CN-121475241-B - Double-layer improved particle swarm unmanned plane path planning method, system, equipment and medium

CN121475241BCN 121475241 BCN121475241 BCN 121475241BCN-121475241-B

Abstract

The invention belongs to the field of unmanned aerial vehicle path planning, and provides a method, a system, equipment and a medium for planning a double-layer improved particle swarm unmanned aerial vehicle path, which comprise the steps of obtaining a starting point coordinate and an end point coordinate of an unmanned aerial vehicle path, carrying out coordinate conversion, determining a path maximum boundary under a new coordinate system, and dividing the path maximum boundary The method comprises the steps of establishing path particle swarm archives in each random boundary, carrying out path optimization iteratively, carrying out boundary optimization once after iteration setting times, carrying out boundary optimization again after iteration carrying out path optimization in each processed random boundary until stopping conditions are met, obtaining an optimal boundary, constructing a combined path particle swarm and a combined path particle swarm archives based on the last boundary optimization result, carrying out path optimization based on the combined path particle swarm iteration until the maximum iteration times are reached, and obtaining an optimal path.

Inventors

YAN ZHIGUO
SONG XIAOFENG
LI AO
HU GUOLIN
ZHU LIYING
SONG YUNXIA
FANG WENJING

Assignees

齐鲁工业大学(山东省科学院)

Dates

Publication Date: 20260508
Application Date: 20260108

Claims (9)

1. The method for planning the path of the double-layer improved particle swarm unmanned aerial vehicle is characterized by comprising the following steps of: acquiring a starting point coordinate and an end point coordinate of the unmanned aerial vehicle path, performing coordinate conversion, determining a path maximum boundary under a new coordinate system, and dividing the path maximum boundary A group random boundary; Creating a path particle swarm archive in each group of random boundaries, iterating for path optimization, performing boundary optimization once after iterating for set times, checking whether path particles in each group of optimized random boundaries exceed the boundaries, performing out-of-boundary processing on the path particles exceeding the boundaries, evaluating the path particles in each group of processed random boundaries and updating the path particle swarm archive, and the method comprises the following steps: initializing the positions of all path particles of the path particle swarm under the new coordinate system in each group of random boundaries to obtain initial path particles, and initializing m groups of path particle swarms and corresponding path particle swarm archives; Evaluating all initial path particles in each group of random boundaries, determining an optimal path particle set at the current moment, adding the optimal path particle set into a corresponding path particle archive to obtain an optimized path particle group, and grading all path particles in the optimized path particle group to obtain a cost evaluation result after path optimization; Dynamically selecting a speed weight parameter updating strategy of the path particles according to the cost evaluation result after each path optimizing, and updating the speeds of the path particles in each group of random boundaries by utilizing the updated speed weight parameters; all path particles and corresponding grades are evaluated after each speed and position update; carrying out the above process on all path particle swarm iteration loops in each group of initial random boundaries, carrying out boundary optimization once after iteration is carried out for a set number of times to obtain optimized random boundaries, checking whether particles in the boundaries cross the boundary to carry out cross-boundary treatment on the cross-boundary particles, re-evaluating the cost of the internal path particles, and updating the corresponding path particle swarm archive in each group of boundaries; Then, carrying out path optimization in each group of processed random boundaries in an iteration mode, and then carrying out boundary optimization until stopping conditions are met, so as to obtain an optimal boundary; And constructing a combined path particle swarm and a combined path particle swarm archive based on the final boundary optimizing result, and carrying out path optimizing based on the combined path particle swarm iteration until the maximum iteration number is reached, so as to obtain an optimal path.
2. The method for planning a path of a two-layer modified particle swarm unmanned aerial vehicle according to claim 1, wherein the acquiring the start point coordinates and the end point coordinates of the path of the unmanned aerial vehicle performs coordinate transformation, determines a path maximum boundary under a new coordinate system, and divides the path maximum boundary A group random boundary comprising: Loading a map, acquiring a starting point coordinate and an ending point coordinate of an unmanned aerial vehicle path under a map coordinate system, taking a straight line where a connecting line segment from a starting point to an ending point in a two-dimensional plane is positioned as an x axis, taking a perpendicular bisector of the connecting line segment in the two-dimensional plane as a y axis, and keeping the z axis unchanged to acquire a new coordinate system and the starting point coordinate and the ending point coordinate of the unmanned aerial vehicle path under the new coordinate system; performing coordinate conversion on four vertex coordinates of the map top view under a map coordinate system to obtain a path maximum boundary under a new coordinate; in the new coordinate system of the present invention, dividing the path maximum boundary into Group random boundaries, as a population of boundary particles, each random boundary being a boundary particle.
3. The method for planning a path of a double-layer improved particle swarm unmanned aerial vehicle according to claim 1, wherein the speed weight parameter updating strategy is dynamically selected according to the cost evaluation result after each path optimization, specifically comprising the following steps: And randomly selecting a strategy from the strategy pool by the path particle swarm in each group of random boundary, keeping the strategy for iteration if the newly obtained pareto optimal path particle number of each group of path particle swarm is improved compared with the previous generation after each iteration, and selecting other strategies from the strategy pool again for iteration if the newly obtained pareto optimal path particle number is less than the previous generation after the iteration.
4. The method for path planning for a two-layer modified particle swarm unmanned aerial vehicle of claim 1, wherein said checking whether each set of optimized random intra-boundary path particles exceeds a boundary, out-of-boundary path particles, evaluating each set of processed random intra-boundary path particles and updating a path particle swarm archive comprises: checking whether path particles in each group of optimized random boundaries exceed the boundaries; If the Y value of the path particle exceeding the boundary range is close to the left boundary value of the boundary particle, the Y value of the path particle exceeding the boundary range is taken as the left boundary value, and if the Y value of the path particle exceeding the boundary range is close to the right boundary value of the boundary particle, the Y value of the path particle exceeding the boundary range is taken as the right boundary value; and re-evaluating all path particle costs in the processed random boundary, and updating the path particle swarm archive.
5. The method for planning a path of a double-layer improved particle swarm unmanned aerial vehicle according to claim 1, wherein the constructing a combined path particle swarm and a combined path particle swarm archive based on the last boundary optimizing result, and performing path optimizing based on iteration of the combined path particle swarm until reaching the maximum iteration number, comprises: Constructing a combined path particle swarm and a comparison path particle swarm and a corresponding path particle swarm archive based on the final boundary optimizing result; And carrying out path optimization based on the combined path particle swarm and the comparison path particle swarm iteration until the maximum iteration times are reached, and obtaining an optimal path.
6. The method of claim 5, wherein constructing a combined path particle swarm and contrast path particle swarm and corresponding path particle swarm archive based on a result of a last boundary optimization comprises: based on the last boundary optimizing result, selecting the boundary particles with the fitness value from low to high and the front three bits, and respectively taking the boundary particles The path particles with the value of 30% before are defined as initial optimal path particles; judging whether all the initial optimal path particles exceed an optimal boundary range, and performing out-of-range treatment on the initial optimal path particles exceeding the optimal boundary range to obtain a treated initial optimal path particle swarm; Generating a part of new path particle definition as an initial random particle swarm in the optimal boundary; The initial optimal path particle swarm and the initial random particle swarm after treatment form a combined path particle swarm with the population size of np= NpB ; Duplicating the combined path particle swarm to obtain a comparison path particle swarm ; And respectively creating a path particle swarm archive corresponding to the combined path particle swarm and the comparison path particle swarm, and storing the path particles with PraetoRank =1 in the two groups of path particle swarms.
7. Double-deck improvement particle swarm unmanned aerial vehicle route planning system, its characterized in that includes: The path boundary dividing module is configured to acquire the starting point coordinates and the end point coordinates of the unmanned plane path, perform coordinate conversion, determine the path maximum boundary under the new coordinate system, and divide the path maximum boundary A group random boundary; The boundary optimizing module is configured to create a path particle swarm archive in each group of random boundaries and iterate path optimizing, after iterating for a set number of times, perform boundary optimizing once, check whether path particles in each group of optimized random boundaries exceed the boundary, perform out-of-boundary processing on the path particles exceeding the boundary, evaluate the path particles in each group of processed random boundaries and update the path particle swarm archive, and comprises: initializing the positions of all path particles of the path particle swarm under the new coordinate system in each group of random boundaries to obtain initial path particles, and initializing m groups of path particle swarms and corresponding path particle swarm archives; Evaluating all initial path particles in each group of random boundaries, determining an optimal path particle set at the current moment, adding the optimal path particle set into a corresponding path particle archive to obtain an optimized path particle group, and grading all path particles in the optimized path particle group to obtain a cost evaluation result after path optimization; Dynamically selecting a speed weight parameter updating strategy of the path particles according to the cost evaluation result after each path optimizing, and updating the speeds of the path particles in each group of random boundaries by utilizing the updated speed weight parameters; all path particles and corresponding grades are evaluated after each speed and position update; carrying out the above process on all path particle swarm iteration loops in each group of initial random boundaries, carrying out boundary optimization once after iteration is carried out for a set number of times to obtain optimized random boundaries, checking whether particles in the boundaries cross the boundary to carry out cross-boundary treatment on the cross-boundary particles, re-evaluating the cost of the internal path particles, and updating the corresponding path particle swarm archive in each group of boundaries; Then, carrying out path optimization in each group of processed random boundaries in an iteration mode, and then carrying out boundary optimization until stopping conditions are met, so as to obtain an optimal boundary; The path optimizing module is configured to construct a combined path particle swarm and a combined path particle swarm archive based on the last boundary optimizing result, and perform path optimizing based on iteration of the combined path particle swarm until the maximum iteration number is reached, so as to obtain an optimal path.
8. A computer readable storage medium, on which a computer program is stored, which program, when being executed by a processor, implements the steps of the two-layer modified particle swarm drone path planning method according to any of claims 1-6.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the two-layer modified particle swarm drone path planning method according to any of claims 1-6 when the program is executed.

Description

Double-layer improved particle swarm unmanned plane path planning method, system, equipment and medium Technical Field The invention belongs to the technical field of unmanned aerial vehicle path planning, and particularly relates to a method, a system, equipment and a medium for planning a path of a double-layer improved particle swarm unmanned aerial vehicle. Background The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art. In the existing unmanned aerial vehicle three-dimensional path planning technology, a particle swarm optimization algorithm is widely applied because of the avoidance of dimension disasters, the influence of the parameter combination of inertia weight, individual learning factors and social learning factors in the particle swarm algorithm on the optimization effect is remarkable, but the prior art mostly adopts a fixed or single parameter adjustment strategy, the global exploration and local development capacity of the algorithm is limited because the path optimization requirement cannot be dynamically adapted according to an iteration process, and the traditional particle swarm algorithm lacks effective constraint on an iteration range. In order to solve the above-mentioned shortcomings, the existing population optimization mechanism mostly realizes combined optimization by introducing different algorithms or combining with other algorithms, for example, the local exploration capacity is enhanced by an attraction searching algorithm, and in this way, the association logic of nodes between different paths is not considered when the attraction is calculated, but the attraction is calculated by the Euclidean distance between paths, so that the node is guided to be averaged, and the different node differentiation requirements cannot be adapted. Disclosure of Invention In order to solve the problems, the invention provides a method, a system, equipment and a medium for planning a path of a double-layer improved particle swarm unmanned aerial vehicle, which can solve the defect that a traditional particle swarm algorithm frequently falls into local optimum in unmanned aerial vehicle path planning and improve the path planning efficiency. According to some embodiments, the first scheme of the present invention provides a path planning method for a double-layer improved particle swarm unmanned aerial vehicle, which adopts the following technical scheme: the double-layer improved particle swarm unmanned aerial vehicle path planning method comprises the following steps: acquiring a starting point coordinate and an end point coordinate of the unmanned aerial vehicle path, performing coordinate conversion, determining a path maximum boundary under a new coordinate system, and dividing the path maximum boundary A group random boundary; Creating a path particle swarm archive in each group of random boundaries, carrying out path optimization in an iteration mode, carrying out boundary optimization once after iteration is carried out for a set number of times, checking whether path particles in each group of optimized random boundaries exceed the boundaries, carrying out boundary crossing treatment on the path particles exceeding the boundaries, evaluating the path particles in each group of treated random boundaries and updating the path particle swarm archive, carrying out boundary optimization again after carrying out path optimization in each group of treated random boundaries in an iteration mode until stopping conditions are met, and obtaining an optimal boundary; And constructing a combined path particle swarm and a combined path particle swarm archive based on the final boundary optimizing result, and carrying out path optimizing based on the combined path particle swarm iteration until the maximum iteration number is reached, so as to obtain an optimal path. Further, the acquiring the starting point coordinates and the ending point coordinates of the unmanned aerial vehicle path performs coordinate transformation, determines the path maximum boundary under the new coordinate system, and divides the path maximum boundaryA group random boundary comprising: Loading a map, acquiring a starting point coordinate and an ending point coordinate of an unmanned aerial vehicle path under a map coordinate system, taking a straight line where a connecting line segment from a starting point to an ending point in a two-dimensional plane is positioned as an x axis, taking a perpendicular bisector of the connecting line segment in the two-dimensional plane as a y axis, and keeping the z axis unchanged to acquire a new coordinate system and the starting point coordinate and the ending point coordinate of the unmanned aerial vehicle path under the new coordinate system; performing coordinate conversion on four vertex coordinates of the map top view under a map coordinate system to obtain a path maximum boundary under