CN-121994250-A - Multi-target unmanned aerial vehicle path planning method and system based on particle swarm optimization

CN121994250ACN 121994250 ACN121994250 ACN 121994250ACN-121994250-A

Abstract

The invention discloses a multi-target unmanned aerial vehicle path planning method and system based on particle swarm optimization, and belongs to the technical field of unmanned aerial vehicle autonomous navigation and cluster cooperative control. The method comprises the steps of constructing a flight environment comprehensive interference map through real-time sensing and geomagnetic and communication dual dynamic interference fusion, further predicting the state stability of an unmanned aerial vehicle and quantifying the cluster collision risk, establishing a multi-target particle swarm optimization model to solve by taking the minimum total path length of the clusters and the minimum estimated maximum cluster collision risk value as cooperative targets, obtaining a pareto front edge solution set, and finally determining an optimal flight path from the solution set according to a preset task safety level. According to the invention, the cooperative optimization of the unmanned aerial vehicle cluster path efficiency and the flight safety under the complex interference environment is realized.

Inventors

ZHANG AIHUA
WANG WANG
ZHOU FANGMING
LI YINDONG
HE CHANGLIANG
DENG WEIMIN

Assignees

四川吉利学院

Dates

Publication Date: 20260508
Application Date: 20260409

Claims (10)

1. The multi-target unmanned aerial vehicle path planning method based on particle swarm optimization is characterized by comprising the following steps of: S1, acquiring three-dimensional structure data, real-time geomagnetic field intensity data, communication interference spectrum data and real-time flight state data and real-time position data of each unmanned aerial vehicle in a cluster of a planning airspace; S2, based on real-time geomagnetic field intensity data, respectively evaluating the degree of decline of the positioning accuracy and the degree of course angle drift of the unmanned aerial vehicle caused by geomagnetic anomaly to generate geomagnetic navigation interference coefficients; s3, integrating the geomagnetic navigation interference coefficient and the communication control interference coefficient to generate a flight environment comprehensive interference coefficient, and predicting the flight state stability of each unmanned aerial vehicle by combining the flight environment comprehensive interference coefficient and the real-time flight state data of each unmanned aerial vehicle; S4, calculating real-time collision risk values between all unmanned aerial vehicles in the cluster based on the predicted flight state stability and real-time position data of each unmanned aerial vehicle, and aggregating to generate the real-time collision risk values of the cluster; S5, establishing a multi-target particle swarm optimization model by taking the minimum total flight path length of the clusters and the minimum estimated maximum cluster collision risk value as collaborative optimization targets, wherein each particle represents a cluster path scheme; And S6, carrying out iterative solution through a multi-target particle swarm optimization model to obtain a pareto front solution set, determining a final flight path from the pareto front solution set according to a preset task security level, and controlling the unmanned aerial vehicle to fly in a cluster.
2. The particle swarm optimization-based multi-objective unmanned aerial vehicle path planning method according to claim 1, wherein the steps of obtaining three-dimensional structure data, real-time geomagnetic field intensity data, communication interference spectrum data, and real-time flight state data and real-time position data of each unmanned aerial vehicle in the cluster comprise: S110, acquiring three-dimensional structure data comprising terrain elevation information and the spatial position of an obstacle; s120, acquiring real-time geomagnetic field intensity data acquired by a distributed geomagnetic sensor in a planning space domain; S130, acquiring the position of a signal transmitting source, the intensity of a received signal and the time delay and intensity characteristics of multipath components, which are acquired by a communication monitoring node in a planning space domain, so as to obtain communication interference spectrum data; S140, acquiring the flying speed and the attitude angular speed of each unmanned aerial vehicle in a cluster, which are reported in real time through an airborne sensor, acquiring a horizontal positioning precision factor and a vertical positioning precision factor which are output through an unmanned aerial vehicle positioning system, calculating to obtain a positioning precision attenuation coefficient, acquiring heading angle data which are output through an unmanned aerial vehicle heading sensor, calculating a short-term fluctuation standard deviation to obtain a heading maintenance precision coefficient, acquiring a received signal strength indication value and a signal-to-noise ratio value which are output through an unmanned aerial vehicle communication module, calculating to obtain a communication receiving quality coefficient, forming an unmanned aerial vehicle performance vector by the positioning precision attenuation coefficient, the heading maintenance precision coefficient and the communication receiving quality coefficient, forming flying state data by the flying speed, the attitude angular speed and the unmanned aerial vehicle performance vector, and acquiring real-time position data output by the positioning system.
3. The method for planning a path of a multi-objective unmanned aerial vehicle based on particle swarm optimization according to claim 2, wherein the steps of respectively evaluating the degree of decline of the positioning accuracy and the degree of drift of the heading angle of the unmanned aerial vehicle caused by geomagnetic anomalies based on real-time geomagnetic field intensity data to generate geomagnetic navigation interference coefficients, respectively evaluating the degree of attenuation of signals caused by terrain and building shielding and the degree of upward lift of communication error rates caused by signal reflection superposition based on communication interference spectrum data and three-dimensional structure data to generate communication control interference coefficients comprise: s210, comparing the real-time geomagnetic field intensity data with a standard geomagnetic field map, and calculating magnetic field deviation values of all the space grids; S220, calculating a positioning error factor of each grid point according to the mapping relation between the magnetic field intensity deviation value and a preset positioning error threshold value, calculating a spatial gradient of the magnetic field intensity of each grid point, and calculating a heading disturbance factor of each grid point according to the mapping relation between the spatial gradient and a preset heading disturbance threshold value; S230, constructing a signal propagation geometric model based on three-dimensional structure data, simulating signal propagation by a ray tracing algorithm by combining the position of a signal transmitting source and the intensity of a received signal in communication interference frequency spectrum data, calculating the signal path loss of each grid point caused by the shielding of the topography and the building, and generating the path loss factor of each grid point; S240, extracting time delay and intensity characteristics of multipath components from communication interference spectrum data, calculating interference intensity factors of all grid points according to the mapping relation between the multipath component characteristics and a preset bit error rate threshold value, and combining the path loss factors and the interference intensity factors according to preset weights to generate communication control interference coefficients of all the grid points.
4. The particle swarm optimization-based multi-target unmanned aerial vehicle path planning method according to claim 3, wherein the geomagnetic navigation interference coefficient and communication control interference coefficient are fused to generate a flight environment comprehensive interference coefficient, and the method for predicting the flight state stability of each unmanned aerial vehicle by combining the flight environment comprehensive interference coefficient and real-time flight state data of each unmanned aerial vehicle comprises the following steps: S310, fusing geomagnetic navigation interference coefficients and communication control interference coefficients for each grid point in a planning airspace to obtain a flight environment comprehensive interference coefficient of each grid point; S320, inquiring from a flight environment comprehensive interference coefficient distribution diagram according to real-time position data of each unmanned aerial vehicle to obtain a flight environment comprehensive interference coefficient corresponding to the position of each unmanned aerial vehicle; S330, analyzing unmanned aerial vehicle performance vectors from flight state data, and calculating anti-interference capacity factors of each unmanned aerial vehicle based on the unmanned aerial vehicle performance vectors; S340, fusing the comprehensive interference coefficient of the flight environment of each unmanned aerial vehicle with the anti-interference capacity factor, and calculating to obtain an effective interference coefficient; S350, calculating the dynamic instability of the unmanned aerial vehicle based on real-time flight state data of each unmanned aerial vehicle; S360, combining the effective interference coefficient and the dynamic instability of the fuselage, and calculating and outputting the stability of the flight state through a stability mapping model.
5. The particle swarm optimization-based multi-objective unmanned aerial vehicle path planning method according to claim 4, wherein the calculating real-time collision risk values between all unmanned aerial vehicles in the cluster based on the predicted flight state stability and the real-time position data of each unmanned aerial vehicle, and the aggregating and generating the cluster real-time collision risk values comprises: S410, calculating the real-time distance between any two unmanned aerial vehicles in the cluster according to the real-time position data, and simultaneously acquiring the flight state stability of the two unmanned aerial vehicles; s420, calculating a real-time collision risk value between the two unmanned aerial vehicles through a collision risk model according to the real-time distance and the flight state stability; and S430, aggregating the real-time collision risk values among all the unmanned aerial vehicle pairs to obtain a cluster real-time collision risk value.
6. The method for path planning of a multi-objective unmanned aerial vehicle based on particle swarm optimization according to claim 5, wherein the establishing a multi-objective particle swarm optimization model with minimized total flight path length and minimized predicted maximum cluster collision risk value as collaborative optimization targets, wherein each particle represents a cluster path plan, comprises: s510, carrying out parameter coding on candidate flight paths of each unmanned aerial vehicle, forming a particle by a sequence of all unmanned aerial vehicle path codes, and determining a particle coding mode; S520, calculating the sum of path lengths of all unmanned aerial vehicles in a path scheme represented by particles to obtain the total flight path length of the cluster, and taking the total flight path length as a first optimization target fitness value; S530, simulating the unmanned aerial vehicle to fly along a path of the particle codes, calculating a cluster instantaneous collision risk value at each simulation moment in the flying process according to the comprehensive interference coefficient and the position relation of the flying environment at the passing position, taking the maximum value of the cluster instantaneous collision risk value in the whole simulation process as a predicted maximum cluster collision risk value of the particles, and taking the predicted maximum cluster collision risk value as a second optimization target fitness value; S540, constructing a solving framework of a multi-target particle swarm optimization algorithm based on the particle coding mode, the first optimization target fitness value and the second optimization target fitness value, and completing the establishment of a multi-target particle swarm optimization model.
7. The particle swarm optimization-based multi-objective unmanned aerial vehicle path planning method according to claim 6, wherein iterative solution is performed through a multi-objective particle swarm optimization model to obtain pareto front solution sets, and final flight paths are determined from the pareto front solution sets according to a preset task security level to control unmanned aerial vehicle cluster flight: s610, initializing a particle swarm based on a multi-objective particle swarm optimization model, and performing multi-objective iterative optimization on a first optimization objective fitness value and a second optimization objective fitness value; s620, after iteration is finished, outputting a non-dominant solution set as a pareto front solution set; S630, calculating the comprehensive evaluation score of each solution in the pareto front solution set based on the first optimization target fitness value and the second optimization target fitness value according to the preset task security level; and S640, selecting the solution with the highest comprehensive evaluation score, and issuing the solution to the unmanned aerial vehicle cluster for execution as the final flight path after decoding.
8. A particle swarm optimization-based multi-objective unmanned aerial vehicle path planning system for implementing the particle swarm optimization-based multi-objective unmanned aerial vehicle path planning method according to any of claims 1 to 7, comprising: the data acquisition module is used for acquiring three-dimensional structure data of a planning airspace, real-time geomagnetic field intensity data, communication interference spectrum data, and real-time flight state data and real-time position data of each unmanned aerial vehicle in the cluster; The system comprises an interference evaluation module, a communication control interference coefficient, a communication interference spectrum data and a three-dimensional structure data, wherein the interference evaluation module is used for respectively evaluating the degree of decline of the positioning precision and the course angle drift degree of the unmanned aerial vehicle caused by geomagnetic anomaly based on real-time geomagnetic field intensity data and generating geomagnetic navigation interference coefficients; The stability prediction module is used for integrating the geomagnetic navigation interference coefficient and the communication control interference coefficient to generate a flight environment comprehensive interference coefficient, and predicting the flight state stability of each unmanned aerial vehicle by combining the flight environment comprehensive interference coefficient and the real-time flight state data of each unmanned aerial vehicle; the risk calculation module is used for calculating real-time collision risk values between all unmanned aerial vehicles in the cluster based on the predicted flight state stability and real-time position data of each unmanned aerial vehicle, and generating the real-time collision risk values of the cluster in an aggregation mode; The optimization modeling module is used for establishing a multi-target particle swarm optimization model by taking the minimum total flight path length of the clusters and the minimum estimated maximum cluster collision risk value as collaborative optimization targets, wherein each particle represents a cluster path scheme; the decision execution module is used for carrying out iterative solution through the multi-target particle swarm optimization model to obtain a pareto front solution set, determining a final flight path from the pareto front solution set according to a preset task security level, and controlling the unmanned aerial vehicle to fly in a cluster.
9. An electronic device, comprising a processor and a memory, wherein the memory stores a computer program which can be called by the processor, and the processor executes the multi-objective unmanned aerial vehicle path planning method based on particle swarm optimization according to any of claims 1 to 7 by calling the computer program stored in the memory.
10. A computer readable storage medium, characterized in that a computer program is stored thereon, which computer program, when being executed by a processor, implements the particle swarm optimization based multi-objective unmanned aerial vehicle path planning method according to any of claims 1 to 7.

Description

Multi-target unmanned aerial vehicle path planning method and system based on particle swarm optimization Technical Field The invention relates to the technical field of unmanned aerial vehicle autonomous navigation and cluster cooperative control, in particular to a multi-target unmanned aerial vehicle path planning method and system based on particle swarm optimization. Background The unmanned aerial vehicle cluster has great application potential in the fields of reconnaissance, logistics, agricultural plant protection and the like, and one of the core challenges is to conduct safe and efficient collaborative path planning in a complex dynamic environment. The traditional path planning method focuses on single targets such as single machine obstacle avoidance or minimized flight distance, and is difficult to adapt to complex requirements of cluster cooperation. The prior art mainly has the following defects that firstly, most methods depend on a static environment model, and the real-time influence of dynamic and non-measurable factors such as geomagnetic anomaly, communication interference and the like in a real flight environment on the navigation and control performance of an unmanned aerial vehicle is not fully considered, so that the reliability of a planned path is reduced in actual execution. Secondly, cluster path planning often takes anti-collision as hard constraint treatment, and quantitative evaluation and active optimization of potential collision risks caused by unstable unmanned aerial vehicle states due to environmental interference are lacking. Finally, the common optimization algorithm is easy to fall into local optimum when processing multi-objective collaborative optimization such as path length and security risk, and is difficult to provide a series of balanced solutions which can be flexibly selected by a decision maker according to task requirements. Therefore, a multi-objective path planning method capable of integrating real-time environment interference sensing, quantitatively evaluating cluster dynamic collision risk and effectively synergistically optimizing path efficiency and flight safety is urgently needed, so that autonomy, adaptability and overall task efficiency of an unmanned aerial vehicle cluster in a complex space domain are improved. Disclosure of Invention In order to overcome the defects in the prior art, the invention provides a multi-target unmanned aerial vehicle path planning method and system based on particle swarm optimization, which aim to realize safe and efficient collaborative flight of unmanned aerial vehicle clusters in a dynamic interference environment. In order to achieve the above purpose, the invention adopts the following technical scheme: In a first aspect, the present invention provides a multi-objective unmanned aerial vehicle path planning method based on particle swarm optimization, comprising the steps of: S1, acquiring three-dimensional structure data, real-time geomagnetic field intensity data, communication interference spectrum data and real-time flight state data and real-time position data of each unmanned aerial vehicle in a cluster of a planning airspace; S2, based on real-time geomagnetic field intensity data, respectively evaluating the degree of decline of the positioning accuracy and the degree of course angle drift of the unmanned aerial vehicle caused by geomagnetic anomaly to generate geomagnetic navigation interference coefficients; s3, integrating the geomagnetic navigation interference coefficient and the communication control interference coefficient to generate a flight environment comprehensive interference coefficient, and predicting the flight state stability of each unmanned aerial vehicle by combining the flight environment comprehensive interference coefficient and the real-time flight state data of each unmanned aerial vehicle; S4, calculating real-time collision risk values between all unmanned aerial vehicles in the cluster based on the predicted flight state stability and real-time position data of each unmanned aerial vehicle, and aggregating to generate the real-time collision risk values of the cluster; S5, establishing a multi-target particle swarm optimization model by taking the minimum total flight path length of the clusters and the minimum estimated maximum cluster collision risk value as collaborative optimization targets, wherein each particle represents a cluster path scheme; S6, carrying out iterative solution through a multi-target particle swarm optimization model to obtain a pareto front solution set, determining a final flight path from the pareto front solution set according to a preset task security level, and controlling unmanned aerial vehicle cluster flight; According to the above technical scheme, the obtaining the three-dimensional structure data, the real-time geomagnetic field intensity data, the communication interference spectrum data, and the real-time flight state data and the re