellent offspring particles through the crossover of parent particles. The preset triangular segmentation method adopted in this embodiment for route segmentation to obtain offspring particles will definitely be superior to the parent particles. Therefore, in this embodiment, all particles are not cross-operated without difference. Instead, only the routes that do not pass the collision test are cross-operated. In order to improve the qualification rate of the offspring routes after crossover, the routes that do not pass the collision test are crossed with the routes that pass the collision test with a certain probability; the preset probability is set according to actual engineering experience and is not limited here.
[0064] After obtaining the updated particle swarm, the step of calculating the fitness value can be returned to continue calculating the fitness value of each individual along the path. This process is repeated iteratively until the optimal fitness value and the optimal solution are obtained. The optimal solution is the optimal UAV route obtained in this embodiment.
[0065] The UAV path planning method based on triangulation technology provided in this application does not obtain UAV inspection routes through a gridded model. Instead, it uses a preset triangulation method to obtain multiple UAV inspection routes, forming a set of UAV inspection routes, thereby reducing computational load. In the process of route optimization using a genetic algorithm, specific algorithms are designed to optimize the selection and crossover operations of individual routes. This not only selects suitable genetic individuals but also reduces computational load, ensuring the reliability of route planning while minimizing computational pressure. Therefore, this application can solve the technical problem that existing route planning methods based on genetic algorithms are limited by large computational loads or large errors, resulting in poor efficiency in engineering applications.
[0066] For easier understanding, please refer to Figure 2 This application provides an embodiment of a UAV path planning device based on triangulation technology, comprising:
[0067] The route generation unit 201 is used to obtain multiple UAV inspection routes based on the preset triangulation method, and form a UAV inspection route set.
[0068] The route selection unit 202 is used to initialize the particle swarm with the UAV inspection route set and perform route optimization calculation using a preset genetic algorithm to obtain the optimal UAV route. The preset genetic algorithm optimizes route selection based on a preset triangular segmentation algorithm and performs crossover calculation based on a preset collision crossover strategy.
[0069] Furthermore, the route generation unit 201 is specifically used for:
[0070] Based on the preset triangulation method, various drone inspection routes are drawn according to the inspection start and end points;
[0071] Collision tests are performed on the drawn drone inspection routes. If the test passes, the drone inspection routes are retained to form a set of drone inspection routes.
[0072] Furthermore, the route selection unit 202 includes:
[0073] The initial computation subunit 2021 is used to initialize the particle swarm based on the UAV inspection route set according to the preset genetic algorithm, and simultaneously calculate the initial fitness value corresponding to each individual route.
[0074] Individual selection subunit 2022 is used to perform individual selection operations on the initialized particle swarm based on the initial fitness value and the preset triangular segmentation algorithm to obtain the genetic particle swarm.
[0075] The cross-iteration subunit 2023 is used to perform cross-operation on the genetic particle swarm based on a preset collision cross-operation strategy, obtain an updated particle swarm, and return to the initial calculation subunit until the iteration condition is met, then the optimal solution is obtained, and the optimal solution is used as the optimal UAV route.
[0076] Furthermore, the cross-iteration subunit 2023 is specifically used for:
[0077] Based on a preset collision crossover strategy, the genetic particle swarm that failed the collision test is crossovered with the genetic particle swarm that passed the collision test with a preset probability to obtain an updated particle swarm.
[0078] This application also provides a drone route planning device, which includes a processor and a memory;
[0079] The memory is used to store program code and transfer the program code to the processor;
[0080] The processor is used to execute the UAV route planning method in the above method embodiments according to the instructions in the program code.
[0081] This application also provides a computer-readable storage medium for storing program code for executing the UAV route planning method in the above method embodiments.
[0082] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.
[0083] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0084] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.
[0085] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for executing all or part of the steps of the methods described in the various embodiments of this application through a computer device (which may be a personal computer, server, or network device, etc.). The aforementioned storage medium includes: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.
[0086] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.
Claims
1. A UAV path planning method based on triangulation technology, characterized in that, include: Multiple UAV inspection routes are obtained based on the preset triangulation method, forming a set of UAV inspection routes. The specific process is as follows: Based on the preset triangulation method, various drone inspection routes are drawn according to the inspection start and end points; A collision test is performed on the drawn UAV inspection route. If the test passes, the UAV inspection route is retained to form a UAV inspection route set. Using the UAV inspection route set as the initial particle swarm, a pre-set genetic algorithm is used to perform route optimization calculations to obtain the optimal UAV route. The pre-set genetic algorithm optimizes route selection based on a preset triangular segmentation algorithm and performs crossover operations based on a preset collision crossover strategy. The specific process is as follows: According to the pre-set genetic algorithm, the UAV inspection route set is used as the initial particle swarm, and the initial fitness value corresponding to each individual route is calculated at the same time. Based on the initial fitness value and the preset triangular segmentation algorithm, an individual selection operation is performed on the initialized particle swarm to obtain a genetic particle swarm. The genetic particle swarm is subjected to crossover operation based on a preset collision crossover strategy to obtain an updated particle swarm. The step of calculating the initial fitness value corresponding to each individual route is returned until the iteration condition is reached, then the optimal solution is obtained, and the optimal solution is taken as the optimal UAV route. The process of performing crossover operations on the genetic particle swarm based on a preset collision crossover strategy to obtain an updated particle swarm is as follows: Based on a preset collision crossover strategy, the genetic particle group that failed the collision test is crossovered with the genetic particle group that passed the collision test with a preset probability to obtain an updated particle group.
2. A UAV path planning device based on triangulation technology, characterized in that, include: The route generation unit is used to obtain multiple UAV inspection routes based on a preset triangulation method, forming a UAV inspection route set. Specifically, the route generation unit is used for: Based on the preset triangulation method, various drone inspection routes are drawn according to the inspection start and end points; A collision test is performed on the drawn UAV inspection route. If the test passes, the UAV inspection route is retained to form a UAV inspection route set. A route selection unit is used to initialize a particle swarm with the UAV inspection route set, and employ a pre-set genetic algorithm to perform route optimization calculations to obtain the optimal UAV route. The pre-set genetic algorithm optimizes route selection based on a preset triangular segmentation algorithm and performs crossover operations based on a preset collision crossover strategy. The route selection unit includes: The initial calculation subunit is used to initialize the particle swarm based on the UAV inspection route set according to the preset genetic algorithm, and simultaneously calculate the initial fitness value corresponding to each individual route. The individual selection subunit is used to perform individual selection operations on the initialized particle swarm based on the initial fitness value and the preset triangular segmentation algorithm to obtain a genetic particle swarm. The cross-iteration subunit is used to perform cross-operation on the genetic particle swarm based on a preset collision cross-operation strategy to obtain an updated particle swarm and return to the initial calculation subunit until the iteration condition is met, then the optimal solution is obtained and the optimal solution is used as the optimal UAV route. The cross-iteration subunit is specifically used for: Based on a preset collision crossover strategy, the genetic particle group that failed the collision test is crossovered with the genetic particle group that passed the collision test with a preset probability to obtain an updated particle group.
3. A UAV path planning device based on triangulation technology, characterized in that, The device includes a processor and a memory; The memory is used to store program code and transmit the program code to the processor; The processor is used to execute the UAV path planning method based on triangular segmentation technology as described in claim 1 according to the instructions in the program code.
4. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store program code for executing the UAV path planning method based on triangulation technology as described in claim 1.